Pascal Reischl — Log

"The paper can't get to me"

Fri, 09 Oct 2026 00:00:00 GMT

Kay-Dietrich is out on the road for work. He pulls over and opens a ring binder. It's a Time/system planner from 1996, bought second-hand for a song, and there was still a car hire advert from the same year tucked inside. He checks what's left for today and what's on tomorrow. Then he drives on.

The remarkable thing is not so much the paper as the person opening it. Kay-Dietrich has been practising GTD since around 2018 and has put dozens of hours into automations. His projects live in Trello, sorted by areas of responsibility. In his Outlook calendar, n8n automatically adds the drive there, the preparation and the drive back. And this same person has spent the last few weeks copying out his days in pencil. His reasoning fits into two sentences: "I can always get to the paper. But the paper can't get to me."

My own notebook and pencil go back to a conversation with him. Even so, the planner was new to me, and by his own account he was rather proud of that.

Why the digital calendar doesn't show you the day

A digital calendar is a store that many people are allowed to write to. Colleagues send invitations, automations add entries. Everything in it is correct, but you didn't write most of it yourself. Kay-Dietrich says of his own: "I can't see my Outlook calendar in my mind's eye." The day he has copied out, on the other hand, is in his head: ten o'clock, one o'clock, half past two.

Then there's the missing frame. The digital day has 24 hours, and seven hours of driving are just one coloured bar among many. On a page whose grid starts at 8 am and ends at 8 pm, they take up more than half the day, and you can see that as you write them in. In Kay-Dietrich's words: "That's when you realise: the day is already full."

As far as the method goes, none of this is new. In GTD, the calendar holds only the hard landscape: whatever has to happen on that particular day or at that particular time. You decide in the moment which next actions fit into the gaps, based on context, time available, energy and priority. The prerequisite is the daily review, a look at your calendar and lists every day. Do it on your phone between two emails and you haven't done it. Copying out the day makes it unavoidable. That's where the main gain lies, not in the paper.

Daily planning on paper: a copy, not a second calendar

Everything hinges on one rule, and Kay-Dietrich phrases it without any wiggle room: "Nothing appears on the paper that wasn't in my digital world first." The original stays in his personal and his work Outlook. Inviting, rescheduling, cancelling, marking something as tentative: only the digital calendar can do that, and the automations hang off the digital calendar alone.

The pre-printed daily page has an hourly grid from 8 am to 8 pm and space for tasks and contacts. Kay-Dietrich fills it in the day before or in the morning, sometimes three days ahead:

Appointments and journeys from the digital calendar, with their times, into the grid.
What should happen today, from his Today list in Trello.
Who he wants to call.

By his account, that takes five to eight minutes. If you already plan the day the evening before, nothing changes about the routine, only the tool.

Two rules I would add

The grid and the task section are not equally binding. Only what's fixed goes into the grid. The task section is an intention, not a commitment. Whatever is still open there in the evening doesn't get carried over to the next page; it's still on the digital list anyway. That way the page stays a copy you simply turn over, rather than becoming a daily list you carry forward by hand every evening.

The page gets emptied in the evening. New things still end up on the paper during the day, a number to call back, say, or a commitment made in conversation. Before tomorrow's page is written, everything that didn't come from the digital system goes into its inbox. For his own thoughts, Kay-Dietrich keeps a second notebook, a bullet journal that serves as a small inbox and a place to think on paper. That keeps the daily page clear.

Why the copy isn't a parallel system

The obvious objection is that it doubles the work and creates a parallel system that practically invites discrepancies. I've written myself that copying tasks onto a second daily board is the surest route to two lists, neither of which is right any more. I stand by that.

The difference is lifespan. A copy on a board lives on: you maintain it, and at some point you no longer know which version counts. The daily page lives for one day and is never the source for anything else. Something is done when it's ticked off in the original. In GTD terms, the page isn't part of organising but of engaging: it doesn't file anything, it helps you choose.

What the paper itself gives him above all is fewer glances at his phone, and in his case four taps are enough for his mind to be elsewhere. In the morning, the phone can stay where it is for the first hour, because the day is written down in a different place.

Where I disagree: calm is not the same as up to date

"The paper can't get to me" describes the method's strength and its weakness in the same sentence. If someone moves the eleven o'clock at half past nine, the page never finds out. For Kay-Dietrich that hardly matters, because by his own description his working days don't shift around much. But if you keep a work calendar that a whole team writes into, at ten o'clock you're holding a page that's reassuring and no longer right. That kind of calm is worse than none at all. There, the copy needs fixed checkpoints, such as a look at the digital calendar before every change of location. Any changes get pencilled in.

The second point is the year planner in the ring binder. Kay-Dietrich doesn't put future dates in it himself, but just in case, he relies on the weekly review to bring things back together. For me, that's too late. A commitment that exists only on paper for a week is invisible for that whole week to anyone looking at your digital calendar, including your automations. New appointments go into the original the same day.

Ring-bound planner, bullet journal or pocket diary

What matters is less the format of the paper than the question of where the original lives.

Format	What's on the paper	Original	Works if
Daily page in a ring-bound planner	a copy of one day	digital	others rarely change the calendar during the day
Bullet journal	your own structure on blank pages, open items carried forward by hand	paper	the paper is the whole system, not a second one
Pocket diary	appointments agreed on the spot	paper	nobody else writes in your calendar
Digital only	nothing	digital	colleagues use your calendar too

Kai sold his planner ten years ago and now carries an A6 diary in his back pocket. When a tradesperson asks when he's free, he pulls it out: "Monday, 4 pm", done. That works as long as this piece of paper is the original. Neither would suit Jörg: his work calendar is shared with lots of colleagues, and on his phone he overlays his personal calendar on top. For him, paper would double the work, and he's right about that.

Where it typically fails

The paper becomes a second original. An appointment gets agreed straight out of the ring binder and never makes it into the digital calendar. From then on there are two calendars and not one you can trust.
The page becomes a wish list. Eight intentions next to a full grid aren't planning, they're a guilty conscience with times attached. Kay-Dietrich's yardstick: "I can't do eight things tomorrow. I can only do two."
Blank pages instead of a template. Kay-Dietrich first tried it in an A5 notebook and found it too much effort to write everything out from scratch every time. Only the pre-printed page with its fixed grid solved that for him.
The page quietly goes out of date. In a shared calendar, the morning copy is a historical document by lunchtime, and you're working to a plan that no longer applies.
Notes stay on the paper. Tuesday's call-back number is still on Tuesday's page on Friday. If you don't empty the page in the evening, you're keeping an in-tray that nobody processes.

Common questions

Is daily planning on paper double the work?

Only if the paper turns into a second original. If the digital calendar remains the only source and you just copy out the day in question, it takes a few minutes and replaces the daily look at appointments and lists. It only doubles the work once appointments are agreed on paper and have to be transferred later.

What belongs on a handwritten daily page?

The hourly grid holds only the appointments and journeys that are fixed for that day, in other words the hard landscape from the digital calendar. Next to it go a few intentions and the calls that are due. Whatever is still open in the evening stays on the digital lists and is not copied onto the next page by hand.

What is the difference between a ring-bound planner and a bullet journal?

A ring-bound planner has pre-printed daily pages with an hourly grid, while a bullet journal consists of blank pages that you structure yourself. In a bullet journal, open items are carried forward to new pages by hand; that's part of the method. As a copy of a digital calendar, the planner is the better fit, because the template keeps the copying short.

When does daily planning on paper not work?

If other people keep changing your calendar throughout the day, a page written in the morning is out of date within a few hours. That mainly affects shared work calendars in teams. There, the copy needs fixed checkpoints against the digital calendar, or you leave it out altogether.

How do you stop paper and the digital calendar drifting apart?

With one fixed direction: appointments are only ever created in the digital calendar and copied from there onto paper, never the other way round. Anything new that gets jotted on the paper during the day is treated as an inbox item and goes into the digital system in the evening, before the next page is written.

Tomorrow morning, copy out just this one day, in pencil, from 8 am to 8 pm: appointments and journeys first, then only the intentions that fit into the gaps. Keep it up for a week, and each time you notice while copying that something no longer fits, make a tally mark in the margin. Every mark is a decision the afternoon would otherwise have made for you.

If you can no longer say for sure which calendar counts (Outlook, phone or ring binder), it's rarely down to discipline and usually down to your tool landscape. In the Digital Workplace Audit we work out where the original lives and what may only ever be a copy. Write to me if you're shuttling between several calendars and no longer fully trust any of them.

Ten records were agreed. Millions ended up in the test system.

Fri, 02 Oct 2026 00:00:00 GMT

An IT firm I know well was supposed to fill a test system together with a large German company. Ten customer records had been agreed, and the data protection team and the works council had both approved it. On each side, a programmer clicked OK. The transfer took a minute instead of three seconds, and afterwards the test system held not ten records but millions.

The corporation had to report the breach to the data protection authority, without undue delay and where feasible within 72 hours. The small service provider had to produce what it had, immediately: its data processing agreement (DPA) and the description of its technical and organisational measures. If an hour like that is when you start writing those documents, you are in serious trouble.

In the small businesses I know, it starts more innocently, with an annoyed customer and a website without a privacy policy. Shortly afterwards the authority starts asking questions, and by then it is too late to go looking for the paperwork.

I am not a lawyer. I have worked in IT since 1998 and deal with these questions almost every day. What follows is the basic kit I recommend to self-employed people from the technical side, not legal advice.

Who the GDPR applies to, and who it does not

If the only people you keep in Gmail, iCloud or Outlook are family and friends, you are out: the GDPR does not apply to purely personal or household activities (Art. 2(2)(c), the so-called household exemption). As soon as customers, clients, members or prospects are in your address book, it applies in full, even to a one-person business, and so do the DSG in Austria and the BDSG in Germany. If you use one account for family and clients at the same time, you cannot rely on the household exemption for the business part.

Clubs and associations overlook this most often. The treasurer keeps the membership list in their private Gmail because it happens to be there, and that leaves the club with exactly the same obligations as a company.

What is in the account	Does the GDPR apply?	Do you need a DPA?
Family and friends	no, household exemption	no
Personal and business mixed	yes, for the business part	yes, but not available for a free account
Customers, clients, prospects	yes, even in a one-person business	yes
Members of a club or association	yes	yes

What a data processing agreement covers

The data processing agreement (sometimes called a data processing addendum; in German Auftragsverarbeitungsvertrag, or AVV) is set out in Art. 28 GDPR. Your client gave her phone number to you, not to Microsoft. Once it is in your mailbox, Microsoft processes it anyway. The DPA makes sure that only happens on your instructions, with fixed rules on confidentiality, security, sub-processors, deletion and audits. A missing contract is an infringement, and Art. 83 provides for fines.

The agreement works in both directions. If you process data for clients yourself, say as a virtual assistant working in other people’s inboxes or as a web service provider, you are the processor. Then many clients will want to see your DPA and your measures.

The list first, then the contracts

The obvious first step would be to go and get the DPA from Google or Microsoft. I would turn the order round. Email is only one of many services that touch client data, and if you start with the best-known one, you often stop with the best-known one.

Start with a mind sweep, just for tools: email, calendar, cloud storage, video conferencing, newsletter, appointment booking, accounting, the contact form on your website. Anything where a name or a phone number lands goes on the list.

That list becomes your record of processing activities under Art. 30 GDPR. The exemption for small businesses only applies to occasional processing, and you process client data all the time. The EU plans to widen this exemption considerably with its Digital Omnibus, but that has not been adopted yet. A table is enough as a framework, plus your contact details and a short description of your measures:

Purpose	Which data	Where, under which contract	Third country	When deleted
Client correspondence	Name, email, phone	Microsoft 365, DPA filed	USA, standard contractual clauses in the DPA	after the engagement ends, unless something has to be kept
Invoices	Name, address, service	Accounting software, DPA filed	none, servers in the EU	when the retention period ends
Appointment booking	Name, email, appointment	Booking tool, DPA filed	as per the list of sub-processors	after the appointment

Many people leave the last column empty because they want to keep client data forever. That is not allowed. And when a client says “delete my data”, you know straight away everywhere it is stored.

The same list gives you your privacy policy, linked in the footer of your website and referenced in your contracts. It names who data goes to, Google or Microsoft for example, and states that it may also be processed in the US.

Data processing agreements for freelancers at Google and Microsoft

Free Gmail, Outlook.com, GMX and Yahoo do not offer a DPA, because the provider there is not working on your behalf but as a controller in its own right, with purposes of its own. And paid does not automatically mean business: Google One and Microsoft 365 Family are consumer plans and change nothing about that. The exception is Google Workspace Individual, a subscription for freelancers on top of a regular Google account. If you use it for business, Google's additional terms say a data processing addendum applies. A business account with your own domain is still the cleaner option.

Google Workspace: As a super admin, open the Admin console and go to Menu → Account → Account settings → Legal and compliance. In the “Security and Privacy Additional Terms” section, click “Review and Accept” next to the “Cloud Data Processing Addendum”. If your contract already includes the addendum, Google says accepting it makes no difference, so it cannot do any harm.

Microsoft 365: There is nothing to click here. The Data Protection Addendum is automatically part of the business agreements through the licensing terms, and the current version is at aka.ms/DPA.

Both publish audit reports and certifications such as ISO 27001, Microsoft in the Service Trust Portal, Google in the Compliance Reports Manager. Download everything as PDFs, put it in a folder called Data Protection and make a note of when you accepted. A box that only gets ticked once the authority asks will not help you any more. The same goes for every service on the list.

What you have to prove yourself

The DPA covers what the provider does. What you do is set out in your technical and organisational measures under Art. 32, TOMs for short. They have to be both set up and written down. Here is what I consider reasonable today:

Sign-in: two-factor authentication wherever it is available, and a password manager instead of one password for everything.
Devices: a lock by password, PIN or Face ID, and an encrypted drive. Current devices can do this, but it is not always switched on. Check on a Mac under System Settings → Privacy & Security → FileVault, on Windows 11 under Settings → Privacy & security → Device encryption, or BitLocker.
Permissions: only people who have to be administrators are administrators. Anyone who leaves the business loses their access the next day.
Backups: ideally weekly, with a second provider that has a data centre in the EU and a DPA.
Sharing: twice a year, check who has access to what, and record it with the date.

For the settings in the Admin console, feel free to ask an AI model: “I recently became the super admin of Google Workspace for two people. Which security settings matter most, and where do I find them?” Working through the list is still down to you.

You document this in a simple file: which service, what is set up, what happens when someone leaves. That does not mean you have won automatically. But when an authority decides on a fine, it expressly takes into account which measures you had taken (Art. 83(2) GDPR). An access list that has been signed off twice a year for years looks rather different from a shrug.

The Data Privacy Framework applies, until further notice

Google, Microsoft and Apple contract with European customers through companies in Ireland and are subject to the GDPR. The problem is US law, which their parent companies are also subject to. That is why the European Court of Justice has already struck down the basis for transfers to the US twice: Safe Harbor in 2015 and Privacy Shield in 2020. Since July 2023 the EU-U.S. Data Privacy Framework has applied, and an appeal against it is pending before the same court.

For you as a business customer of Google or Microsoft, this means less than it sounds, because the EU standard contractual clauses are in your DPA anyway. If the framework falls, that does not make the transfer unassailable, but you are not left without a contract.

Jörg works for an Australian company and knows the other side: the US is only the best-known case. Every transfer to a third country without an EU adequacy decision needs its own legal basis. There is no such decision for Australia, so transfers there need the European Commission’s standard contractual clauses, adopted unchanged. The same applies to a German IT service provider whose team is partly based in Kosovo. Where your data really goes is in the list of sub-processors that comes with every DPA.

Using only European providers is a legitimate choice, but not the only permissible one, and in my experience you pay for it in features with many tools. To me it makes more sense to do what is legally possible with every service, and to be able to prove it.

Data protection is an area of responsibility, not a project

Setting it up is a project with a clear outcome: list complete, contracts filed, measures set up and described, privacy policy online. The next actions almost write themselves. “Download the DPA from Zoom” is as concrete as a next action gets.

After that, data protection is an area of responsibility, and an area runs on reminders in your tickler file rather than on good intentions: the sharing review twice a year, the list against the services you actually use once a year. New services go on the list first and into use second. There is a checklist for when someone leaves, because it happens too rarely to keep in your head.

Where this typically falls apart

The account from the early days. Your first clients land in your private Gmail because it was already there, and they stay there. There is no DPA for it, and a consumer plan like Google One changes nothing about that.
The treasurer with the private address. Members’ mail runs through the inbox of whoever holds the post at the time. When the post changes hands, the list does not move with it. It doubles.
Paperwork on demand. DPAs only get downloaded and TOMs only get written when someone asks. A document dated yesterday only proves that you started yesterday.
Sharing with no end date. The folder for a project that finished long ago is still shared, and nobody has looked at it since.
The third country through the back door. The provider is in the EU, its support team is not. If you have never read the list of sub-processors, you do not know where your data ends up.

Common questions

Is a data processing agreement mandatory for freelancers?

Yes, as soon as a service provider processes your clients’ personal data on your behalf, such as your email provider or your cloud storage. There is no minimum size: the GDPR applies to one-person businesses too. The only exception is purely private data, such as the addresses of family and friends.

Can I use a free Gmail account for business?

Not for client data, because Google does not offer a data processing agreement for free accounts. The same goes for free accounts at Outlook.com, GMX or Yahoo and for consumer plans like Google One. You get the agreement with a business account such as Google Workspace or Microsoft 365 Business, and at Google also with Workspace Individual.

Does the GDPR apply to clubs and associations?

Yes, a club or association that holds member data falls fully under the GDPR, whatever its size. The association is the one responsible, and the board has to see that it is implemented. Membership lists therefore belong in an association account with a data processing agreement, not in an office holder’s private email account.

Do I need a data protection officer as a freelancer?

Usually not. The GDPR only requires one in specific cases, for example when large-scale processing of health data is part of your core activity. In Germany one is also mandatory if, as a rule, at least 20 people are constantly involved in the automated processing of personal data (Section 38 BDSG), while Austria has no such threshold. All other obligations apply regardless.

Is the Data Privacy Framework enough for Google and Microsoft?

For now it is a valid basis for transfers to the US, but a contested one: an appeal against the adequacy decision is pending before the European Court of Justice, which struck down both of its predecessors. Google and Microsoft have also included the EU standard contractual clauses in their data processing agreements. If the framework falls, their business customers are therefore not left without a contractual basis.

Open your email provider’s admin console today and look for the data processing agreement. If you find it within ten minutes, put it in the Data Protection folder and move on to the next service on the list. If you cannot find an admin console at all, you already have the more important answer.

Which services touch your client data is rarely written down anywhere in full, and often nobody has checked the settings behind them. In the Digital Workplace Audit I go through your tools with you from the technical side and record what is in place and what is missing; the legal review stays with your lawyer. Write to me if you are self-employed and would rather sort out the paperwork now than on the day someone asks for it.

Your Claude plan isn't too small. It's set to the wrong level.

Sun, 27 Sep 2026 00:00:00 GMT

Usually one that's too high. People who burn through their weekly limit by Wednesday rarely ask too many questions. They ask them with a setting that is far too expensive for the job.

You don't need a different model to fix that. With Claude Opus 5.5, the same task can cost up to eleven times as much of your plan depending on the effort level. And in most cases the more expensive result is not better.

What the effort setting does

Next to the model name, Claude has a second control: effort. It has five levels, Low, Medium, High, Xhigh and Max. It decides how long Claude thinks before answering, how often it calls tools such as web search, and how detailed the answer gets.

Opus 5.5 ships set to Medium. Most people leave it there or turn it up on principle, because more must be better. The second habit is expensive.

Effort is not a fixed budget. On a simple question, Claude thinks only briefly even on Max. A high level gets expensive on tasks with a lot to chew on, and that is exactly where it shows up in your limit.

Why your plan runs out so fast

Your plan doesn't count messages. It counts tokens, the chunks of text Claude reads and writes. Thinking produces tokens, even though you never see them. Every level up means more thinking tokens per answer.

On top of that, chat, the desktop app and Claude Code all draw from the same pool. An afternoon of coding on a high level leaves less for chat in the evening. Anthropic's help article on usage limits names the effort level explicitly as one of the factors that decide how fast your allowance goes.

The numbers for Opus 5.5

Anthropic doesn't publish how much of your plan a level costs. The best approximation comes from Artificial Analysis. They run Opus 5.5 through the same tests on every level and measure what a task costs at the official API prices. Since your plan is also metered in tokens, the ratios carry over well.

I use Medium, the default, as the baseline.

Level	Usage per task	Tasks solved out of 100
Low	0.4×	42
Medium (default)	1×	51
High	1.4×	54
Xhigh	2.6×	56
Max	4.5×	58

Low and Max are a factor of 11 apart. On Max, Opus 5.5 writes around 119,000 tokens per test task on average, most of it thinking, according to Artificial Analysis. The biggest saving sits at the bottom: Low uses about 60 percent less than Medium.

What the scores mean

The right-hand column comes from the Artificial Analysis Intelligence Index, as of September 2026. It combines ten very hard tests: expert-level science questions, real office tasks with documents and spreadsheets, programming in a terminal. Roughly speaking, the number shows how many out of 100 such tasks Claude solves.

So 51 means about every second expert-level task works out. Your email, your filled-in form or your meeting notes are far easier. Every level handles those.

One caveat belongs here. Because the test tasks are so hard, Claude thinks a lot on every level. On everyday tasks the levels sit closer together in usage. The order stays the same.

Does it get better if I turn it up?

Barely. Performance rises clearly only at the start. After that you pay more and more for less and less.

Step	Extra usage	Extra tasks solved out of 100
Low → Medium	+144 %	+9
Medium → High	+36 %	+3
High → Xhigh	+90 %	+2
Xhigh → Max	+73 %	+2

High to Xhigh nearly doubles usage for two tasks out of a hundred. Those are two expert-level tasks, not two emails.

Anthropic's own documentation on the effort parameter says Max adds significant cost for relatively small gains on most workloads, and can lead to overthinking on simple or structured tasks. The code review company CodeRabbit reached the same conclusion in its own tests: Max gave mixed results, and the default level was the better choice overall.

Important is not the same as difficult

This is the actual mistake. We pick the level by how important a task feels. The proposal for the biggest client, the contract, the tax form. That one has to be right, so High, just in case.

But a text doesn't get more correct on High if Medium already gets it right. More thinking only helps when a task fails for lack of thinking. How much it matters to you plays no part.

The better question is: does the result have gaps? If not, the level was right. If it does, go up one level and try again.

Which level when

Level	When
Low	Enough more often than you'd think: anything whose result you can quickly check yourself
Medium	Everything else, including important and demanding work
High	Not a starting point. Only once Medium has demonstrably missed something
Xhigh	Long jobs with many steps: large research tasks, agent runs of half an hour or more
Max	Only when Xhigh has failed

Switching takes seconds: in chat through the menu next to the model name, in Claude Code with /effort low or /effort high. Getting used to a quick look at which row fits before you start saves more than any model switch. It pays off twice for recurring work, such as routine tasks AI handles in the background.

Making your allowance last

One level up, not straight to the top. If the result doesn't fit, try High first. The jump to Max is almost always the most expensive one with the smallest gain.
Sharpen the request before touching the setting. A shallow answer often comes from a vague request. A line like "This has several layers, think it through carefully" often gets more out of Medium than High does with a fuzzy description.
New chat for a new topic. With every message, Claude rereads the conversation so far. A chat with eighty messages makes every further question more expensive.
Tools only when needed. Web search and connected apps pull a lot of text into the chat. Switch them off when you don't need them.
Check Max results against High. Run the same task once on High and once on Max. If you can't find the difference, you have your answer.

The principle holds outside AI too: tuning the tool feels like progress but is often just a more expensive way of not doing the work.

Common questions

Where do I set the effort level in Claude?

In chat and the desktop app through the menu next to the model name, in Claude Code with the /effort command followed by the level. In Claude Code, Max only lasts until the end of the session.

Is Max the best choice if usage doesn't matter?

No. Max is slower, and according to Anthropic it can overthink simple or structured tasks, which can make the result worse. Max only pays off when Xhigh has failed on a hard task.

Why doesn't Anthropic publish exact numbers for plan usage?

Usage depends on conversation length, attached files, tools used, the model and the effort level, so Anthropic only gives relative figures between plans. The numbers in this post are an approximation based on measured cost per task.

Should I use a smaller model than Opus to save usage?

There's no need. According to Artificial Analysis, Opus 5.5 on Low costs about as much per task as Sonnet 5 on Low and solves more tasks than Sonnet 5 on Max. Staying on Opus and choosing the level deliberately saves more than switching models.

Set Claude to Low for your next email and compare the result with what you get on your usual level. If you can't spot a difference, you've just freed up allowance for a lot more work.

Todoist wants a project for every project. Martina gets by with one.

Fri, 25 Sep 2026 00:00:00 GMT

Martina runs her personal GTD system on the free version of Todoist. According to Todoist, that version allows five active projects, and Martina gets by with four: "Single next actions", "Projects", "Goals" and "Other lists". Her actual projects all sit in the second one. Preparing for a pole dance competition is a single task in there, and the steps towards it hang underneath as sub-tasks.

I hadn't expected the free version to stretch that far, and I told her she had bent Todoist well out of shape. That is true, but only from Todoist's point of view. From GTD's point of view, she has straightened it out.

A Todoist project is not a GTD project

A GTD project is an outcome that takes more than one step, and the project list is exactly what its name says: a list you read from top to bottom in your weekly review.

Todoist uses the same word for a container with its own page and its own entry in the sidebar. Set up every GTD project as a Todoist project, and your weekly review turns into clicking from page to page instead of reading a list. On the free version, the fifth container is the end of the line. I see this regularly in coaching: a sidebar that is longer than any context list.

The second trap is called "Today". The view shows what is due or overdue today. If you want to see a next action there, you give it a date that is not a real deadline, and before long the view consists of red entries nobody takes seriously any more.

Martina's setup sidesteps both traps: her projects are tasks, and a step becomes visible through a label, not through a date.

GTD with Todoist: projects as tasks

In the Todoist project "Projects", the sections are her areas of responsibility. Each GTD project is a task in there, and underneath it, as sub-tasks, hangs whatever came to mind when she created it, "just slapped it all in", as she puts it. The description of the project task holds the project notes, including links to files. For the competition prep, the link to the rulebook sits right in the sub-task that reminds her to read it.

With the sub-tasks collapsed, this one page is the whole project list, grouped by area. Everything else maps like this:

GTD	In Martina's Todoist
In-tray	Todoist Inbox, also filled by dictation on her phone
Project list	Todoist project "Projects", each GTD project a task
Areas of responsibility	Sections in that project
Project plan	Sub-tasks without a label
Next action	Sub-task with a context label
Context lists	Labels "To-do" and "At training"
Agenda, Waiting For	Labels as well, attached to the project
Project support material	Description of the project task, with links
Single actions	Project "Single next actions"
Routines	A section of their own in there, repeating from completion
Goals	A project of their own, same sections in the same order
Someday/Maybe, checklists	Project "Other lists"

The label is the switch

In Martina's system, a label decides which step is up. Whatever she could do right away gets "To-do" or "At training" and so appears in the label view, without losing its connection to the project. All the other steps stay in the project and show up nowhere else. When she ticks off a step in the label view, it is done in the project as well.

That is the separation between project plan and next action that gets lost so easily in list tools. The plan is allowed to be unordered and incomplete. The context list only holds what is possible right now. "Agenda" and "Waiting For" are labels in her setup too, so even an open question to someone stays visibly attached to the project.

Two context labels are enough for her, because in her personal life she does almost everything from home, and her working life runs in a separate system, in Microsoft To Do. There, too, projects are tasks with steps underneath. Using the three-dot menu, you can promote a step to a task of its own and move it to a context list, but in doing so it leaves the project. The connection that the label holds in Todoist breaks in To Do exactly where you need it.

Routines don't get a label

In the "Single next actions" project, Martina has a section for tasks that are never finished: the housework. That is where the kitchen worktop lives, along with the fact that James needs his water changed every week. I had taken James for a goldfish and wondered when the fish tank would get its turn. James is her robot vacuum, the kind that also mops.

These tasks carry a recurring date but no context label. They never appear on a context list, only in the date views "Today" and "Upcoming". When she sets aside time for her next actions, she doesn't want to see a worktop.

The trick is repeating from completion. In the English version of Todoist, you type every! 4 days into the date field; in the German version, it is alle! 4 Tage. Without the exclamation mark, Todoist keeps counting from the original date; with it, from the day you tick the task off. If you have been away from home for four days, you tick it off once and the task comes back in four days, not tomorrow. This is exactly what gave Martina trouble in Microsoft To Do: leave it for ten days, tick it off ten times.

For her, the date is a reminder, not a deadline. "Nobody's going to die because I didn't wipe down the kitchen worktop yesterday." I told her I like that calm, and I stand by it, with one caveat: it only works because the red entries are nothing but routines. Anything that has to happen on a particular day belongs in the calendar; at work, Martina enters such tasks as all-day events. Once a reminder date lands on a real next action, it shows up in the same red as the worktop, and red stops meaning anything.

The question for the weekly review

The flip side: a project whose labelled step is done also disappears from every list. That is why, in her weekly review, Martina goes through the project page and checks for each project whether a label is attached anywhere. If none is, a step gets one: for the competition prep, say, "Come up with choreo" with "At training", so it is up next time she is at the studio. Or the project rests for another week.

This is where I would tighten her system. Her project list also holds projects that are deliberately resting because they are not a priority right now. Without a label, they look exactly like a project whose next action she has simply forgotten, and the more resting projects pile up, the easier it gets to skim past the ones that ought to be moving. Anything deliberately on hold belongs on the Someday/Maybe list or in a section of its own called "Paused". Then "no label" means only one thing: a decision is missing here.

Goals in the same grid

One level up, Martina repeats the pattern. The Todoist project "Goals" has the same sections in the same order as "Projects", and the areas follow the wheel of life she has been filling in once a year for years. In her quarterly review, she opens both projects side by side in two windows and goes through them area by area: here is the goal, over there a project has to be running for it, or at least a next action. A goal with no project under it stays a dream, and this comparison is the moment you notice.

She cannot get a real link between goal and project in Todoist without giving up the grouping by area. She has tried. The same grid in two windows manages without a link, and it is enough.

Where this typically falls apart

Every GTD project becomes a Todoist project. The sidebar grows, the weekly review turns into clicking through pages, and the free version is full after five projects. A project list has to be readable in one go.
Ticked off, but no next label. Complete the labelled step and you leave behind a project with no visible next action, and the earliest you notice is in the weekly review. Labelling the next step the moment you tick one off saves the project up to a week of standstill.
Reminder dates creep onto next actions. Then "Today" mixes routines and real deadlines with mere intentions, all in the same colour. Only what is genuinely due on that day gets a date, plus the routines.
Labels just in case. Create a label for every conceivable situation and you end up maintaining lists with almost nothing on them. Martina has two for her personal life and only splits one when it gets too full.
The project plan outgrows the limit. According to Todoist, a cap of 300 active tasks per project applies on every pricing tier, sub-tasks included, so upgrading won't help. By her own account, Martina is just under it. If you keep all your projects in one Todoist project, move dormant ones out and put thoughts that are not actions yet into the description instead of sub-tasks.

Common questions

Does GTD with Todoist work on the free version too?

Yes, as long as the GTD projects are not set up as separate Todoist projects but as tasks with sub-tasks in one shared Todoist project. According to Todoist, the free version allows five active projects, and four are enough for a complete system: single actions, projects, goals and other lists. The context lists come from labels, which are available in the free version as well.

How do I set up context lists in Todoist?

With labels. Each label stands for a context, and the label view shows every task with that label across all projects. If you label only the step in a project that is possible right now, you get a context list that holds nothing but next actions.

How do I spot a project without a next action in Todoist?

In the weekly review, by checking for each project whether one of its steps carries a label. A label such as "Waiting For" or "Agenda" counts too, because then the project is not standing still either. If there is no label at all, either a step gets one, or the project deliberately moves to the Someday/Maybe list.

How do I make a task in Todoist repeat from the completion date?

With an exclamation mark in the date field: "every! 4 days" in the English version of Todoist, "alle! 4 Tage" in the German one. Without the exclamation mark, Todoist keeps counting from the original due date; with it, from the day you tick the task off. For household routines that is the better choice, because a task that has been left lying around does not fall due again straight after you complete it.

How many tasks fit into a Todoist project?

According to Todoist's overview of usage limits, 300 active tasks per project, sub-tasks and recurring tasks included, and that applies on every pricing tier. Completed tasks do not count. If you keep all your GTD projects in a single Todoist project, keep the project plans lean and move dormant projects out.

Go through your project list once with a single question: does this project have a step with a label? Where it doesn't, decide on the spot, for a labelled step or for the Someday/Maybe list. Keep count of how often you had to decide. The number says more about your weekly review than about your projects.

Which tool it ends up being is rarely the hard question. The hard part is separating project plan and next action inside it so that the lists still hold true six months from now. That is exactly what I set up with you in the Workflow Setup, in Todoist or in whatever tool you already have. Write to me if your system keeps growing in the sidebar and shrinking on the lists.

Two keys empty the Gmail inbox. There is no key for the rest.

Fri, 18 Sep 2026 00:00:00 GMT

I hit send. Two seconds later I realise the link is missing. Down in the bottom left corner, “Undo” sits there for thirty seconds, I click it, the draft is open again, and the mistake never happened. Perhaps our brain needs the click on send to see what is missing.

The send cancellation period is shorter out of the box. You find it through the cog in the top right, “See all settings”, the “General” tab, and pull it up to thirty seconds. It is the most harmless of the settings on this list.

Why the defaults work against processing

Gmail arrives in the state that works for someone who reads mail. Every incoming message announces itself on your desktop. The inbox comes pre-sorted into categories. After every action the view jumps back to the list. Keyboard shortcuts are off.

For a reader, all of that fits. For someone who wants to empty an in-tray, every single point is the wrong way round. Processing needs one in-tray instead of five, no interruption, one movement per item, and at the end the decision has to sit somewhere other than the mailbox. The difference between processing your email and working it is not settled in your head alone. It also depends on which of the two your tool makes easier.

And one quirk that everybody keeps forgetting: Gmail runs in a browser. The settings page is a web page, so Ctrl+F works. With the number of options on each tab, that is the fastest way to a particular one.

Setting up Gmail for GTD

Keyboard shortcuts sit under “General” and are off out of the box. This is the most important setting of the lot, because without it everything else stays mouse work. Four keys carry most of it: # deletes, e archives, r replies, f forwards. Plus Shift+T, which turns the open message into a task with a link back to the mail.

Desktop notifications are on by default at Google. Off. One alert per incoming mail is the exact opposite of processing in blocks, and it is guaranteed to arrive during the one decision that needed quiet.

Categories live on the “Inbox” tab. The detection is good, and that is precisely the problem: one in-tray becomes five, and you only notice the four extra ones once something has piled up in them. The spam filter is untouched by this, it works on a different level. At Google, Promotions is a pre-sort; real spam still ends up in the Spam folder.

Offline has a tab of its own. Gmail then keeps part of your mail in the browser cache, you can read, reply and decide on the train, and everything goes out the next time you have a connection. The part almost everyone forgets: without internet you cannot type in the address that would get you there. Set a bookmark while you are still online. Otherwise you have a feature switched on that you cannot reach. In Chromium-based browsers this works reliably; with others it is worth a test before you rely on it.

Templates sit under “Advanced” and are off. They are useful for more than boilerplate: if you let an AI reach into your mailbox, you get drafts that are hard to produce without this setting.

The reading pane I would leave off, even though people coming over from Outlook look for it first (“Inbox” tab, further down). It shows you the current message and the list of all the others at the same time. That is handy when you are searching and in the way when you are deciding, because the next one is always already sitting in the corner of your eye.

The two settings that are faster than your decision

Two of the most popular settings deserve a second look. Both are good. Both speed up exactly the point where the method has to be slow.

Auto-advance

It sits under “Advanced”. After you archive or delete, Gmail does not jump back to the list but straight into the next message. With a full inbox in the morning, that feels great.

The side effect: deleting and archiving are one-key decisions, and one-key decisions are exactly what auto-advance makes faster. “This is a next action, context phone, and it comes after Tuesday’s quote” is not a one-key decision.

That is easy to miss, because Shift+T is so quick. The task is created in a second — but what it says is the subject line of the mail, and a subject line is not a next action. “Re: Quote wk 38” has to become “Call the supplier: settle the discount on item 4”, and that one rewrite is the whole job. It takes a few seconds longer than any keystroke, it takes your head — and while you are in it the next message is already open and pulling at you. Go through thirty messages with auto-advance and you have cleared thirty messages. What you have processed is the subset that one key could finish.

The counterweight is not a setting but a rule: as soon as a message needs more than one key, the next movement is not archiving, it is the entry. Shift+T, rewrite the subject into a real next action, put it on the right context list — and only then e. Archiving is the full stop at the end of the decision, not a substitute for it. If your list system does not live in Google Tasks, you really do leave Gmail at this point; the stop is the same one, it just takes longer.

Send and archive

“General” as well, hidden out of the box. When you reply, the button replaces the usual send: the reply goes out, the message you received moves to the archive, the inbox shrinks by one.

What it does not do: it does not remember what was in your reply. A large share of the replies I write contain a question, or a commitment somebody else has made. The moment you press send and archive, something is waiting, and it is written down nowhere. The original mail is in the archive, the reply is in “Sent”, and neither of those is a list. The Waiting For list carries it, and in Gmail it costs you one extra movement: Shift+T before you send, and then the item sits on a list with a link to the mail and is free to disappear where it belongs.

Decision	Keys in Gmail	What is in your system afterwards
trash	`#`	nothing
reference	`e`	nothing, the archive is the filing
under two minutes	`r`, send and archive	nothing, unless you asked something
delegated	`Shift+T`, then `f`	an entry on Waiting For
next action	`Shift+T`, then `e`	an entry on the right context list
more than one step	`Shift+T`, then `e`	a project plus exactly one next action

The right-hand column is the part there is no key for.

What AI in your mailbox takes over, and what it doesn’t

Having a long client email summarised, along with a request for the tasks and open questions it contains, works well and saves real minutes when the thing runs to four screens. What comes out, though, is a summary, not a decision. Whether the three extracted points are your points is not in the mail.

With appointments taken out of mail I would be more careful. I had an event created from a message that said “next Wednesday”. The entry landed in the calendar a week too late, and one more click would have sent the invitation out. Relative dates are ambiguous even between people. If a date is in the mail, the result is usable. If there is none, you read it yourself.

The line does not run between “AI can do this” and “AI cannot do this”. It runs between a handling step and a decision. Recurring handling steps you hand over, and you hand them over completely. The decision about what a message means for you stays with you.

Where this typically falls apart

The shortcuts are switched on but never practised. After two days your hand goes back to the mouse. Start with two keys, # and e, and ignore the rest for a week.
The empty inbox becomes the metric. It is a by-product. Make it the goal and you optimise for the keys that empty, which are exactly the ones that leave nothing on a list.
Archived counts as decided. Archiving means: no longer in the in-tray. It does not mean anybody has thought about the content.
Google Tasks turns into a second system. Shift+T is tempting because the link back to the mail works so well. If your list system lives somewhere else, you now have two places for next actions and you only look in one — the usual start of a system you have to pull back together later.
A hidden list goes missing without a trace. Google Tasks can hide lists, but it cannot collapse them. When you look through them, a hidden list looks exactly like a list that does not exist, and that state is stubborn.

Common questions

Which settings do you need to set up Gmail for GTD?

Four carry most of it: switch keyboard shortcuts on, switch desktop notifications off, switch the inbox categories off, and set the send cancellation period to thirty seconds. All four sit in the cog menu under “See all settings”. The rest is polish and taste.

Do I lose spam protection if I switch the Gmail categories off?

No. The Promotions, Social and Updates categories are a pre-sort inside the inbox, not the spam filter. Messages recognised as spam end up in the Spam folder regardless. Switching the categories off only brings everything back into a single in-tray.

What does the “Auto-advance” setting do in Gmail?

After you delete or archive a message, Gmail opens the next one immediately instead of jumping back to the list view. The setting is on the “Advanced” tab and is off out of the box. It speeds up clearing considerably; for messages that lead to an entry in a list system you still need a deliberate stop.

Is Google Tasks enough as a complete list system?

For a split setup yes, for a complete one probably not. The view is built as a board, with several context lists you scroll sideways all the time, and lists can only be hidden, not collapsed. If you work in Google Workspace professionally and keep your private system elsewhere, the handful of lists on the work side will do the job nicely.

Does Gmail work without internet?

Yes, if you switch on offline mail on the “Offline” tab. Gmail then stores messages in the browser cache, you can read, reply and process, and everything goes out as soon as there is a connection again. Set yourself a bookmark beforehand, because without a connection you cannot reach the address by typing it.

Look at the mail you sent over the last seven days and mark every message in which you asked a question or asked for something. For each one: is the item written down anywhere outside the mailbox? The gap is not a discipline problem, it is the place where your tool was faster than your decision.

If the settings are in place and the inbox still refuses to empty, it is rarely Gmail. It is the missing stretch between inbox and list. That is exactly what I build in the Workflow Setup, with your tools and your own work. Write to me if you work in Google Workspace and your system stops at the mailbox.

I had the most beautiful GTD system in the world. It did nothing for me.

Fri, 11 Sep 2026 00:00:00 GMT

A few years ago I had the most beautiful GTD system I have ever come across. It ran in OmniFocus, and every task trailed a little row of emoji. I had just learned that there is more than one clock emoji: one for the full hour, one for quarter past, one for half past. So a task that takes fifteen minutes got the quarter-past clock. On top of that, symbols for whether something needs a lot of thinking, a little thinking, or none at all because it is quick. It looked colourful, it felt creative, and I liked showing it off.

Someone I demonstrated it to looked at it for a while and then asked: what does this actually do for you?

My answer came quickly. The emoji tell me how long a task takes and how much head it costs me. The follow-up was uncomfortably precise. If you already know while writing it down that vacuuming takes fifteen minutes, you will still know it later when you only read the line. And then the second one: how long do you need to pick the right emoji and stick it on? Ten seconds. On every task, every day.

The system collapsed in on itself during that conversation. It had a price per capture and a benefit of zero.

Why systems only ever grow in one direction

The overstuffed systems I get to see almost never came out of one wrong decision. They came out of a long run of right ones. You read about a trick, you try it, it helps you that week, and it stays. Nobody puts an expiry date on a new list.

On top of that, lists rarely start out wrong. They turn wrong. When context lists were invented, the computer stood in the office, which made it a place: if you wanted to be at the machine, you had to travel there. Today the phone in your pocket is a computer, and the context no longer answers any question. I kept an office list for years, even though I was self-employed and working from home. The list was not set up wrong, it simply became pointless at some point, and I carried it around anyway.

You pay for that in two places, and both are quiet. At capture, every extra decision costs you seconds. In your weekly review, every list costs you attention, even when there is nothing on it.

The question that decides every list

There is one question that settles this in a few seconds. When did I last go into my system deliberately, for this one list alone, to find something?

Not: when did I last see it. Not: when did I last write something into it. But: when did it take a decision off my hands? For the pretty tags, the honest answer is usually never.

Underneath sits a distinction that rarely gets said out loud. A context is a hard constraint. It answers whether you can do a thing here and now at all. An attribute is a preference. It answers whether you want to do it right now. Constraints earn their own list, because in a particular moment they do the searching for you. Preferences only earn one if you genuinely filter by them.

Some people do. If you regularly have twenty minutes of dead time before your next appointment and you actually want to use them, a time estimate is not decoration, it is the filter that makes the list usable. So the test is not a matter of taste. Two people with the same tag can come out of it differently.

List or tag	What it answers	Deserves its own list
Agenda per person	Who am I seeing or speaking to next?	Yes, the window is short
Errands per place	Am I there right now?	Yes, otherwise you go twice
Waiting For	Who owes me something?	Yes, otherwise it vanishes quietly
Computer	Am I sitting at a machine?	Once yes, now almost never
Low energy, 15 minutes	Do I fancy it right now?	Only if you really filter by it

For most people, a handful of context lists is what survives, plus one agenda for each person they deal with regularly.

Why "not used in three months" is the wrong test

The moment you start clearing out, you want a rule that makes the decision for you. The obvious measure is time: anything you have not needed in three months goes. I think that is wrong, and it is wrong precisely for the lists that are worth the most.

Take a list for the hardware shop. If you keep one, you might need it three times a year. By the time rule it would have to go, and that would be a mistake, because in exactly the moment when you are standing there anyway, it saves you the second trip. The same goes for everything that hangs on a rare opportunity: the tax paperwork, the appointment at the doctor's surgery, the person you see twice a year.

The better measure has two parts. Does the list hold up in the moment it was built for? And what does it cost you in all the other moments? A place list costs almost nothing, because when you write the thing down you already know that drill bits come from the DIY shop. A time estimate costs you a few seconds of thinking at every single capture. Frequency is a measure for attributes, not for opportunities.

Simplifying a GTD system means deleting, not reshuffling

When a list disappears, its tasks do not disappear with it. That is where the clear-out usually gets stuck, and that is where it gets interesting what the list really was.

If every task immediately finds another list that already exists, the old one was a duplicate and you have lost nothing. If a task finds no home, that is the most valuable result of all: either you are missing a list for a situation that genuinely exists, or the thing has been dead for ages and has to beg for a place before anyone notices. And if you cannot decide, the task was never processed. It is sitting there as a topic, not as a next action, which is why it fits everywhere and nowhere.

In my own system the test caught a list that stands in every textbook: Out and about. I kept it because it gets recommended. At some point I noticed that I have to look at it every single day, because I am out and about every single day, and that it holds exactly one task. I need to print a page and I have no printer. My idea behind it was that while I was out, I would happen to walk past a printer that was desperate to print something. That is not going to happen. The task was on the wrong list; it belongs to the person who has a printer, and therefore on their agenda. Without it, Out and about has no content left and no reason to show up every morning.

With projects the same move has a different aim. Anything that is guaranteed to sit untouched for the next month or two does not get deleted, it goes on the Someday/Maybe list, and once a month you check what is allowed back.

The meta layer costs you at capture, not at search

The second price of all that superstructure is not paid when you search. My rule of thumb: if capturing something takes longer than ten seconds, the friction is too high. Honestly, I now think ten seconds is generous.

The moment in question is always the same one. I sit down on the sofa in the evening, and the second I am sitting, something good occurs to me. All that matters is holding on to that thought in a way that lets me find it again tomorrow. "Mum" is entirely enough. You will still know tomorrow why she came to mind.

If you have to decide in that moment which list the thing belongs to, how long it takes and which tag it gets, two things happen. Sometimes you just do not do it, and the thought stays in your head. And when you do do it, you rob yourself of the chance to process it properly.

The obvious objection is fair: you cannot assign anything in ten seconds. True, and that is exactly why assigning does not belong there. Capturing is not clarifying, and clarifying is certainly not organising. If you type the idea straight into the right list, you skip the step where the real question gets asked: what is this, and what is the next action?

Where this usually goes wrong

The clear-out happens in the middle of the work. On a Tuesday morning you notice that a tag does nothing for you, and you strip it out everywhere on the spot. It feels productive, and it is the reason nothing is finished by the evening. The decision belongs in the weekly review, the tidying up in an appointment of its own.
Tinkering gets banned instead of scheduled. Building on your system is fun, and there is nothing wrong with that. Ban it and you will do it anyway, just secretly, at nine o'clock on a Monday morning. Set up a project that is named for what it actually is, give it a fixed slot, and put it behind the work, not in front of it.
The tasks from the deleted list end up in a leftovers box. A catch-all list with no criterion is not a simplification, it is the old state under a new name. When a task finds no home, that is a result, not a problem to be filed away.
The big cull is followed immediately by the next rebuild. Whether a structure holds up shows after a few weeks at the earliest, and most clearly in the weekly review. Bring in the next concept in the meantime and you will never know which one did the work.
Afterwards the system looks better than it works. Tidy tiles, good names, matching colours: the effect is real, it just sits in the wrong place. The test stays the same. Which decision did this list take off your hands this week?

Common questions

How do I know it is time to simplify my GTD system?

The clearest sign is that you push the weekly review back even though you have the time. Then there is the hesitation at capture: if you would rather keep a small thing in your head than type it in, the friction is too high. And if you have lists you only look at because they exist, the system is no longer carrying you, you are carrying it.

How many context lists make sense?

There is no right number, but there is a test that holds: every list has to take a decision off your hands in a specific situation. For most people a handful survives, plus one agenda for each person they deal with regularly. Anything that only describes how a task feels is an attribute, not a list.

What do I do with the tasks when I delete a context list?

You move them one by one before you remove the list, and only onto lists that already exist. If a task finds a home straight away, the deleted list was surplus. If it does not, that is the interesting exception: either a list is missing for a situation that genuinely exists, or the task is done and nobody ever ticked it off.

Is it wrong to put a time estimate on tasks?

No, as long as you filter by it. The estimate costs you a few seconds at every capture and only pays off if you really do go looking for short tasks while you are waiting around. If you do that regularly, keep it. If you cannot remember a single time you did, you are maintaining it for the look of it.

When should I tinker with my system, if not on the side?

In a planned slot that does not sit at the start of a working day. A fixed slot after lunch works well, or a monthly appointment, so the urge has somewhere to go. What matters is the order: first the work that is due, then the system you enjoy fiddling with.

Open your system and find the one list you have only been looking at for weeks because it is there. Do not delete it. Write down what is on it instead, and next to each item, where that thing would go if the list no longer existed. Whatever you cannot place is your answer.

If the clear-out leaves you with the question of whether a list can really go, that is rarely about the list and usually about the structure underneath it. That is exactly where the Workflow Setup starts: one pass through your system, until all that is left in it is what you actually use. Write to me if you have spent years working with a system that grew along with you and now weighs more than the work inside it.

When I started with GTD, I had no idea why I was on this planet

Fri, 04 Sep 2026 00:00:00 GMT

I ended up with GTD because I was drowning in work. Not because I wanted to know why I was put on this planet. Back then I had never given that question a thought. That is exactly where several other productivity methods lost me: you open the book, and before anything practical happens, you are supposed to work out what you are here for. After that it is about how to get there. At that point I had an inbox that had long since got away from me, and no idea what came first on Monday.

It is the same for most people who start out. And yet the idea persists that you have to sort out the upper levels first. Anyone who has not sorted them feels unfinished, and so leaves them alone entirely.

The GTD horizons of focus aren't a build order, they're a checking order

The image that does the damage is the pyramid. At the bottom, the level you are standing on: next actions, calendar, the day you are in. Above it five horizons, from projects up through areas of responsibility and goals to your vision and the question of what all of this is ultimately about. It gets drawn from the top down, and that is exactly how it gets read: as the order in which you build the thing.

It isn't a build order. It's a checking order. At heart, the horizons above your projects ask a single question: is what you are doing right now going in the direction you want to go? A checking question needs something to check. If nothing is running down below, you are checking fog.

There are exceptions. A colleague once had a client, a priest, who turned up with a thick book of handwritten notes: everything he had in mind for himself and his parish, spelled out over years. For him, starting at the top made sense, because there was already something up there.

This is where I would argue against the convenient explanation. The reason to start at the bottom isn't that most people have no vision written out. The reason is that every decision you make at the top has to arrive at the bottom, or nothing comes of it. Even the priest with the full book needs a working list. Otherwise his vision stays exactly that: a full book.

Four rounds of why, from the front door to your vision

The way up is shorter than most people think. You don't need an away day and you don't need three weeks in a monastery. You need an existing list and one question: why.

Take an item that has been staring at you for a while and ask it that question four times in a row. A chain like that might look like this.

Horizon	The question it answers	Answer in the example
Next action	What am I actually doing?	Paint the front door
Project	What outcome do I want?	Living in a nice house
Area	Which part of my life does this belong to?	Home
Goal	What should be different there in a few years?	A house that barely needs any work
Vision	What is this a part of?	A home in old age I can rely on

Anyone looking closely at the method will spot a problem in the second row straight away. Living in a nice house isn't a project. It never gets a tick, it's a state.

That is precisely where the value of the exercise lies. For a great many tasks, the first why doesn't land on a project at all, it lands straight on an area. That is not a glitch, it is a finding: the task is not attached to an outcome, it is attached to a standard you hold for that part of your life. For the odd task, that is perfectly fine. But if half your list looks like this, you are working on states instead of outcomes, and that is why the list never feels any shorter.

The one horizon where nothing ever gets ticked off

Paul Martin asked me the best question about this picture I know of: why do goals sit above areas of responsibility, when the areas are the space in which those goals happen in the first place? Shouldn't the two levels swap places?

The honest answer about the order is: because that is how the method defines it. The better answer is that the question exposes a flaw in the picture. An area isn't a floor at all, it's a column. Draw the areas as vertical lines running through the whole pyramid and it works again: family, health, work and finances each have their own next actions, their own projects, their own goals and their own share of your vision.

A practical difference hangs on this. An area is the one horizon where nothing ever gets ticked off. A next action gets done at some point, a project gets completed, a goal gets reached. An area is a part of your life where you want to hold a standard, and that is never finished. Which means the question is a different one. With a goal you ask whether you will reach it. With an area you ask whether something is running here right now that matters to you.

Coaching has the wheel of life for this: a circle divided into areas of life, each segment coloured in according to how full it currently feels. At the end you look at a wheel that isn't one, it's jagged. If you sort your projects and goals by area, you get the same information without the coloured pencils, because area by area you can see where nothing has happened for months.

An empty area isn't an accusation. I regularly have areas where nothing at all is running for a while. That is a decision, not a gap. The only difference is that I make it deliberately once a quarter.

Thinking upwards means changing something down below

The why chain is half the work. The other half is the way back down.

Years ago my friends complained that they never heard from me. They were right, I was bad at it. Today the same people ask me how I actually manage to stay in touch with everybody. The solution is unspectacular: every weekly review carries the question of who I have not heard from the longest, and that person lands on my calls list. An area that matters to me has been translated into a recurring question that spits out a concrete name every week.

That is the test for every half hour you spend up at the top. If afterwards no project has appeared, none has been struck off and no recurring question has been added, it was pleasant reflection and nothing more.

And the results need somewhere to live. If areas and goals sit in a document of their own that you open twice a year, they do not exist in your system. They belong where your review is looking anyway, right next to the projects. Otherwise the urgent wins every time.

The quarterly review is shorter than the weekly review, not longer

The weekly review asks whether your system still matches your reality. You go through the context lists, you go through the projects, and for each one you check whether a next action is captured. The quarterly review asks the reverse: whether your reality still matches your goals.

That is why it takes less time. I do not look at a single next action in it, only at projects, goals and areas, and for each area I ask myself whether what matters to me is actually running, or whether I am just pushing tasks along. My weekly review has a fixed slot on Friday. The quarterly review is a recurring task every three months, and it is allowed to sit for a week until I am at home and have some quiet. With that one I deliberately ease off on the discipline, because it does not take well to the wrong surroundings.

Where this typically goes wrong

The horizons live outside the system. Areas and goals in a separate note, on a whiteboard or in a presentation. The review never looks there, so nothing ever happens there.
Thinking at the top, changing nothing at the bottom. After the why chain, no new project appears and none is struck off. Then the exercise was a feeling, not a result.
The quarterly review gets planned like a big weekly review. If you block half a day for it, you will never find the time. Count on a fraction of that, because you are leaving out an entire level.
All the areas at once. That turns the exercise into a project, and the project migrates to Someday/Maybe. Take on one area, no more.
An empty area gets read as failure. An area with no running project only becomes a problem when you have not deliberately left it empty.

Common questions

Do I have to build the GTD horizons of focus from the top down?

No, for most people the workable route is from the bottom up. The level of next actions and projects already exists, and the horizons above it are checking questions that need something to check. Starting at the top is only worth it if there is already material written out up there.

What is the difference between an area of responsibility and a goal?

An area of responsibility is the space in which something happens, a goal is an outcome inside it. The area of family is not a goal, it is never reached and never ticked off, it describes a standard you want to hold. A goal within that area, by contrast, has a recognisable state that tells you you have arrived.

How long does it take to think one area of responsibility through?

Give yourself half an hour per area and take on only one. Many people put off the work on the upper horizons for years because they expect it to take a multi-day retreat. After half an hour you usually know whether something that matters to you is running in that part of your life.

How does the quarterly review differ from the weekly review?

The weekly review reconciles your system with your reality, the quarterly review reconciles your reality with your goals. In the weekly review you go all the way down to the level of next actions, in the quarterly review you only look at projects, goals and areas. That makes the quarterly review the shorter of the two.

What do I do with a task that has no answer to the why question?

Then it has no place in your system, and that is a useful result. Either you strike it out, or it belongs on the Someday/Maybe list, because it is a wish and not a commitment. Either one is better than dragging it along between real commitments for another few months.

Take on a single area this weekend, ideally one of the quiet ones: friends, personal development, home. Half an hour, a sheet of paper, four rounds of why. If a new project is standing there at the end, or an old one is crossed out, then the horizon has changed something at the bottom, and that is exactly what it is for.

The upper horizons rarely fail for lack of thinking. They fail because areas and goals have no place in the system where the review would ever find them. That is exactly what I set up as part of the Workflow Setup, from the context lists to the levels above them. Write to me if your system runs cleanly at the bottom and nothing is connected at the top.

Nobody emails you to say you have stopped calling your friend

Fri, 28 Aug 2026 00:00:00 GMT

"You have put on weight. You need to do something about it." That sentence sat in the inbox of a lawyer who has been working with GTD for over twenty years, in among deadlines, case notes and dictations. He put it there himself. His active project list now has losing weight on it, right next to the live cases, and beside that a second project about his own social media use. Writing sentences like that down is, he says, the hardest part of capturing. And, looking back, the most valuable.

You never have to chase work material. It has a sender. An email lands in your inbox, a client rings, a deadline sits in a contract, a colleague asks for the second time. The channel does half the work for you: the material comes to you whether you want it or not. Personal material has no such channel. It exists only as a thought, and thoughts turn up in the shower, at a red light and at three in the morning. That is not when you have your system open.

This is not a discipline problem, even if it feels like one. It is a problem with the way in. Whatever never gets into the system does not lose out to work. It never even enters the race. You did not decide against ringing your friend, against the check-up, against the conversation with your father. Those things were never on the table. They still cost you capacity, just at a point where you cannot decide anything. In coaching I see the pattern all the time: everything captured, clarified and sorted on the work side, while the same intentions live on in the private half as sentences that start with "I really should".

The mind sweep is the way in that personal life never got

The mind sweep is the only kind of capture that is not triggered from outside. Everything else in your inbox was sent to you by someone. That is exactly why it is the tool for the topics with no sender, and exactly why it is the first thing to go when time gets tight.

In practice it falls down at the trigger list. The lists that get passed around are sorted by work: projects, colleagues, clients, deadlines, equipment, travel. You work through them, end up with twenty entries, feel thorough, and have not touched half your life. You have to build the second half yourself. Five categories are enough:

Body: check-ups, prescriptions, the ache you have been carrying around for months, exercise, sleep.
People: contacts that have gone quiet, conversations you still owe, promises you have made, an argument nobody ever ended.
Home: broken, planned, put off, uninsured.
Money: contracts, subscriptions, tax, anything you have not looked at in years.
Intentions: whatever you once set out to do and never looked at again.

The result belongs in the same inbox as your work. The lawyer has exactly one inbox for both, work and private, and therefore only one shortcut on his phone: two voice commands for dictation, one German, one English, both callable by voice while he is driving. The dictation is transcribed, gets a timestamp and lands as a file in the inbox folder. Nothing bought, no second app, no data on someone else's server. The point is not the technology. The point is that there is no longer a decision to make about where a thought belongs.

Write down the sentence, not the solution

Capturing and clarifying are two separate steps, and if you merge them, personal material is exactly what you lose. The lawyer describes his own beginner's mistake like this: he wanted to write something down, started clarifying it while he was writing, it took too long, and in the end he did not write it down at all.

The effect is not evenly spread. A work task almost writes itself, because it arrives with its outcome attached: quote by Friday, call him back, check the contract. A personal topic arrives as a feeling. "Something with my father" is not a task, and once you are in the habit of formulating a next action while you capture, you reliably filter out everything that cannot be turned into an action in five seconds. That is always the personal stuff.

So: the raw sentence, in the words it turned up in. "You have put on weight" is a good inbox entry. "Find a nutritionist" is a premature answer to a question you have not even asked yet.

The level it sits on decides how often you look at it

Personal topics almost always end up one floor too high, and that is where they disappear. "Health" is an area of responsibility. "Back to my old weight within a year" is a goal. Both sound like clarity, and both make sure you will not look at the topic again any time soon.

Level	How it sounds	When you next look at it
Area of responsibility	health	when you go through your horizons
Goal	back to my old weight within a year	when you next make time for your goals
Project	lose weight, with an outcome and a next action	in every weekly review

The lawyer said the decisive sentence himself: at project level, the topic competes for attention. At goal level, it waits for three months to pass until he happens to look at his goals again. That is why his weight sits as a project on the same list as his cases. The criterion is unspectacular: a concrete outcome, several steps, achievable within a year. If all three apply, it is a project, even if it sounds like a new year's resolution.

That does not mean everything has to go on the project list. Some of it becomes a Someday/Maybe when you clarify it, some of it becomes an appointment, some of it becomes an entry on one particular person's agenda, because personal plans nearly always need a second human being. And some of it goes in the bin. A deliberate no is a clean outcome. The weekly review then goes through everything you kept, once a week, and that is the only reason the project level works at all.

The sentence I never click on

There are two places in my system with sentences that are not tasks and never get done. They sit above lists I look at several times a day anyway. I do not click on them, and most of the time I do not even consciously read them. I catch them out of the corner of my eye, and that is the whole point. The things that matter most to me rarely make any noise, so the reminder makes the noise for them.

The lawyer does the same. Above his context list it says that his wife matters more than anything listed below, and next to it a note that he has to process his inbox, because without processing no review works. It costs nothing; a sticky note on the monitor does the same job. It only works for what you already know. For everything else you need the sweep.

Where it usually goes wrong

You clarify while you capture. Then you only write down what translates into an action straight away, and the personal things fall through the net every single time.
You capture and then never clarify. A mind sweep with no processing afterwards leaves you worse off than before, because your inbox turns into an archive of your own omissions. The sweep is only finished when every entry has an outcome, and binning it is one.
The trigger list stops at the office door. For most people the system stops at exactly the same place, and nobody notices, because the work third of it runs impeccably.
Personal material sits one floor too high. Phrased as an area of responsibility or as a goal, it comes past every few months and loses to work every week in between.
Two systems, one review. If you keep work and private life in separate systems, you end up maintaining only one of them, and it is rarely the private one.

Common questions

How does a mind sweep work for personal topics?

You sit down with a trigger list and write down every thought the categories set off, without judging it and without formulating a solution. For personal topics you need your own triggers, because the usual lists are sorted by work: body, people, home, money, old intentions. It all lands in the same inbox as anything you capture at work, and it is clarified afterwards in the normal way.

How often do I need a mind sweep?

A full mind sweep pays off whenever something bigger changes: a new role, a house move, an illness, a project that ends. In between, the short pass in your weekly review is enough, as long as your day-to-day capturing works cleanly. If you notice the same thoughts coming back to you several times a day, you have put it off too long.

Do personal plans belong in the same system as work?

At the level of the inbox, yes. Separate systems mean that one of them gets maintained regularly and the other does not, and in practice it is the private one that drops off. Separate lists or separate views, on the other hand, make good sense, as long as both come up in the same weekly review.

Is losing weight a project or a goal?

It is a project as soon as it has a concrete outcome, needs several steps and is achievable within a year. Phrased as a goal, it moves up to a horizon that most people only look at every few months. On the project list it stands among all your other plans and has to hold its own there, which is precisely the point.

What do I do with thoughts that have no next action?

You capture them anyway and clarify them afterwards like everything else: some become a project, some a Someday/Maybe, some an entry on one particular person's agenda, and some go in the bin. A deliberate no is a clean outcome. The only thought that comes back is the one that never reached a decision.

Take your project list and go through it with a single question: who sent me this? For almost every entry someone will answer, a client, a boss, a deadline, a contract. Count the entries where nobody answers. That number is the most honest measure you have of how far your system actually reaches.

If everything from work is in your system and nothing from your private life is, what is missing is rarely discipline and almost always the way in. That is exactly what I build with you: every channel into one inbox, and a review that takes hold of both halves (Workflow Setup). Write to me if your system currently stops at the office door.

A goal with no project under it is a dream

Fri, 21 Aug 2026 00:00:00 GMT

Years ago I watched every series I could get my hands on. There was no Netflix yet, you got the seasons from friends, and I worked my way through them, every evening. At some point I asked myself the question I have asked about almost everything since: why, actually? I couldn't find a why. Not a social one, I never talked to anybody about those series. Not the simple one either, that I had too much time on my hands. The only thing I came up with: you can become world champion at watching series. I didn't want that. From then on I barely watched a series at all and went back to reading a great deal more.

In GTD the why question isn't a motivational trick, it's a lift. You ask it about a next action (why does it say "buy milk" here?), you land on a project, you ask again, you land on an area of responsibility, and eventually you're standing on Horizon 3, at your goals. Most people have something written there. It has just been sitting there unchanged for quite a while.

The level that never speaks up on its own

Every level below your goals has a trigger built into it. You look at next actions the moment you ask yourself what to do now. Projects speak up in the weekly review at the latest, and also whenever a context list runs dry: a project without a next action stands out. Areas of responsibility speak up when something in your life shifts, a new job, a move, a child leaving home.

Goals have no trigger. Nothing runs dry, nothing turns red, no list stalls. A goal nobody has touched for six months looks exactly like a goal that is being worked on right now: one line of text. The most common state at this level is therefore not the badly worded goal, it's the declared one. Neatly written down, properly filed, and underneath it nothing happens. In coaching I see this regularly: the project list is well kept, the weekly review has been running for years, and the goal list dates from the year the system was set up.

The done test for GTD goals and areas of responsibility

Before you start working on wording, sort the levels. There is exactly one useful question for that: what actually gets finished here?

Level	What gets finished	How you recognise it
Next action	a single action	you could film it with a camera
Project	a desired outcome	several steps, one ending
Area of responsibility	nothing	a standard you hold
Goal	an outcome made of several projects	projects fit underneath, not actions
Vision	open	you don't yet know how you get there

The interesting entry is the third one. The area of responsibility is the only level in the whole model where there is nothing to complete. It's a standard: being healthy, keeping a good relationship, keeping the flat liveable. Nothing there is ever closed off, which is why there is never a tick beside it.

This is exactly where most goal lists fall apart. "I want to be fit" feels like a goal because it sounds important. It just never gets finished, so nothing on the goal list ever moves, so at some point you stop opening the list. Filed one level down, the same sentence becomes useful: the area is called fitness and health, and inside it there may well be a goal that has an ending.

My own areas have barely changed over the years, only the names got more precise: adventure, leisure and creativity. Development. Family and friends. Finances. Fitness and health. Self-employment. Home. My children had an area of their own for years. They are 17 and 18 now, the older one is moving into a flat of his own, and nobody wants to go to the theatre with dad every other week any more. The area hasn't disappeared, it has been absorbed into family and friends. Areas are allowed to do that.

A project without a goal is a hobby

I say that sentence often, because it sticks. It's still useless as a diagnosis, and I'll correct it here myself, straight away.

Most projects in a well-run system contribute to no goal at all, and that is perfectly normal. My tax return contributes to none of my goals. If I hadn't done it I would have paid a fine, and that's the entire reason. That isn't a gap, it's an obligation. An obligation is a goal that comes from outside, that's how Paul Martin put it, and I've never read it put better anywhere. The rest are hobbies, and hobbies are good. The sentence is a sorting tool, not a verdict.

The other direction is the expensive one. A goal with no project under it is a dream. A dream costs nothing as long as it's labelled as a dream, and then it belongs on Someday/Maybe, where the weekly review reliably puts it in front of you again. On the goal list it does damage: it takes up space, it produces a guilty conscience every quarter, and it dilutes the list until the real goals disappear in it too.

The rule behind it is unspectacular. A goal isn't made of next actions, it's made of projects. If you can't get a single project underneath a goal, then either it isn't a goal, or you haven't yet decided to really pursue it.

When no project will form at all

Then it's usually down to how the goal is worded. Three formats help with that, and they help with different problems.

Most people know SMART from work. The core is good: a goal is worded usefully when two people can tell independently of each other whether it has been reached. The weakness is well known and real. SMART rewards whatever is measurable and tempts you to cut goals so small that a tick is guaranteed. I still like using it, because it delivers the traceability this level otherwise doesn't have.

WOOP comes from psychology, from Gabriele Oettingen, and takes the opposite route to positive thinking. Wish, Outcome, Obstacle, Plan. The active ingredient sits at the back: the obstacle is explicitly the inner one, not the circumstance, and you put an if-then plan on top of it. If I get cravings in the evening, then I drink a large glass of water and eat the bowl of fruit I prepared. In the moment of temptation your head no longer has to decide anything.

One objection to that, and it isn't mine: in the standard example (wish: ten kilos lighter, outcome: feeling awake in the morning) a few steps are missing. Between weight and morning energy sits sleep, and whether that improves depends more on what's on your plate in the evening than on the number on the scales. The point is a methodical one: the outcome has to be the state you actually want, not a chain of cause and effect you have no evidence for.

HARD, from Mark Murphy, starts somewhere else entirely. The assumption is that goals don't fail because of poor wording, they fail because of a missing emotional bond. Heartfelt, Animated, Required, Difficult: it has to mean something to you, you have to be able to picture it, it has to feel necessary rather than nice, and it has to stretch you, otherwise you'll get bored with it.

SMART when you lack clarity. WOOP when you know exactly what's in your way. HARD when what you're missing isn't the clarity but the energy.

The check where you don't look at a single next action

I look at my goals once a quarter. The difference from the weekly review is the altitude, and it's stricter than it sounds: I don't open a single context list and I don't read a single next action. That's exactly why the run-through doesn't take long.

It runs in two directions. Top down, I go through every goal and ask which projects contribute to it. If I can't think of one, I have two options and both are fine: I word a project now, or I delete the goal. Deleting isn't failure. A goal that no longer pulls a quarter later was either set wrongly or has been overtaken.

Bottom up, I go through the projects and ask of each one whether it still contributes to the goal I filed it under back then. A project should be worded so that I see its desired outcome immediately. If it isn't, that's the first repair, before the goal level.

If your project list is sorted by areas of responsibility, you see both at a glance: which area something is moving in and which one has had nothing for months. That's the real reason for sorting it that way, not tidiness.

For me the check sits on a list as a recurring task, not as an appointment in the calendar. It's allowed to be two weeks overdue and glowing red. I need the right day and a quiet spot for it, and an appointment I move three times gives me neither.

Where it typically goes wrong

Standards are sitting on the goal list. "Stay healthy", "keep a good relationship", "keep up professionally". Nothing there ever gets finished, so nothing ever moves, so the whole list feels dead. The done test settles that in ten seconds per line.
Not a single project underneath. Test it hard: over the next four weeks, is there at least one next action that contributes to this goal through a project? If not, it's a dream, and that's where it belongs.
Other people's goals aren't labelled as other people's. Some goals are on the list because somebody else wanted them there. That's fine, as long as you know it. What isn't fine is the state in between: declared, not wanted, a guilty conscience every quarter. That isn't a question of discipline, it's a question of labelling.
The goal level rides along in the weekly review. Weekly is too often for this altitude, and in a run-through with a hundred next actions it goes under. It needs a run-through of its own, and a rarer one.
The list never gets shorter. Goals may be trimmed, changed and deleted. If you never delete, you collect ballast at exactly the point where clarity would have the biggest leverage.

Common questions

How do GTD goals differ from areas of responsibility?

Ask yourself what gets finished. A goal has an outcome that can happen, an area of responsibility is a standard you hold permanently and where nothing is ever completed. "I want to be fit" is an area of responsibility, "by the end of next year I'll run a half marathon" is a goal within that area.

How long is a goal allowed to take?

The original definition in GTD puts it at twelve to twenty-four months. For my own goals that's too short, I think in three to five years there, and I consider the time horizon the weaker criterion anyway. The structure is more reliable: what you reach with a few actions is a project, what you need several projects for is a goal.

Does every project need a goal?

No. Most projects contribute to no goal, they're obligation or hobby, and both are legitimate. The only thing that matters is that you know which of the two it is, because a project you wrongly believe is bringing you closer to a goal distorts your sense of how far along you actually are.

What do I do with a goal I might never reach?

First check whether it's in your hands at all. What you can't influence is not a goal but a wish, and that belongs on the vision level or on Someday/Maybe. If it is in your hands, the honest question is whether you have so far done everything in your power, and the project list underneath answers that more reliably than your gut does.

How often should I look at my goals?

Once a quarter is enough, as long as you really limit the run-through to goals and projects and don't look at next actions. You look at next actions daily, projects weekly, goals quarterly. Every level needs a different frequency, because every level changes at a different speed.

Take your goal list and go through it with a single question: which project in my system contributes to this goal? For every goal where you can't think of one, decide one of two things right now, word a project or delete the goal. Both are a result. Leaving it standing is not.

Setting up the goal level is little work, and the check takes half an hour a quarter. It almost always jams one floor further down: wording and sorting projects so that a check becomes possible at all. That's exactly what the Workflow Setup is for. Write to me if your goal list has been sitting there unchanged for months and you aren't sure whether that's down to the goals or to what's missing underneath.

"I don't really feel like it" is not a no

Fri, 14 Aug 2026 00:00:00 GMT

At work I have been able to say no for years. My boss asks whether I can still take that on by Monday, and I say what will be left lying around if I do. In private I am a specialist in the opposite. Someone suggests something, and I say: I don't really feel like it. The second time: no, I genuinely don't enjoy that. The third time I say, all right, let me see whether we can make it work somehow. That is not friendliness. That is a no I broke myself in the first sentence.

Last week I worked with a communication trainer who has spent years practising difficult conversations with managers. Her point landed with me: people rarely react to the word no. They react to what they suspect sits behind it. And if you offer them a loophole, they take it. Don't be surprised when the other person turns your no into a yes, she says. You handed them the invitation.

The most common hedges

Hedges are words that soften the core of a statement without you noticing. You mean them as courtesy; they arrive as negotiating room. The tricky part is the asymmetry: you remember your sentence, the other person remembers the word that suited them.

What you say	What arrives	What actually carries
I don't really have time	So there is an exception	I have no capacity for that this week
I'll try	He has agreed	I won't get that done by Friday
I'll do my best	He has agreed and he is motivated	I'll take the analysis, the proposal stays where it is
Maybe I can manage it	He didn't say no	No, I'm not taking that on
I'll see what I can do	He is on it	I'll come back to you Thursday with an answer

The most expensive sentence in that table is "I'll try". It sounds like a commitment and is not one. Whoever hears it plans around you. Whoever says it has committed to nothing. Two people leave the same conversation with two different pictures, and when it goes wrong, formally nobody lied.

The conditional is not politeness

"I would suggest." "I'd like to invite you." "Could you maybe take a look at some point." In our speech culture that counts as considerate. She disagrees: the conditional is not a form of politeness, it is a form of indecision. "I'd like to invite you" becomes "I'm inviting you", and the sentence loses none of its warmth.

There is one exception. The conditional is right when it describes a real condition rather than your uncertainty. "If we push the proposal to next week, the analysis is doable by Friday" is a clean sentence. It names a trade. Nothing is watered down.

Hedging is rarely a courage problem

Here I see it differently. Her diagnosis is that many people have too little commitment and too little courage for a clear no. In coaching I rarely saw it that way. What is missing is usually not the courage, but the cover.

Look at the replacement sentence she uses instead of a bare no: "I can't take that on this week, because I have already committed to three other things." The sentence only works if you can name those three other things on the spot. If you can't, you lose the conversation at the first follow-up question. That is exactly where "actually" comes from: it is the linguistic form of "I didn't have the overview just now".

That is why saying no is easy for me at work and hard at home. At work I have a project list, context lists and a calendar that holds only what genuinely has to happen on that day. In private, a lot of it runs past all of that. Cut the hedges without the overview behind them and you don't get a better no, only a harder one that holds just as poorly.

In practice that means: before you answer, thirty seconds on the project list. Then you don't say "I have a lot on", a sentence nobody can hear any more, you name two commitments and a date. That is checkable, and checkability is the difference between a boundary and an excuse. How to defend your own day against exactly these requests is in your plan for the day survives until quarter past ten.

Separate fact from interpretation

"You keep giving me extra work" is an invitation to disagree. The other person will defend themselves, and from then on the two of you are talking about your relationship instead of the work. Priorities are never decided on the relationship level.

The fact-based version: "Over the past week I received three additional tasks on top of what was planned. How do you want to handle it? What can I take, what do we move?" That cannot be disputed, because nothing is being interpreted. Complaint mode ("more again, how am I supposed to manage all this") reliably produces resistance, even when the complaint is justified.

When your no is not accepted

The sentence that follows is almost always: but it still has to work somehow. At that point don't cave, and don't get harder either — ask. What makes Monday so important? The nicest example of this comes from her own practice: the answer was "because I'm on holiday from Tuesday". A piece of information you did not have before, and it changes the entire conversation. Suddenly it is no longer about everything by Monday, but about the part he needs before he leaves.

A no does not need a counter-offer. It needs a continuation. Whoever says "let's talk about the order of things" has created an open loop, and open loops without a date die. That sentence belongs on your waiting-for list or in your day as a next action while you are still in the conversation, otherwise your no is a yes again in three days, only this time without the negotiation.

The rescuer reflex

"Could you take the difficult client call for me, you're so much better at those." You feel flattered, you say yes, and the work has been rerouted. She suggests three steps: show understanding ("I understand that the conversation feels unpleasant"), set the boundary ("I won't take it over for you"), offer development ("sit in on one of my calls, and afterwards we'll talk about it if you're interested in my view").

That offer has a useful side effect: it separates two requests that sound the same. Whoever genuinely wants support takes it. Whoever wanted to offload suddenly has no time to learn. A quick yes would have prevented that distinction and taught the colleague that it works again next time.

Where this typically fails

You cut the words without having the list. The no sounds firm until someone asks what is actually on your plate. Without nameable commitments it collapses at exactly that point.

You say no too quickly and don't ask. Clarity without curiosity is defensiveness. Asking why costs two sentences and often delivers the information that dissolves the problem.

You transfer spoken sentences one to one into writing. In an email the tone is missing that keeps your no friendly in person. One line of appreciation, one sentence of fact, one question. No justification spiral that leaves the recipient looking for leverage.

You negotiate and write nothing down. "We'll talk about prioritisation again" with no date and no entry has vanished into the universe.

You teach the other person that pushing works. Whoever gets a yes on the fourth attempt has learned that asking four times is enough. That is no longer a communication problem, it is a pattern you installed yourself.

Common questions

What is hedging language?

Hedges are words that weaken a clear statement without changing its content. Typical examples are "actually", "maybe", "I'll try" and "I'll do my best". They are meant as courtesy, but they arrive as negotiating room and reliably turn a no into a maybe.

How do I say no without sounding rude?

In three movements: acknowledge the other person's request, state your position without hedges, keep the conversation open. "I understand this is pressing. I have no capacity for it this week, because the analysis and the proposal are already committed. If one of those can wait, let's talk about the order." The friendliness sits in the acknowledgement and the offer, not in the softness of the wording.

What do I do when my no is not accepted?

Don't repeat it and don't cave — ask what makes the deadline so important for the other person. The answer usually contains information you did not have before and turns a demand into a solvable problem. You are not giving up your boundary, you are only moving the conversation from the commitment to the priorities.

Isn't the conditional simply polite?

No, in most cases it signals indecision rather than respect. "I'd like to invite you" and "I'm inviting you" are equally polite; only the second one is a decision. The conditional is justified when it expresses a real condition, for instance which trade would make a commitment possible.

How do I turn down a colleague who wants to hand me their task?

Say clearly that you are not taking the task on, and offer support with doing it themselves instead — sitting in on a conversation, and a debrief afterwards. That combination separates the two requests that sound alike: whoever really wants to get better takes the offer, whoever only wanted to hand it off has no time for it.

An observation task for the next seven days: open your sent folder and search for "actually", "maybe", "I'll try" and "I'll see". In Gmail that is in:sent actually, in Outlook the search box in "Sent Items". Go through the hits from the last four weeks and mark the ones where you really meant no. At the end you will know whether you have a wording problem or a list problem.

The words are one thing, the overview behind them another. If you want to go through both against your real commitments, sparring is the place for it. Write to me.

I don't get up in the morning looking forward to the property tax notice

Fri, 07 Aug 2026 00:00:00 GMT

The electricity bill and the gas bill are sitting on my desk. Both are paid, from the app on my phone, then scanned and filed. All that is left is to throw them away. I don't get up in the morning looking forward to filing the property tax notice. It still has to happen, otherwise the piles get bigger and eventually expensive.

In the last Q&A call I was asked what this "Admin" context actually is that I keep in my lists. Kay filed it under "Beyond GTD", as something that is not in the method as such. Christian first thought of elevated permissions when he heard "Admin". Paul Martin suggested "back office" as a term. The argument about the word is harmless, the thing behind it is not: my context lists have long stopped being named after a place or a tool, and are named after the kind of time a task needs.

Why place and tool no longer sort anything

A context has exactly one job: to exclude. It should make sure that in a given moment I only see the entries I can actually do right now. That is precisely what the classic contexts no longer deliver. I carry my office around with me, in the shape of a laptop. So the split between "office" and "laptop" does not help me, it just produces two lists with the same content.

With the phone it was the same, only earlier. My list stopped being called "phone" a long time ago and became "mobile", because I do far more with the device than make calls. Today it is called "communication". On it is everything that involves a second person: make an appointment with the GP, sort out after-school care, call someone and agree a date. At work a Teams message is exactly the same thing. For me communication is a block, not a device.

What is left, once place and tool exclude nothing, is the state I am in. Do I have half an hour of remaining strength between two meetings, or do I have an uninterrupted hour of quiet? That is what I sort by.

Three lists, three block sizes

List	What is on it	Time block	Size (roughly, in my case)
Admin	Small stuff that has to happen: scan an invoice, raise a standing order, file a notice	30 minutes, daily	around 18 entries
Focus	Bigger chunks that will definitely get done and need room	around 1 hour, three times a week	a handful
Creative	Bigger chunks I am genuinely looking forward to	around 1 hour, twice a week	a handful
Communication	anything that involves someone else	no block of its own	varies

Admin is the stuff nobody feels like doing and that still has to move forward a little every day. If I still had an assistant, I would hand a lot of it over. I don't, so it gets a fixed half-hour slot. In half an hour you get surprisingly far with small stuff.

Focus and Creative are both "big"; the difference is motivation. Focus has to happen and only needs the room for it. Creative is the things I look forward to: working a natural planning model through for a private project, defining a method properly. That is exactly why they must not simply slip in between, otherwise I eat the nice things first and the property tax notice stays where it is.

The calendar block is the mechanism, not the list

Tasks do not belong in the calendar. An admin block does not break that rule, because it is not a task. It is an empty container with a time on it. What I put into it I decide when I open it, not when I create it. An appointment for a specific task is lost when it falls through. A container is only capacity, and capacity can be moved.

Two properties make the block work. Its length: I don't want to do admin work for two hours straight, I would never start. Thirty minutes I will start. And its bindingness: move it if you must, cancel it never. If you don't have the time to get things done, your lists will not get shorter, no matter how cleanly they are sorted.

Where this construction does not hold

I think the setup is good, but two things about it are wrong, and both only became clear to me during the call.

First, it is not a sort by energy, it is a sort by the length of block required. Focus and Creative demand the same concentration and the same hour; they differ only in whether I feel like it. It would be more honest to have a half-hour list and an hour list plus a marker for appetite. Naming it that way gets you to the real question faster: how much contiguous time does this entry need, and do I have it today?

Second, during the call I said that everything on the focus list will definitely get done and in the foreseeable future. Hendrik then asked where the difference to Someday/Maybe lies. My answer was: bungee jumping is on Someday/Maybe and may be deleted at any time, the documents for a court hearing are on Focus, with deadlines and money attached. That is a clean answer to the wrong question. Because "I am definitely doing this" is not a context, it is a commitment decision. What I have done there is build a second axis into the context layer. It works for me because I carry both axes in my head. Anyone adopting the structure sorts by feel, and the focus list fills up with everything that sounds important.

And one more thing needs putting straight: energy is not an extension of the method. It is one of the criteria GTD uses to decide what you do in a given moment, on equal footing with context, time available and priority. The only new thing about my solution is that I take that criterion out of the head decision and build it into the structure of the lists. That is not a deviation, it is a materialisation.

Where this typically fails

The lists get renamed, the blocks never get entered. Then only the label has changed. The structure carries nothing, the calendar carries.
The admin block is set too long. Nobody sustains two hours of small stuff. The block becomes the place where you hide from the real work behind trivia.
The focus list holds sub-projects instead of next actions. "Get a valuation for the old estate car" was the disputed case in the call. For one participant a clear next action, because he drives to the dealer he trusts. For me a sub-project, because I would first have to research what that even means. An hour-long block is worth nothing if the entry cannot be finished within that hour.
The block gets cancelled instead of moved. An appointment with yourself that has been cancelled once is cancelled more easily the second time.
Focus becomes the collection point for everything important. When it holds thirty entries at an hour each, the list is a calculation that does not add up. Then two honest routes remain: delete, or tell whoever is waiting that it is not happening. Both are unpleasant, both are better than a list that lies.

Common questions

What are GTD contexts by energy?

They are context lists named not after a place or a tool, but after the state and the amount of time a task demands. Typical names are Admin for small-scale work, Focus for concentrated work and Creative for tasks that need appetite and room. The purpose is the same as with classic contexts: in a concrete moment, to see only what is actually doable now.

How many context lists should I keep?

Enough that each list can be taken in fully in a real moment, and few enough that no two lists hold the same content. If two lists are always relevant together, they belong merged. If a list gets so long that you stop reading it to the end, it is no longer a selection aid, it is an archive.

Is an admin block the same as time blocking?

No, and the difference matters in practice. In classic time blocking a specific task gets a specific time, and if the slot falls through, the task is undone and the calendar entry worthless. An admin block only reserves capacity; the selection happens when the block opens. That is why it survives disruption without anything having to be rescheduled.

How do I separate the focus list from Someday/Maybe?

By commitment, not by task size. Someday/Maybe holds what could be deleted without anyone noticing, and what comes up for reconsideration in the weekly review. The focus list holds what is tied to a promise, a deadline or a consequence. If you cannot decide where an entry belongs, it is the commitment behind it that is unclear, not the list.

What do I do if I have no time for blocks like these at all?

Then that is the result of the reflection, not an argument against the structure. Someone who sits in meetings from nine to five every day has no time to get things done, and no ordering of lists will change that. The next action in that case is a conversation with your manager or your team, not a new way of sorting.

Take your longest context list and, for one week, write just one number after every entry: the minutes it needs in one go. After seven days you will see whether your problem is your list or your calendar.

A yoga teacher asks about neck pain. More than half the hands go up.

Mon, 03 Aug 2026 00:00:00 GMT

Last week our regular Friday call had, for once, neither a tool nor a list at its centre, but a guest: Ninni, yoga teacher, nurse and sound therapist. Her title for the session was the kind of extension you have to think of first: Mind like Water, Body like Bamboo.

Before she showed the first exercise, she put two questions to the group. The first: what comes to mind when you think of bamboo? The chat delivered resilience, stable but flexible, light but tough, roots without end (and pandas). The second: who has tension or pain in the neck and shoulders? More than half the group raised a hand. All people who think, plan and decide for a living. And for most of them the neck tightens while they do it.

Why this is a method topic

The water metaphor in GTD describes a mind that responds to every impulse appropriately: not too much, not too little, and then quiet again. What gets lost in the purely methodical discussion is that this mind runs on hardware. Sit at the machine with a headache, a locked-up neck or shallow breathing, and you may have the best system in the world, but no focus.

Screen work is structurally one-sided. Arms forward, shoulders rolled in, gaze down, for hours. On top of that comes a detail Ninni, as a nurse and yoga teacher, explained cleanly: when you work under concentration or stress, you breathe almost entirely into the chest, and pure chest breathing is read by the body as an alarm signal. Tension rises further, the loop reinforces itself. Footballers breathe that way on purpose before a penalty, someone offered on the call: maximum activation, maximum adrenaline. A working day is not a penalty kick, and anyone who spends it entirely in that mode is correspondingly finished by the evening.

The methodical anchor is the Engage step. You choose the next action by context, time available, energy and priority. For context and time we have lists and the calendar. Energy most people treat like weather, something you accept. Yet it is the only one of the four you can influence directly in two minutes.

Three pillars, two of which fit at the desk

Ninni sorts physical health work into three pillars: strength, stretching, mobility. Strength belongs in training, not between two video calls. The other two fit at the workplace. Mobilisation keeps joints moving and improves circulation, stretching releases tension that is already there. The order matters: mobilise first, then stretch.

If you would rather be guided than read: Ninni has guided meditations on YouTube.

Her most important rule: gentle and long rather than hard and short. Forcing yourself into a stretch creates new stress in the tissue instead of releasing it. She holds stretches for many breaths at low intensity, well below the pain threshold.

Desk yoga concretely: one breathing technique and four exercises

The entry point is full yogic breathing, and it comes with a self-test built in. Left hand on the chest, right hand on the belly. First a few breaths into the belly only: on the inhale the belly wall pushes outward, on the exhale it draws gently back in. Then a few breaths into the chest only, against the upper hand. Then both as a wave: inhale low, carry the breath upward, exhale at the top, finish emptying at the bottom.

The call showed immediately that this is not self-evident. Hendrik said openly that he simply could not separate belly and chest breathing. Ninni's answer: entirely normal, most of us never learned to breathe differently, which is exactly why the hands are there. Nicki added the practitioner's trick: a slight hollow back pushes the breath almost automatically into the chest, which is how you feel the difference in the first place.

The four exercises that followed cover the regions where screen workers typically collect tension:

Exercise	Region	How	Dose
Half moon roll	Neck	Ear to the right shoulder, then a slow half circle across the front to the left side and back. Shoulders stay down, the movement comes from the cervical spine.	5 to 10 slow passes
Neck stretch	Side of the neck	Draw the ear towards the shoulder and hold. Turning the chin towards the shoulder as well shifts the stretch to a different strand of muscle.	5 to 10 breaths per side
Shoulder opener	Shoulders, chest	Interlace the fingers behind the back, keep the palms together as far as they go, draw the hands towards the floor, open the chest.	5 to 10 breaths
Wrist circles	Wrists, forearms	Close the fists with the thumbs inside, circle from the wrist. The forearms stay still, otherwise the elbow cheats.	10 circles per direction

The shoulder opener is the direct counter-movement to the laptop posture: everything the screen pulls forward and inward goes backward and open here. If you like, add the lemon face from Jacobson's progressive muscle relaxation: tense every facial muscle for five seconds as if you were biting into a lemon, then release. It sounds silly, but it noticeably lowers baseline tension, especially in the jaw.

The exercises are not the problem. Remembering is.

The decisive objection came from Ulrich: under tension you pull one thing through after another, and suddenly two hours are gone, two hours spent fully braced. This is exactly where the yoga topic becomes a GTD topic. Remembering your own intentions is precisely the work we otherwise hand to the system without hesitation. For the body most people make an exception, and then wonder why it comes away empty.

Two mechanisms work. First, the recurring task: mine would read "desk yoga, 5 minutes", would sit in the system every day and stay visible until it is ticked off. No popup, no alarm, just an entry that looks back at me every time I open the day's list. Second, and probably more effective: couple the exercise to a transition that happens anyway. End of a work block, last call before lunch, completion of a task. The head needs the break at those points regardless, and the body collects it at the same time.

What I would warn against is turning this into a project. "Become more flexible", with twenty collected exercises, three subscribed channels and a new mat, is the same mechanism as the perfect board that never completes a single task. Ninni herself set the bar low: if it is the same neck exercise every day for two weeks, because the neck is the current issue, a lot has been achieved. She drew the comparison without prompting: the weekly review does not work because it is big, but because it reliably takes place.

Where it typically fails

The reminder lives in your head. "I'll do that here and there from now on" is an intention, not a component of a system. Without a recurring task or a fixed trigger, the next email always wins.
One exercise turns into a collecting project. First collect exercises, then save videos, then buy equipment. It feels like progress and replaces the ninety seconds it was actually about.
Stretched too hard. The body answers pain with counter-tension, and after three days the exercise is dropped. Gentle and long works, hard and short injures.
All or nothing. People who cannot find a quarter of an hour do nothing at all. The version for bad days is a single exercise, and it counts.
"Here and there" is not a point in time. Without a defined trigger (end of a block, last morning call, a finished task), here and there never happens.

Common questions

What does desk yoga do for concentration?

Desk yoga lowers the physical tension that eats concentration: a tight neck, raised shoulders, shallow breathing. Short mobilisations and gentle stretches improve mobility and circulation without you leaving your workplace. The effect replaces neither breaks nor sleep, but it is felt immediately and costs only a few minutes.

How do I tell whether I am breathing into the belly or the chest?

Put your left hand on your chest and your right on your belly and keep breathing normally. The hand that rises more clearly shows your pattern. During concentrated screen work, most people's breathing migrates upward into the chest. The same hand position also trains full yogic breathing: inhale against the lower hand first, then against the upper one.

What is the difference between mobilisation and stretching?

Mobilisation moves a joint through its available range, for instance as a slow half circle with the head, and improves mobility and circulation. Stretching holds a position across several breaths and releases tension in muscles and fascia. The combination is what works: mobilise first, then stretch gently.

How often do I have to do the exercises for them to help?

A single exercise every day beats the ambitious programme that dies after four days. The effective variable is consistency, not volume. If your neck is the issue, it may be the same neck exercise every day for two weeks.

Do I need a yoga class to start with this?

No, simple mobilisations, stretches and breathing exercises work on an office chair without a mat and without prior experience. If you want to go deeper, calm styles such as Yin Yoga or Iyengar Yoga are a good fit, both work with props and suit beginners. One fixed session a week that reliably happens is worth more than occasional intensive phases.

Pick a single exercise from the table and hang it on a transition that occurs in your working day anyway. Two weeks, every day, only that one. And today, three times in the middle of work, briefly put your hands on your chest and belly and notice where you are breathing. Do not change it, just observe. What you find there tells you more about your working day than any review of your lists.

The night her head spent five hours clearing itself out

Tue, 28 Jul 2026 00:00:00 GMT

A participant in our sessions recently showed us her system. Two children, now seven and nine, self-employed, and she started at the end of 2023.

The trigger was not an interest in productivity. She describes the state before it like this: running after things all day, sorting out the world outside, functioning. Then in the evening the realisation that she also has a body, and a headache she had simply pushed aside for hours.

When she started capturing, something happened that I rarely hear put so plainly. She cut up scrap paper and put it everywhere — next to the bed, in the bathroom, in the kitchen. And then her head began to clear itself out. Lie down, a thought arrives, sit up, write it down, lie down. The next thought. For five hours.

The next morning she took the children to kindergarten, came home, and there was quiet. Her sentence for it: I had the feeling I could choose again what I do, and that I was no longer being driven.

Why it is different with children

For parents it is not only the volume that explodes. The nature of the commitments changes.

A large part of what has to be done is not yours at all. The green pencil for the pencil case, the broken ruler, the present for Saturday's birthday party. These are other people's needs turning into your tasks, and nobody enters them anywhere. They get mentioned in passing.

On top of that, these things are individually laughable in size. Which is exactly why they do not land in a system but in your head. And your head does not distinguish between big and small. Every open loop costs the same, whether it is a product launch or a pencil.

That is why exhaustion in parents often does not correlate with the amount of work. It is the forty small things nobody wrote down.

One system for everything

She does not separate private from professional. Four boards in Trello, and both sit side by side there: the travel idea, the book idea, the client call, the school matter. Whatever is not current sits on someday/maybe and does not clutter the active lists.

This is the point where GTD differs from most methods, and with parents it pays off most. Run two separate systems and you have to decide for every single thing where it belongs — and you make that decision on a day when both are happening at once anyway. Her phrasing: everything has a place. That was what convinced her.

What needs separating is the view, not the filing. You do not want to see the shopping list while you work. But that is a question of contexts and filters, not of two systems.

The board on the wall

The part that occupied the group most: at her home a kanban board covers an entire wall, a good two metres wide, magnetic and writable.

The columns are the usual ones: inbox, to do, waiting for, done. And every family member has their own row, a swimlane.

On it are sticky notes with exactly the things from above. New green pencil. Replace ruler. Present for the birthday party.

Why analogue, when she has everything in Trello anyway? Because it has to be present in family life. On a phone it is gone the moment you lock the screen, and nobody wants to look at a phone over dinner to find out what is due. You cannot fold a wall shut.

The second effect matters more and gets overlooked: a visible board makes the volume visible. When one person carries it all in their head, nobody else sees how much it is. When it hangs on the wall, everyone sees it — including the people who are otherwise surprised that somebody is spent by the evening.

Your partner does not have to join in

Her husband does not do GTD. He uses the family board, and that is that.

Asked how she had convinced him, she gave the most relaxed answer of the evening: not at all. He sees the board, we use it together, and if he wants to know more I am happy to tell him. She has no interest in converting anyone.

I consider this the most underrated point in the whole construction. A family system does not work because everyone knows the method. It works because there is a shared place where it says what is due. Anyone trying to teach their household the method before anything improves will be waiting a long time.

Two context lists you only get with a family

Alongside the usual contexts, her next-action lists carry two you rarely see in office setups: discuss with husband, and discuss with the children.

These are agenda lists, and in private life they are at least as effective as at work. Everything that does not warrant its own occasion but should be raised at the next shared moment lands there. You stop trailing your partner with seven small things across the day, and you stop leaving a shared evening thinking you had actually meant to ask something.

With children a second benefit appears. When you know you have four topics to raise with your child, you also notice when three of them are reprimands.

Chore cards the children choose from

For the household she built cards with pictures, made with an AI image generator. Each card is one task: clean the sink, tidy the children's room, empty the letterbox, take out the rubbish.

Saturday morning she goes through what is currently due. Some drop out because they do not work that day — plastic waste not on a Sunday, for instance. Then she lays the cards on the table and each child picks two or three.

The difference between assigning and choosing is the entire trick. A child who chose the sink cleans the sink. A child assigned the sink negotiates. And there are tasks children are surprisingly happy to do if you let them. Cleaning the toilet is not among them; that stays with her.

Two to three hours on Saturday, everyone at once, and then it is done. Her observation: together it is considerably more fun than spread out alone across the week.

One detail that shows the system really runs: laundry into the machine is a card. And afterwards she is waiting for the machine to finish, so the thing moves to waiting-for. That sounds excessive until the one time you notice at ten in the evening that the washing has been sitting wet in the drum for six hours.

Anyone can be busy

She experimented with a today list and stopped again. The reason is instructive: she never knew whether to move a card there or enter it anew, things were left standing, and then came the question of whether to move them back or simply recreate them tomorrow. The upkeep outweighed the benefit.

What works instead are markers for the things that are genuinely due. And she puts the insight behind it like this: one or two success-critical tasks, which then actually get done. Anyone can be busy. You can spend the whole day running after things, feel like you are doing something, and not have worked on anything that mattered.

For a day with children, appointments and self-employment, one to two is the honest number. Anything above that is a list that proves in the evening that you failed, even though you worked all day.

Waiting for, sorted by people

A small detail with a large effect: her waiting-for list is sorted by groups of people. Clients, family, school, network, plus open invoices and ordered parcels.

The advantage shows when going through it. You do not check thirty individual items; you go group by group and see immediately where something is jamming. And if you are on the phone with someone anyway, you have everything open with that person in one place.

A technical tip from the group for anyone using Trello: set up an automation that sets the start date to today whenever a new card is created on the waiting-for board. Then every entry says how long you have been waiting, without you typing it. And how long is the only information that turns a waiting-for into an action.

Where it typically fails

The board hangs there, but nobody keeps it up. A wall board needs the same weekly pass as a digital one. Otherwise the Easter cards are still hanging there in May.
The partner becomes a project. Try to re-educate your household before your own system works and you will give up on both. First the shared place, then perhaps the method.
Everything gets done on paper. The wall board works for family matters because they are short-lived and should be visible. Projects, reference material and anything professional do not belong there.
The small things still land in your head. If the pencil for the pencil case is not written down somewhere within ten seconds, it is written down nowhere. Capture has to be possible everywhere, which is why the notepaper sits in every room.

Common questions

How do you organise family tasks sensibly?

With a shared visible place instead of inside one person's head. A kanban board on the wall with columns for inbox, open tasks, waiting for and done, plus one row per family member. The decisive effect is that the volume becomes visible to everyone.

Should I keep private and work tasks separate?

In capture no, in the view yes. A single system means every item has a place immediately. Separate views through contexts and filters make sure you do not see the shopping list while you work.

How do I get children involved in housework?

Choosing instead of assigning. Write or draw tasks on cards, select the ones that are currently due, and let the children take two or three from those. One shared fixed appointment works better than individual instructions spread across the week.

What if my partner does not want to join in?

Force nothing. A shared visible board works even when only one person knows the method behind it. Participation comes from the shared place, not from being convinced of the method.

Why do small tasks weigh so heavily?

Because your head does not distinguish between big and small. Every unfinished thing that is not reliably noted anywhere consumes attention, regardless of its size. For parents, many of them accumulate every day.

If you live with children and feel constantly drained without being able to name what from: for one week, write down every small thing somebody mentions to you in passing. The length of that list is usually the answer.

A system that carries work and family at the same time is something you build properly once and then never again. That is exactly what a workflow setup is — with your real data. Write to me if everything is currently sitting in your head.

The tandem jump sat on my list for a year and a half

Mon, 20 Jul 2026 00:00:00 GMT

During a break in a training, someone told me I absolutely had to do a tandem paragliding jump, that it was the best thing ever. I found the idea middling. I wrote it down anyway, and it landed on my someday/maybe list.

It sat there for a year and a half. I saw it every month, thought about it briefly every time, and every time left it standing. Until the month the thought came: either I delete this now, or I do it.

I made a project out of it. It was called "complete a tandem jump without booking a course", because I know myself. Five minutes after the first step I was on a flight school's website looking at the advanced course for cross-border flights, about to book it. I just about managed to turn around. On the first attempt I then waited four hours for wind and drove home empty-handed; two weeks later it worked.

This is not a story about parachutes. It is one about a list doing exactly what it was built for.

What the list exists for

It keeps capturing cheap.

If every idea you wrote down were a commitment, you would eventually stop writing ideas down. And that is precisely what happens. I regularly heard sentences in coaching like: I inwardly refuse to write down any more, because then I have to process and push all of it forward too.

That is economising at the wrong end. Capturing has to remain the simplest action in the whole system, because everything else builds on it. Stop collecting and you no longer have a complete picture, and your head notices immediately and takes over again itself.

The someday/maybe list is the place where an idea is allowed to exist without you committing to it. That is its entire purpose, and it is a bigger one than it sounds.

What separates it from an idea graveyard

The term came from one of our sessions, a participant used it, and it describes the risk exactly: a list where things land and are never looked at again. If that is the case, you might as well print it out and throw it away. The effect is the same.

The difference between a working list and a graveyard consists of exactly one thing: an appointment.

I go through mine once a month. Not thoroughly, I fly over it. For each item there are three options: delete, leave standing, activate. For me that takes a few minutes. Where this pass sits in the larger process is in the weekly review.

Some people deliberately put creative time in the calendar for it, which I think is good, because then the pass does not look like system maintenance but like what it is: the only occasion on which you occupy yourself with things nobody is asking of you.

The guilty conscience about leaving things standing

This part took me a long time to learn. If I fly over the list and leave an item untouched, I do not need a guilty conscience about it.

For a long time I had one. These are unfinished things, I thought, and every time I read past them there was that small tug. This is exactly the point at which most people give the list up, because a monthly appointment with a guilty conscience does not become a habit.

The mistake is in the category. An item on someday/maybe is not an unfinished task. It is an option. And you are allowed to decline options month after month without it becoming a failure. With the tandem jump it was eighteen times.

What belongs on the list

Three different kinds, which often get mixed up.

Genuine someday ideas. The language course, the trip, the book project, the tandem jump. Things you might want, but not now.

Dormant projects. Undertakings that have begun or are pending, but where nothing has moved for weeks and nothing will move in the next four or five either. They do not belong on the active project list, where they trigger an individual review with a guilty conscience every week. With forty active projects that is typically ten to fifteen of them.

Everything where nothing happens if you forget it. The criterion comes from a participant and is the most usable one I know. If you leave the house tomorrow and forget the thing, nothing happens. It should still be written down somewhere. That is exactly what the list is for.

The backlog as an intermediate stage

Many people use someday/maybe exclusively for private ideas, holiday dreams and things that are fun. If you then put a dormant company project on the same list, it feels wrong, and the list loses its character.

The solution some have built: a third list between active projects and someday/maybe. From programmer jargon, the name backlog has established itself for it. That is where projects sit that are real and will run at some point, but are currently dormant. Not dreams, just paused.

Whether you need it depends on how your someday/maybe list feels. If you hesitate to put a project on it, you need one. Which lists otherwise belong in the set is in GTD lists: the ones you actually need.

Secure the next step when you move a project

A practical detail that often gets overlooked and saves time on reactivation.

When a project leaves the active area, its next step gets deleted from the context list. It should not show up there any more, otherwise the move has no effect.

But do not delete it without replacement. Write it into the project's support material, into the note, into the Trello card, wherever the project keeps its filing. Then later it says there: this was the state, this is what I wanted to do next.

When you reactivate the project in eight months you will want to think it through afresh anyway. But you start from a point where somebody has already done some thinking — namely you.

Deliberate deletion is not failure

On the last pass I deleted two items. They had been sitting there for three weeks, they had mattered to me at the start, and looking at them it became clear: I did not get to them because other things were more important, and that will stay that way for the next five weeks. So, gone.

The decisive part of this is not the deletion. It is that the things sat there for three weeks and I had the chance every time to decide whether I wanted them. That is the difference between a conscious decline and forgetting.

Forgotten things come back, mostly at night. Declined ones do not.

The list as permission

A benefit I had not had on my radar before, and which came out of the same group.

One participant described how she tends not to do things for herself. The opportunity is there, and then comes the reflex: oh no, better not, too expensive, over the top, I do not need it.

What makes the difference for her is the fact that the thing is on her someday/maybe list. She can then argue with herself: I have thought about this before. It mattered enough to be written down. So I am doing it now.

The list is thereby not just a filing place but a permission you granted yourself at a moment when you were clearer than you are in the moment you hesitate. That is a good reason to keep one, even if not a single item on it ever gets activated.

Where it typically fails

No pass. Without a fixed appointment it is a graveyard, and within a few weeks your head entrusts nothing to it any more.
The list becomes a second inbox. Push everything unclear onto it instead of clarifying it and you have merely renamed the inbox. What goes on the list is what has been decided, namely: not now.
Too many categories. Set up six different maybe-lists and you go through none of them. One, at most two with a backlog alongside.
Everything sits there forever. If you have deleted nothing and activated nothing for a year, the pass has become a formality. Two items per pass where a decision genuinely falls are enough as a sign of life.

Common questions

What belongs on a someday/maybe list?

Ideas you might want to realise at some point, projects that are real but currently dormant, and everything where nothing happens if you forget it but which should still be written down somewhere.

How often should I go through the someday/maybe list?

Once a month is enough, as a skim with three possible decisions per item: delete, leave standing, activate. Without that fixed appointment the list becomes an idea graveyard that nobody entrusts anything to any more.

What is the difference between someday/maybe and a backlog?

Someday/maybe holds options without commitment, often private and idea-like. A backlog contains real but paused projects. The separation is worth it if you hesitate to put a company project on the same list as a holiday idea.

Should I feel guilty if things sit there for years?

No. An item on this list is not an unfinished task but an option you decline regularly. One undertaking sat on mine for a year and a half until the right moment came.

What happens to the next step when a project moves to someday/maybe?

Delete it from the context list, otherwise the move has no effect. But note it in the project's support material first, so that the last state of thinking is available on reactivation.

Open your someday/maybe list and check when you last deleted or activated an item. If you cannot find one, you are not keeping a valve, you are keeping an archive.

If your lists jam in several places at once, it pays to look at the whole rather than at one list. That is sparring. Write to me if you want to sort it out once.

"It took me five hours last week"

Sun, 19 Jul 2026 00:00:00 GMT

The sentence came up in one of our live sessions, where every week I take a volunteer through their weekly review. Not as a complaint, more as a confession, in the tone of someone who genuinely wanted to do it properly.

My answer at the time: that can even be fine, if you reserve Friday for project work anyway. You just must not tell anyone. Anyone who hears that a weekly review takes five hours will never do one again.

The reason for the five hours was found quickly, and it is the same one as for almost everyone whose review runs out of control.

Reviewing and planning are two different activities

While going through his projects, the participant was planning at the same time. A client meeting had left signed documents and notes, out of which came a good fifteen next steps, and those had to be structured and put in order. That is real work, it is necessary, and it has no business being in the review.

Reviewing means checking whether the system still reflects reality. That takes seconds per project. Planning means thinking a undertaking through, and that takes half an hour. Do both in the same appointment and you will not get through.

I would also prefer to pull the whole thing through immediately whenever I set up a project. You have to be careful there. What I do instead, and quite often: during the review I write down a single next step, "plan project XY" or "run the natural planning model for XY". I know exactly what that means for me — set up notes and think the thing through properly once. And I take separate time for that.

The walkthrough, step by step

First of all: there is a common template for the weekly review, and I always recommend copying it into your own notebook and making your own version out of it. Mine is not the perfect version. It is the one that works for me.

Arriving

Distractions away, headphones on, music on. Sounds like a side issue and is a fixed part of it for me.

One participant told us about his creative playlist, purely instrumental, which he uses exclusively for the review so that it does not wear out. It gets him into flow considerably faster. I went through a phase of always listening to the favourite playlist of whoever I was coaching at the time. Two months of Metallica straight through the weekly review. It was hard, but it worked.

Getting clear, physically

Empty all physical inboxes. For me that is a small basket on the desk, for others the briefcase with the plastic sleeves from client meetings.

If there is little material, I process it right away, meaning I decide the next step immediately and enter it. If it would take an hour, it makes more sense to separate collecting from processing and postpone the latter.

Getting clear, digitally

This is where it gets interesting, because this is where most inboxes get overlooked. The question is not "have I got through my mail", but: where can something land that I would have to answer or do?

From practice, mine and participants':

Email, obviously.
WhatsApp, including for business. For many people it is the channel where things actually happen, and at the same time the most frequently neglected one. One participant takes a screenshot of the message and sends it into his capture tool, so the matter lands in the normal inbox.
The spam folder. Something relevant regularly lands in mine. That is why it is fixed on my checklist.
Chat programs. I work with several companies and have three Teams instances and one Slack. Some of them I do not look into at all during the week, because there is no need. Once a week I fly over them.
The notes app and transcripts. If you dictate a memory protocol after meetings, there is material sitting there that wants processing.
The downloads folder. More important than the desktop for me, because I use it the way other people use their desktop.

On that last point I always said the same sentence in trainings, and I meant it: no matter how new, expensive or sexy your computer is, nobody can guarantee it will still exist tomorrow. The desktop and the downloads folder are the worst possible place for something that matters to you. Go over it once a week, fish out what is important and put it where it belongs.

Mind sweep

Paper and pen, or digital, and write down everything currently in your head. Do not evaluate, do not sort, do not distinguish between big and small.

I work with a trigger list and read the categories out loud: errands, hardware store, pharmacy, bank, presents. Mobility, car, bike, repairs. Health, doctor, check-ups. Neighbourhood, friends, relatives, colleagues, partner, upcoming birthdays. Communication, open calls, mail, reports, minutes. Staff, hiring, appraisals, personnel development. For the self-employed: freelancers, outsourcing, everything you ought to think about before you urgently need it.

Anyone with a good capture habit during the week finds only two or three things here. That is exactly the point. The mind sweep is not a capture method, it is a check.

One participant asked what to do when ideas always come in places where he has no device on him. The answer is unspectacular: if you have a place where things regularly occur to you, make sure there is something to write with there. For years I had a waterproof board with a suction cup in the shower. The note did not go into the inbox afterwards but onto my keyboard, because I knew I would be sitting there in ten minutes anyway. The later attempt to digitise this with a voice assistant in the bathroom failed. Arguing with a device from inside the shower made me more aggressive than I normally am, and the humidity did the speaker no long-term good.

Context lists

Go through, tick off, delete, defer. Which lists these should even be is in GTD lists: the ones you actually need. More on this below, because one of the three critical points sits here.

Calendar backwards, then forwards

Unfamiliar for many: I start with the past week. Just skim it, click nothing. What kind of meetings were those, who was in them. Almost every time, something comes up for me that got left behind.

Then the coming two to four weeks. And this is the moment the blockers get set. If you do not defend your time, nobody will do it for you. In companies, colleagues simply book themselves in when they see a gap, and privately it works exactly the same. What else ends up in the calendar despite not belonging there is in your calendar lies about your time.

In the session came the objection that always comes: the dentist only has this one slot. Then that is a decision, and you are making it. But it is yours, not his. He is not going to tell you that you need two hours for reading next week. And if a service provider is never available, there are other service providers.

Projects

The main question when going through them is: does this project have a concrete next step? If yes, on to the next one. That goes fast.

It gets interesting with the others. If no next step is defined, it gets defined now. And if a project has been standing for a long time and still nothing is moving, it is worth asking whether the step is cut too large.

If nothing has moved for weeks and nothing will move in the next four or five either, the project does not have to stay on the active list. It moves to someday/maybe, so that the project list stays a usable tool instead of growing overwhelming. Why the sheer number is not the problem is in too many projects?

Anyone who uses someday/maybe exclusively for private ideas and holiday dreams can set up an intermediate stage. Some call it a backlog: a list for projects that are currently dormant but are not dreams. It works the same way, and for some people it feels more right.

Waiting for

For me, everything I have handed off goes in here, and handing off means two things. Real delegation in the team, meaning someone takes a topic on and I want to know the status regularly. And everything where I am waiting for an answer: sent mail, applications, reimbursements, orders.

One example that shows nicely why this pays off: in February I ordered brake levers for the motorbike. Two weeks later I looked at the order confirmation more closely; it named a calendar week, and that week was in July. It moved to waiting-for, and on the very day of the session the delivery arrived and I could tick it off.

Reimbursements likewise. If I submit a medical invoice and the money arrives some time in the next few weeks, I need a reminder to check the account occasionally. Otherwise nobody notices.

One bonus question at the end

At the end I like to ask one more that has nothing to do with the system. For instance: which of your friends have you gone longest without calling? And then that exact name goes onto the calls list.

The three places where it gets stuck

The context lists do not get emptier

In the session, someone said it openly: he inwardly refuses to write new ideas down. Because everything he writes down then wants to be processed and pushed forward too, and it is already too much as it is.

That is economising at the wrong end. Capturing is the part that has to stay cheap, because everything else builds on it. Stop collecting and you no longer have a complete picture, and your head notices immediately and starts carrying things itself again.

The useful question is: where exactly is the jam? If the inboxes keep getting fuller, there is not enough time for processing. If the context lists keep getting fuller, there is not enough time for doing. Those are two completely different problems with different solutions. The second one is mostly about blocking time, saying no, delegating — and sometimes about admitting that it is too much for one person.

A line from the group that captures it well: full context lists always have a reason. Either we do not want to do the things, or they are simply not relevant. Both are information, and both lead to a decision.

And another one I have kept: the flip side of inbox zero is full context lists. The empty mailbox feels like being finished, when the work has merely moved somewhere else.

The idea graveyard

The term came up in the group, and it describes the danger precisely: a list where ideas land and are never looked at again. In that case you might as well print it out and throw it away.

The difference between a someday/maybe list and a graveyard is an appointment. I go through mine once a month, fly over it, and decide for each item: delete, leave standing, or activate.

What took me a long time to learn: if I fly over it and leave something untouched, I do not need a bad conscience about it. For a long time I had one, because these are unfinished things after all. That is the point at which many people give the list up.

My favourite example shows how well it works: a tandem paragliding jump sat on my list for a year and a half, left standing on eighteen consecutive passes — and then I did it. Without pressure, without a guilty conscience. That is the whole point of it. The full story, and how I keep the list, is in someday maybe: the list that must not become a graveyard.

The client as a project

For someone who runs one project per client, going through the project list automatically becomes long, and that slows the entire review down.

The solution we found during the session, which the participant implemented immediately: separate active and dormant clients. One board for clients where something is genuinely running right now, and a pool for all the others. After ten minutes the feedback came: instead of a long list, there are now five clients on the active project list.

This holds beyond the client case. If your review regularly gets sticky at one point, it is usually not the method's fault but the structure at exactly that point.

How long the whole thing takes

With explanations, questions in between and discussion, the session took a good hour and twenty minutes. The participant's own estimate afterwards: without the interruptions he would have been through in about three quarters of an hour.

That is a realistic order of magnitude for a well-kept system. The first few runs take longer because they are catching up on backlog. If it permanently takes several hours, planning or processing is happening in it instead of checking.

The trick that makes it actually happen

My recommendation has been the same for years: find somebody. A colleague at the company, someone you know who is also starting with GTD, and do the review together at the same time.

Not because you learn something from each other, but because a shared appointment gets kept and a solitary intention does not. That, incidentally, is exactly how our weekly session came about. During the pandemic, sitting together was not possible, so I did my own weekly review online and other people could dial in.

Common questions

How long should a weekly review take?

With a well-kept system, about an hour; with practice, less. The first few runs take considerably longer because they are catching up on backlog. Several hours almost always means that planning or processing is happening inside the review.

What is the difference between reviewing and planning?

Reviewing checks whether the system reflects reality, and takes seconds per project. Planning thinks an undertaking through and needs concentrated time. If the review reveals that planning is needed, "plan project X" goes onto the list as the next step and gets scheduled separately.

Which inboxes are most often forgotten in a weekly review?

WhatsApp, the spam folder, chat programs like Teams or Slack, the notes app, dictated memory protocols and the downloads folder. If you only empty the mailbox, you have caught the smaller part.

What do I do when my context lists keep getting fuller?

First distinguish where the jam is. Full inboxes mean too little time for processing, full context lists too little time for doing. For the second, only blocking time, delegating, saying no or deliberately deleting things helps. Stopping capture is not a solution.

How do I stop my someday/maybe list becoming an idea graveyard?

With a fixed appointment. Go through it once a month and decide for each item: delete, leave standing or activate. Leaving something untouched is entirely fine and needs no bad conscience.

If your next review runs long again, note down at which point it happens. Usually it is two or three projects, and those do not need checking — they need an hour of thinking on a different day.

The fastest way to improve a review is having someone sit beside you and see where it snags. That is what sparring is, one fixed appointment a month. Write to me if yours regularly overruns or gets skipped.

"You are there, but you are not here"

Wed, 01 Jul 2026 00:00:00 GMT

Knut, a fellow trainer and a guest in one of our sessions, heard that sentence from his wife years ago. At the dinner table, his son still small, himself physically present and mentally with unread mail and open project states.

That is the kind of feedback you cannot argue away. And it describes a problem that time management does not solve, because it is not a time problem.

Why your head keeps running in the evening

Your brain does not remind you of unfinished things because it wants to annoy you. It reminds you because it does not trust the place where the thing is supposed to be sitting. If a commitment is not captured cleanly anywhere, or if you have not looked at the system holding it for two weeks, your head takes the job on itself.

And it does not speak up during working hours, when you would be on it anyway. It speaks up when things go quiet. At dinner, in bed, on a walk with the children.

That is why mental absence in your private life is a symptom of a work system and not a question of mindfulness. You cannot decide to be present while forty open loops are running in the background.

The lever: separate deciding from doing

Knut's approach reverses the usual order. He plans the next day the evening before, not in the morning.

It sounds like a small thing and it is not, because it solves four problems at once.

In the morning, the decision competes with the execution. Plan in the morning and you spend the clearest time of the day on a decision you could have made the night before. With the plan already standing, you start instead of choosing.

The evening has more information. At five o'clock you know what was left undone today, who has not got back to you, and what is realistic tomorrow. At eight in the morning you are reconstructing all of that.

Your head works on it overnight. If you know in the evening that you will write a particular concept tomorrow morning, you wake up with approaches to it. This is not an esoteric argument; anyone who has gone to bed with an open question knows it.

And the most important point: it is a line under the day. The working day gets an ending that consists of more than closing the laptop. What is due tomorrow is settled and therefore taken care of. That is precisely the condition for your head being free at dinner.

The third effect is the one Knut is after. The other three are by-catch.

What it looks like in practice

Five to ten minutes at the end of the working day:

Look at tomorrow's calendar. What is fixed, how much is left.
Fix one to three tasks that genuinely have to happen tomorrow. The test is put harshly: if you do not do them tomorrow, you have a problem. Anything that description does not fit is a candidate, not a selection.
Skim the waiting-for list. What has been open for more than a week and needs a call tomorrow.

Done. Laptop shut.

In the morning, the question of where to start simply disappears. You start. What pushes against the plan at that point usually comes from outside — more on that in saying no at work.

The cockpit: your software does not open where the requests are

The second part of Knut's setup is a configuration hardly anyone makes, even though it takes five minutes.

His Outlook does not start in the inbox but in the calendar and task view. And it starts offline. Only once the one to three focus tasks are done does he go online.

The principle transfers to any tool. The opening view of your mail program decides what you see first, and what you see first determines what you think about first. If you work in a different program, you get the same effect through your browser start page, a different default tab, or simply by leaving the mail window closed in the morning.

The sentence behind it is the actual point: email is an asynchronous medium. It is built for sender and recipient not being present at the same time. Treat it synchronously and you have voluntarily set up a chat with an obligation to reply, using the worse tool for the job. How the inbox gets worked through afterwards is in process email instead of working it.

Friday, right after lunch

His weekly review sits on Friday afternoon, but deliberately straight after lunch and not at half past four.

The reasoning is unspectacular and therefore right: at half past four nobody has the energy left to resolve open ends. You skim, tick things off and go home. The review that actually does something requires the willingness to look at unpleasant projects and make decisions, and that willingness is still there in the early afternoon.

He describes the result as a holiday feeling. Friday evening does not start for him with a remainder in his head, but with a closed system.

That matches what I saw again and again in trainings: put your review too late and you lose it first in substance and then entirely. It does not fall away because there is no time, but because in the state you do it in, it achieves nothing — and you notice.

My own take

Two remarks, because at one point the model departs from the pure doctrine.

A fixed daily selection is not part of GTD, because it forces a pre-decision that can be obsolete by eleven o'clock. In practice it works anyway for people who otherwise lose themselves in options, under one condition: the selection comes out of the existing lists and does not become a second system alongside them.

And the number. One to three is considerably more realistic than the usual five or six, especially in a role with many meetings. Take on three things and finish two and you have had a good day. Take on six and finish two and you feel in the evening that you failed, at identical output.

Knut's line on this, which I have been quoting ever since: it has to be stupidly simple so that it works in a storm. Any system that looks good on an ordinary Tuesday and collapses on a chaotic Thursday is too complicated.

Where it typically fails

The evening review turns into a second work block. Answer three more mails in the evening and you have missed the line under the day and merely postponed the end of it. Five to ten minutes, then stop.
The selection is made next to the system. A note with three items that exist nowhere else is a parallel system. Selecting means marking, not copying out.
The focus tasks need more time than the day has. Three tasks on a day with six meetings is not planning. Look at the calendar first, then choose.
The weekly review is missing. Without it you plan in the evening on the basis of lists that no longer hold, and your head notices faster than you do.

Common questions

Why should I plan my day the evening before?

Because you have more information in the evening, you no longer have to decide in the morning, and the working day gets a clear ending. That last point is the decisive one if work follows you mentally into the evening.

How long does evening planning take?

Five to ten minutes: tomorrow's calendar, fix one to three focus tasks, skim the waiting-for list. Longer usually means you have started working again.

How many tasks should I fix for one day?

One to three, depending on the free time left after meetings. The test: if you do not get them done tomorrow, a real problem appears. Everything else is a candidate, not a selection.

Why should Outlook not open in the inbox?

Because the first view determines what you think about first. Start in calendar and tasks and you begin with your priorities instead of other people's. Email is an asynchronous medium and tolerates an hour of delay.

When is the best time for the weekly review?

While there is still energy to resolve open ends. Early Friday afternoon works better than late, because a review without decision-making capacity does take place but changes nothing.

If work regularly follows you to the dinner table, it is rarely because too much is open. Usually it is because nowhere does it say what happens with it tomorrow.

For the evening pass to hold, calendar, lists and inbox have to fit together. That is what a workflow setup puts in place. Write to me if your day starts from zero every morning.

Your plan for the day survives until quarter past ten

Mon, 22 Jun 2026 00:00:00 GMT

You did everything right. Looked at the calendar before the inbox, counted the free windows, picked four tasks that make the difference today. Twenty minutes, done properly, good start.

At 10:15 a colleague appears in the doorway, or a message arrives that opens with "got a minute?". And then the plan is gone.

The usual advice is to plan the morning better. That does not help here. The plan was fine. What was missing was the ability to defend it, and that is a different competence from planning.

The morning start in three steps, briefly

Since it is the precondition, here is just the frame:

The calendar first. What are today's hard boundaries, how much free time actually remains after meetings, travel and follow-up. If there is more in there than hard appointments, that is a topic of its own: the calendar lies about your time as soon as tasks are sitting in it.

Then the context lists, filtered down to what is even possible today. If you are working from home, you do not look at the office list.

Then the selection. Not by a fixed number, but by the free time from step one. On a day with five meetings, two tasks are ambitious.

And only then the inbox. How that part runs in detail is in process email instead of working it.

That is a good routine, and it has a known weak point: it assumes the day belongs to you.

The plan does not die from the inbox, it dies from people

Email can be managed with discipline. You close the window, you turn off notifications, and then it is gone.

A person who wants something is harder to switch off. And most people say yes even though the day is full, because in that moment a no feels more expensive than three extra hours in the evening.

That is exactly where the thinking goes wrong. With a request, the question is never "can I do this". Almost everything is doable. The question is "where do I take the time from", and that question always has an answer. It just usually stays unspoken, and then you pay it with whatever was actually due today.

Once you see it that way, you say no differently. Not because you do not want to help, but because you know what the yes costs.

The honest no, in three parts

A usable no has three components, and the order matters.

Show the good intention. "I'd be glad to help you with that." That is not a platitude, it is the information that the refusal applies to the matter and not to the person. Without this part every no sounds like a brush-off.

Give the reason, briefly. "Not today, I have to get the proposal out." One sentence. Not three.

Offer a real alternative. "Tomorrow at nine I'd have an hour." Or: "Ask Martin, he worked on that last week."

The third part decides whether the no lands as cooperative. A no without an alternative is a door closing. A no with an alternative is a detour, and nobody takes a detour personally.

Where saying no goes wrong

Too much justification. The longer you explain, the more it sounds like an excuse and the more surface you offer. Name three reasons and you invite all three to be argued away. One sentence is enough, and it should be concrete.

The reason first. If you open with the justification, for the first few seconds the other person hears a defence without knowing what it leads to. Intention first, then reason.

The alternative is not one. "Maybe some other time" is not an alternative, it is a postponed no that comes back as a follow-up in two days. A concrete slot or a concrete name, nothing else.

Fabric softener. "I'm afraid I might possibly have some difficulty there." The other person hears half a yes and asks again. Being clear is the friendlier option, because it creates work for nobody.

The answer that is almost always the right one

In practice it is rarely about a fundamental no. It is about today.

"Yes, but not today" is the answer that fits most cases and that hardly anyone uses. It costs nothing, it preserves the relationship, and it protects exactly what deserves protecting: the time you allocated to yourself this morning.

The question behind it is whether a request really has to happen today. For a surprisingly large share, the honest answer is no, but nobody asks, because urgency rides along in the tone and gets adopted unchecked.

When you cannot say no

With a boss, an important client or a genuine escalation, declining is not an option. There is still a way, and it is the most useful sentence in this whole article.

Make the displacement visible and let the other person decide.

"I can take that on. Then the proposal moves to tomorrow. Is that all right?"

You are not saying no. You are only making explicit what your yes costs, and handing the decision to where it belongs. In half the cases you get back an "oh, no, then finish the proposal" — from exactly the person who would otherwise have thought you unreliable when the proposal was left sitting.

That is also why this phrasing matters more than any no-technique: it works upward, and most priority conflicts come from above.

The smartphone on the nightstand

One detail decides the whole sequence. Anyone who looks at the display first thing in the morning has other people's priorities in their head before they have formulated their own. After that the morning routine is no longer planning, it is a reaction with structure.

The device belongs in another room, and an alarm clock is an alarm clock. This is not a lifestyle topic, it is the order in which information enters your head.

Where it typically fails

The plan exists only in your head. If what you intended for today is not visible, you cannot point at it during the conversation. Then you say yes, because nothing comes to mind in that moment that speaks against it.
No becomes a default stance. Someone who deflects everything becomes the colleague nobody asks any more, and thereby loses the requests they would have wanted. It is about the selection, not the deflection.
The no comes too late. A yes withdrawn three days later costs more than any no on day one.
There is no consequence. If you say it will not work today and then work on it at eight in the evening anyway, the people around you learn that your no is negotiable. And they are right.

Common questions

How do I say no at work without damaging the relationship?

In three parts: show the good intention, give the reason in one sentence, offer a concrete alternative. The alternative is the part that decides whether it lands as a refusal or as a detour.

What do I do when I cannot say no?

Make the displacement visible instead of declining: name what the new yes pushes back, and leave the decision to the person asking. This also works with superiors.

Why should I not read email first thing in the morning?

Because you then have other people's priorities in your head before you have formulated your own. Every incoming message is judged without a yardstick after that, and seems more urgent than it is.

How many tasks should I plan for one day?

As many as fit into the time that is actually free. First subtract meetings, travel and follow-up from the day, then choose. A fixed number like five ignores that days are differently full.

Next time someone asks "got a minute?": do not ask yourself whether you can, ask where you would take the time from. And say the answer out loud.

The phrasing is one thing, the actual situations another. If you want to walk through those conversations beforehand, sparring is the place for it. Write to me.

A well-kept context list can paralyse you just as much as chaos

Sat, 13 Jun 2026 00:00:00 GMT

Just for the opposite reason. In chaos you find nothing. On a well-kept @Computer list you find forty-eight things that are all doable, all sensible, and all not urgent enough to push themselves to the front.

That is not a flaw in the system. Context lists are sorted by resource, meaning by what you need in order to carry something out. They are storage and a menu of options. They are not sorted by what you intend to do today, because they cannot know that.

The intention for today is missing, and it has to come from you. A kanban board for the day is a usable place for it, provided you avoid the three mistakes that normally sink hybrid setups like this.

The objection first: GTD has no daily list

Anyone who knows the method has an objection ready here, and it is a fair one. A list of today's tasks forces a decision in advance that can be obsolete by eleven. That is why GTD works with contexts rather than daily plans.

A daily board still stays clean if three conditions hold:

It is a window, not a copy. Whatever is on the board is not simultaneously somewhere else as a task.

It can be restocked at any time. When the day turns, cards come off and others go on. Without guilt and without the board turning into a list of failures.

It has no truth of its own. The weekly review goes over the system lists, never over the board. The board exists for one day and is empty by evening.

Hold to those three points and you have no second system, only a view. Break them and you have made yourself work instead of clarity.

The layout

Four columns, no more.

Today. What you set out to do on this day.

In progress. Exactly one card.

Waiting for. A holding bay for cards that will not move today because you are waiting on somebody.

Done. What you got through, visible until the evening.

That last column looks like decoration and is not. Knowledge work leaves no traces. Somebody who cannot see in the evening what they did will consider the day unproductive even when it was not. That is why tradespeople go home with that feeling less often.

The rule everything hangs on: no copies

This is where the majority of these setups break after two weeks.

If you transfer five tasks from your context list onto a second board every morning, every task exists twice. You then have to maintain both places, tick off every completion in two spots and enter every change twice. Nobody keeps that up, and when it breaks you no longer know which list is right. From that point you trust neither.

Two ways to avoid it:

Digital: filter instead of copying. Not a second board but a view onto the same data. A label "today", a filter on it, and the columns as a status on the card itself. In Trello a label plus the board filter is enough; in other tools it is saved views. The task stays one object, you just look at it differently.

Analogue: move it physically. On paper or with cards, the card is the thing. It is either in the drawer or on the daily board, never both. That is exactly what makes analogue boards strong here, even though they are technically primitive.

What you should not do is a second digital board with tasks copied across. That is the route by which a magnifying glass becomes a parallel system.

WIP limit: the actual point of kanban

The columns are not the valuable part of kanban. The valuable part is the limit on how much may be in progress at once. Personal kanban can be reduced to two rules: make your work visible, and limit what runs at the same time.

One card in "In progress" is the hard version, and it is right. Not because multitasking is morally objectionable but because every switch between two tasks costs run-up time and you do not feel that cost. You only feel that the day was full.

Two clarifications, without which the rule fails in daily life:

Cards in "Waiting for" do not count against the limit. Otherwise you block yourself as soon as something sits with somebody else. Waiting is not work.

Interruptions are not a new card in "In progress". When the boss calls and something comes up, the current card moves back to "Today" before the new one comes in. The detour takes three seconds and is the difference between a board that reflects the day and one that lies.

Waiting for is only a holding bay

There is a trap here that does not show up at first. The column on the daily board is not a substitute for the waiting-for list in the system. It is a place for "no further today, because I am waiting for an answer".

By evening this column has to be empty. Every card sitting there moves to the real waiting-for list, with a date and a person. Skip that and delegated work disappears along with the board, and that is the most expensive way to lose trust in a system.

What arrives during the day does not go on the board

The whole point of choosing in the morning is that it was a decision. If everything new goes straight onto the board during the day, you have a reactive list again, just more colourful.

New things go into the inbox. Full stop. At the next processing block you decide what becomes of them, and if it has to happen today it goes onto the board — and something else comes off. That movement, having to take something down, is the real value of the limit. It makes visible that a day does not get bigger just because more arrives.

Capacity, not count

Three to five cards is a common recommendation and still the wrong measure, because it bears no relation to the time available.

Look at the calendar first. How many hours are left after meetings, travel and follow-up. On a day with six meetings maybe one hour is left, and then one card is the right number. Five cards on a day like that is not planning, it is a list that makes you look bad in the evening.

Closing the day, in two minutes

Empty the board. What is done disappears. What is not done goes back onto the context lists, not onto tomorrow's board. Waiting-for cards move to the system list.

The reason for putting them back: a card that has sat in "Today" for three days is not a reminder, it is a reproach. And the decision about whether it is due tomorrow belongs to tomorrow, not to today.

Where it typically fails

Duplicate data. The classic, see above. Filter instead of copying.

The board becomes the system. As soon as the weekly review goes over the board rather than the project and context lists, the separation is gone and the completeness check is missing.

Too many columns. Add "Review", "Blocked" and "This week" and you have a project management tool rather than a window on the day.

Cards without a concrete action. If a card holds a topic rather than an action, even the finest board will not help. The problem then sits one level down, in how the next actions are phrased.

The board does not get emptied in the evening. After a week "Today" is full of backlog, and the thing is more demoralising than the long context list it was meant to replace.

Common questions

How do I combine GTD and Kanban?

The system lists stay in GTD, meaning projects, context lists and waiting for. The kanban board is a daily view with four columns, stocked in the morning and emptied in the evening. What matters is that the board holds no copies but displays the same tasks differently.

What is a WIP limit?

The upper bound on how many tasks may be in progress at once. For individuals one task is the sensible limit, because every switch costs run-up time. Waiting tasks do not count.

Does a daily board not contradict the GTD method?

Only if it becomes a second task list. As a view onto the existing lists, restockable at any time and emptied in the evening, it complements the method rather than replacing it.

How many tasks belong on the daily board?

As many as fit into the time that is genuinely free. Look at the calendar first, subtract meetings, travel and follow-up, then decide. On full days one card is the right answer.

Analogue or digital?

Analogue enforces the card's singularity, and with it the most important rule, by itself. Digital is more practical if you move between devices, but it needs filters rather than a second board.

Build the board for a single day and note in the evening what you learned. Usually it is the number: that less fits into a day than you believe in the morning, and that better planning does not change it.

Whether a board stays a view or becomes a second system is decided when you set it up. A workflow setup does that with your data. Write to me if you would rather build it properly once.

An afternoon designing boards feels better than two finished tasks

Tue, 09 Jun 2026 00:00:00 GMT

And that is the entire trick by which this behaviour sustains itself.

When you rebuild a board, link dependencies and assign colours, you get immediate feedback. It looks better after every change than it did before. It cannot fail. And nobody can contradict you while you are doing it.

The task that is actually due has none of that. It has unclear feedback, it can go wrong, and whether it was any good you only learn later. Between those two options your head reliably decides against you, and because configuring looks like work, nobody notices. Least of all you.

How to spot it

Six symptoms, of which two are enough:

You have tried more than two task managers in the past twelve months.
You know your tool's feature roadmap.
There are fields in your system you have not filled in for weeks but do not delete either.
The thought "let me set this up properly first" comes up more often than "let me do this now".
You read tool comparison articles although your current one works.
After the weekly review you feel good, but in the week that follows no more happens than before.

The last one is the most reliable. A system does not exist to feel tidy, it exists so that something happens afterwards.

The two habits that actually change something

In one of our sessions a community member showed the evolution of his system over several years. What was interesting was not the list of tools he had worked through but what made the difference in the end.

It was two things, and neither of them is software.

The weekly review. Once a week, go through the system instead of hoping it keeps itself current. Without that appointment the lists drift so far from reality within two weeks that you no longer trust them, and then you are carrying things in your head again.

Notifications off. All of them. Not out of focus romanticism but because a system you use between two interruptions never actually gets used.

These two habits work in any tool. They work on paper. Whoever does not have them will not find them in any software, and whoever does have them will get by with almost anything.

And the system change was still right

Here it gets interesting, because the same person then rebuilt their entire setup. That sounds like the contradiction I described above, but it is not. The difference lies in the reason.

His previous tool got more expensive, worked only moderately offline, and the data sat with a provider he had no control over. Those are reasons that lie outside the system.

From which a rule can be drawn that I have used for years:

A system change is justified when the reason lies outside the system. Cost, data protection, a company mandate, a platform change, a provider discontinuing the product.

A system change is suspect when the reason is a feature. "The other one does nested subtasks" is almost always tool procrastination with an argument attached.

And the order matters. Habits first, then the system question. Work the other way round and you migrate your chaos into a new interface and wonder about it six weeks later.

His setup, and what it actually costs

He moved to infrastructure he controls himself:

A Nextcloud at a German host, by his account for just over five euros a month with a terabyte of storage, with full administrative rights.
Nextcloud Deck as a kanban board for the project overview.
Super Productivity as a local list manager on desktop and phone, synced over the Nextcloud.
Plus a document scanner and cloud storage for the paperless archive.

That is a tidy, cheap and field-tested setup. And now the part that is almost always missing from articles like this.

The five euros are not the price. Your own Nextcloud means applying updates, setting up backups and occasionally checking that it still runs. When syncing breaks after an update, nobody is responsible except you, on the day you had other plans. Someone from an IT background factors that in and finds it fine. Someone who does not has taken on a project and is calling it a productivity system.

My honest calculation: hosting plus roughly two hours of operations per quarter, plus half a day for the initial setup. Compared with many subscriptions that is still cheap, but it is not free.

Data ownership, without the drama

The question is a fair one and usually gets asked too broadly. It is more useful in concrete form: which entries in your system would bother you if they sat with a provider in another legal jurisdiction?

In a task manager that is usually not the tasks themselves but the names. Client names in project titles, notes from conversations, in some professions patient or case references too. Anyone working with those has a substantive reason for their own infrastructure. Anyone managing shopping lists and renovation projects does not, and is allowed to see it that way.

Two clarifications, because too much gets claimed here:

A server location in the EU makes things easier but is neither a condition nor a guarantee. Processing outside the EU is lawful when the basis for it is right, for instance an adequacy decision or standard contractual clauses. And a location in Frankfurt makes nothing compliant: compliance depends on what you process, on what basis, how long you keep it and whether the necessary contracts exist. Self-hosted also means you are responsible yourself, including for security. That is more control and more duty, not less.

And the practically most important point in any tool choice is almost always overlooked: can you get out again? An export in a readable format, Markdown or CSV, decides whether the next price rise leaves you over a barrel. Systems whose content sits in text files outlive their providers. That is why I keep my own reference material in Markdown and not in a database only one app can read.

The test before switching

Three questions before you migrate. Answered honestly this takes ten minutes and may save you a weekend.

What exactly can my current system not do that I genuinely needed in the past four weeks?
Is that a tool problem or a habit problem? If your weekly review gets skipped, it will get skipped in the new tool too.
What does the migration cost in hours, and what would I otherwise have done with those hours?

If question one has no concrete answer, the matter is settled.

Where it typically fails

The move turns into system work rather than a migration. "I'll move my lists" becomes "I'll rethink my system", and three weekends later the setup is prettier and the projects are where they were.

Self-hosting without backups. The moment you find out is always the same one.

Forcing everything into one tool. Reference material, projects, tasks and archive do not have to live in the same software. What has to fit together is the logic.

The habits get postponed until after the move. They then do not arrive, because after the move there is the next move.

Common questions

What is tool procrastination?

Working on your productivity system instead of the work it is meant to organise. It sustains itself because configuring delivers immediate, visible feedback and cannot fail, while real tasks offer neither.

When does switching task management tools make sense?

When the reason lies outside the system: rising cost, data protection requirements, a company mandate, a platform change or a discontinued product. A missing feature is rarely a sufficient reason.

What does a self-hosted Nextcloud really cost?

On top of hosting in the single-digit euro range come setup, updates and backups. Realistically half a day at the start and a few hours per quarter in operation. Cheaper than many subscriptions, but not effort-free.

Is an EU server location enough for GDPR compliance?

No, and it is not strictly required either. Processing outside the EU is lawful with the right basis, for instance via an adequacy decision or standard contractual clauses. Compliance depends on purpose, legal basis, retention period and the necessary contracts. With self-hosting, responsibility for security lies entirely with you.

What should I look for when choosing a tool?

The export. If you can get your data out in a readable format such as Markdown or CSV, the next price rise cannot hold you hostage. Systems made of text files outlive their providers.

Look at your calendar for the past four weeks and find the hours you spent working on your system rather than with it. If you cannot find them because they were never entered anywhere, that is already the answer.

Whether your tool really is the problem can be settled in one conversation — cheaper than a migration. That is what sparring is for. Write to me before you move.

More tools do not make unclear people clearer

Wed, 03 Jun 2026 00:00:00 GMT

The sentence comes from one of our sessions and it drew objections there. Fair enough, because at first it sounds like scepticism about technology. It is not. I automate wherever I can and have scripts written for me that used to cost me two evenings.

The point is a different one, and it is arithmetical.

The costs shift, they do not disappear

Writing a decent email used to cost ten minutes. Today it costs thirty seconds, because a model does the phrasing.

Reading that email, understanding it, deciding what it means for me and phrasing a next action out of it costs exactly as much as it did three years ago.

That is the whole mechanism. Costs on the sending side have fallen, costs on the receiving side have not. And because whatever gets cheaper gets done more, the volume rises. Your inbox does not get lighter because you use AI, it gets heavier because everyone else does too.

Anyone who had no process for turning input into decisions now gets the same problem at a higher frequency.

The summary fallacy

A second effect is subtler and therefore more dangerous.

When an AI condenses a mail thread with fourteen messages into five sentences, it feels like processing. But it is only reading, in short. The actual work is not knowing the content, it is deciding what it means for my commitments. Whether I have to do something. What exactly. By when. Whether I want to at all.

That decision stays entirely with me. It just feels done after the summary, because the sense of having an overview is there. That is why people with AI assistance have often read more and clarified less.

What AI can genuinely do in self-management

There is a clear line, and on one side of it the benefit is considerable.

Capturing. A voice note while out and about, automatically transcribed into the inbox. The hurdle between thought and capture is where GTD most often breaks in practice, and that hurdle can be lowered technically.

Extracting raw material. Pulling everything that sounds like a commitment out of a long thread or a meeting note. Important: the result lands in the inbox, not on a project list. It is material, not a decision.

Preparing the weekly review. Tasks completed this week, calendar entries, projects without a next action, waiting-for entries older than ten days. A script can gather that, and you start the review thinking rather than searching.

A sparring partner for thinking things through. In brainstorming on a stuck project, quantity is what counts and evaluation comes later. That is exactly where a model is useful, because it gives you twenty approaches of which three are any good. The selection stays with you, but the selection was never the expensive part.

Opening up reference material. Making what you have filed searchable. Unspectacular, but it genuinely saves time.

And on the other side of the line stands exactly one thing that no model takes over.

Only a person can make a commitment

GTD does not manage tasks, it manages commitments. The reason unfinished things circle in your head is not that they are unfinished but that you committed to them without deciding what becomes of them.

That decision cannot be delegated, not even to a very good model. Whether a project is worth it. Whether you withdraw a commitment rather than postpone it. Whether you tell a colleague you will not manage. That is the heart of the matter, and it consists of decisions that have consequences for you.

A model can help you make them better prepared. Making them is yours.

Commitment has a price, and software has none

Ignoring a push notification costs nothing. Standing up a running partner who is outside your door at six in the rain costs something.

This is not a motivation topic but plain economics. Intentions hold when breaking them has a cost, and social costs are the most effective ones available in daily life. That is why study groups, training partners and weekly sessions with the same people work, while reminder apps get swiped away after eleven days.

The same applies to habits in self-management. The weekly review you plan to do alone on a Friday gets skipped. The weekly review three colleagues meet for at one o'clock on a call happens.

Self-management has long been a team sport

The larger part of what is on your lists depends on other people. You are waiting for an approval, you need information, you delegated, somebody delegates to you.

That makes the quality of your system more than your problem. If you have precisely phrased next actions on your list and your team passes around topics without actions, you spend your time clarifying on behalf of others.

This is exactly the point at which individual work hits a limit. One person with a good system inside a team without shared vocabulary is well organised and still blocked constantly.

What actually changes with a shared language

Culture is a soft term, so let us be concrete. When a team uses the same terms, four things change visibly:

The question "what is the next action and who does it" gets asked at the end of every meeting, without anybody having to facilitate it.

Mails contain an action in the text, not just a topic. The difference between "FYI" and "I need your approval by Thursday" is made explicit rather than guessed.

Delegated tasks are tracked by both sides, once as a task and once as a waiting-for. That ends chasing out of nowhere.

And there is a difference between "I've seen it" and "I've clarified it" that can be spoken out loud without sounding like an excuse.

Internal trainers, and when they achieve nothing

More and more companies are training their own people rather than buying trainings in. The reason is not price but context. Someone who knows the internal processes, the tool landscape and the quirks of how people communicate can bring examples that are true for the people in the room. Nobody from outside has that context on day one, and an AI does not have it at all.

Two qualifications belong with that, otherwise it sounds like a sales pitch.

Without a time budget an internal trainer is a job title. Give somebody the role on top of a full-time job and they will run two workshops and then never another. The role needs hours in the calendar, otherwise it belongs in the category of well-meant.

Without leadership nothing happens. If tasks are handed out by chat at eleven at night with an expectation of immediate replies, no shared vocabulary will help. People then learn a language in which they can describe why their system does not work. That is worse than before.

What holds up in practice is a small recurring session. With us that is a weekly one; in companies half an hour every two weeks with the same people works. Not as a class but as a place where you can say what did not work this week. Same mechanism as the running partner.

Common questions

Can AI take over my self-management?

Partly. Capturing, pre-sorting, extracting raw material and preparing the weekly review can all be handed over well. The decision about whether you take on a commitment, keep it or withdraw it stays with you, because it has consequences for you.

Why does AI make the flood of information bigger rather than smaller?

Because AI lowers the cost of creating, not of processing. Writing a mail takes seconds; clarifying it takes as long as ever. The load shifts from sender to recipient.

Why is an AI assistant not enough for new habits?

Because ignoring a notification has no cost. Commitment comes from social costs, meaning from people who notice. That is why fixed sessions with the same people work better than reminder apps.

What does an internal trainer bring to a company?

Shared terms, examples from their own context, and a recurring session that keeps habits alive. The prerequisites are a real time budget for the role and a leadership team that follows the rules itself.

If you use AI in your self-management, look at where it makes decisions easier and where it merely gives you the feeling of having decided. The difference lies in whether something ends up on a list afterwards.

Where the line runs in your organisation depends on your processes and your works agreement. I clarify that under automation and AI. Write to me.

Two weeks before the trip, the campervan was a write-off

Sun, 24 May 2026 00:00:00 GMT

It was meant to be the big father-and-son holiday. Possibly the last in this constellation, before my son reaches an age where he has other plans. Everything prepared, route roughly in my head, dates fixed. Then the campervan was written off.

The first reaction was not a methodical one. Frustration, a bit of panic, and that sense of standstill in which you think a great deal and decide nothing.

What got me out of it was a look at my own project setup. And there I noticed that in the entire plan the word campervan did not appear a single time. The purpose I had written down was time with my son and having an adventure. The campervan had never been the goal, it was a means of transport.

We ended up travelling the north with rucksacks. Looking back, better than the original plan.

Why that was not luck

The reason I did not spend weeks stuck is banal: the purpose was written down. Not in my head but legible, in a place I looked at during a moment of panic.

Keep the purpose in your head only and you confuse it with the route. And when the route collapses, it feels as if the goal is gone. That happens in companies every week. The supplier drops out, the software does not deliver what was promised, a key person resigns. And then people try to rescue the plan instead of asking which other route serves the same purpose.

The test is simple and uncomfortable: look at your project names. Is there an outcome there, or a vehicle?

"Roll out Salesforce" is a vehicle. "New website" is a vehicle. "Plan campervan holiday" is a vehicle. When the vehicle is in the project name, you committed to a route before you phrased the goal, and nobody on the team knows what the effort is for.

The five steps, in the order your head uses anyway

The Natural Planning Model is called that because your brain approaches projects in this order by default. Deciding to eat out with friends, everybody does it without thinking: first the purpose (we want a nice evening), then the picture (Italian, not too loud), then the ideas (three places come to mind), then the organising (book a table, who picks up whom), then the first action (call).

1. Purpose and principles. Why are we doing this, and what are the standards? The purpose is the one sentence that still holds when the plan tips over.

The principles are the underrated part. The most useful question for them: what would have to happen for this project to drive me up the wall? The answers are your rules. "I don't want to be packing the night before." "Nobody on the team works weekends for this." "We want to enjoy this" is a perfectly legitimate standard, including at work.

2. Vision. What does the outcome look like when it has gone well? Not as a target but as a picture. What do I see, what do I hear, what are people saying?

That sounds more esoteric than it is. The practical benefit: while the picture is missing, nobody can judge whether an idea is any good. Teams working without a vision debate every detail from first principles.

3. Brainstorming. Everything out of your head, without judgement. The most common mistake is gathering and evaluating at the same time. You then brake yourself at every third thought, and the good ideas that sound stupid at first never reach the paper.

Paper usually works better than a tool for this, because no structure is imposed. In a team: everybody writes silently for themselves first, then you collect. Otherwise whoever speaks fastest prevails, rather than whoever has the best thought.

4. Organising. Only now do you sort. What belongs together, what depends on what, what is sequential and what runs in parallel.

5. Next actions. The very first physical action you can take now. A project with several strands can have several, one per strand, but each has to be concrete. "Research" is not an action. "Pull three suppliers out of the trade directory" is.

The most important point: you do not need this for every project

This is where most attempts fail. Try to force your sixty open projects through five stages and you will stop after the fourth and conclude the model is bureaucracy.

The vast majority of projects need exactly two things: a phrased outcome and a defined next action. That keeps them running. The full model is the tool for the projects that feel vague, where nothing is moving, or where something has broken.

You typically have two to five of those at any one time. For those, an hour is worth it.

Diagnosis: which step is it stuck on

When a project is stuck, the symptom tells you which step is missing.

What you notice	What is missing
Endless debate, every decision reopened	Purpose
Everyone is working, nobody knows if the outcome is enough	Vision
The feeling of forgetting something important	Brainstorming, not everything is out
Plenty of ideas, no recognisable route	Organising
It feels clear, but nothing moves	Next action

From which follows a rule of thumb that almost always holds: if the project needs more clarity, go up, to purpose and vision. If it needs more movement, go down, to the next action. The middle, brainstorming and organising, is rarely the problem.

The unnatural route most people take

The counterpart to the natural model is not worse planning but none at all. You start straight at step five, work through action after action and hope the meaning emerges along the way.

That works for small undertakings and, on large ones, is the reason projects suddenly turn heavy in the third month. Not because people have become less capable, but because nobody can say any more what this is for, and motivated people depend on exactly that.

Anyone who has ever asked in a company why a rollout is actually happening knows the answers. Four different ones, from five people.

Where it typically fails

The purpose stays in your head. A purpose that is not written down does not help in a crisis, because in a crisis you do not think calmly. Write it down, attach it to the project, done.

The why is stated once on the kickoff slide and never again. After three months half the team no longer knows it, and the new people never heard it.

Brainstorming with simultaneous evaluation. Costs you half the ideas, and the more interesting half.

The model gets applied to everything. A clarifying tool then becomes a ritual, and nobody keeps rituals up for long.

Common questions

What is the Natural Planning Model in GTD?

A planning sequence in five steps: purpose and principles, vision, brainstorming, organising, next actions. It is called natural because the brain approaches projects in this order by itself, as long as it is not prevented from doing so.

Do I have to go through all five steps for every project?

No. Most projects only need a phrased outcome and a next action. The full model is meant for projects that are stuck, feel vague, or whose plan has tipped over.

What do I do when a project is stuck?

Check whether clarity or movement is missing. For missing clarity, go back to purpose and vision; for missing movement, define the next physical action. The symptom shows the direction.

Why should I write down a project's purpose?

Because it is the only part of the plan that still holds when the route collapses. Keep the purpose in your head only and you confuse it with the chosen route, and you consider the project failed when only the vehicle has dropped out.

Take the project you least want to think about right now and write down in one sentence what it is for. If that sentence will not come to you, you know why it feels the way it does.

A stuck undertaking often comes loose in an hour with someone who asks the right question. That is the purpose of sparring. Write to me if you have one sitting there.

The most common answer to "which lists do you keep?" is "one"

Tue, 19 May 2026 00:00:00 GMT

And that one list then holds "nail for the picture in the kitchen" right next to "plan family holiday" and "company antivirus rollout". Plus an entry called "tax" with no further hint of what was meant.

Look at a list like that and you cannot start. Not because it is too much, but because every entry demands a different kind of thinking. The nail needs a hammer and four minutes. The holiday needs an hour of thinking with your partner. "Tax" first needs a decision about what it even refers to. Your head bounces between these demands and ends up doing nothing.

The answer is not sorting by priority. The answer is that each list answers exactly one question.

The question is the organising principle

Every list discussed here can be defined by the question you open it for. If you cannot say which question a list answers, you do not need it.

List	Question
Calendar	What has to happen today at a particular time?
Next actions, by context	What can I do now, here, with this energy?
Waiting for	What am I waiting on, from whom, since when?
Projects	What have I committed to?
Agenda, per person	What do I want to raise the next time we speak?
Someday/maybe	What do I not want to forget but not do now?
Reference	What do I want to find again when I need it?

Seven questions, seven places. The rest of this article covers what goes wrong with them in practice.

Calendar: the hard landscape

The calendar takes only what has to happen on a particular day or at a particular time. Appointments, flights, day-bound information.

Tasks do not belong in it, because the calendar knows nothing about backlog. Whatever did not get done in a past time slot disappears unnoticed instead of staying open. That is why people with full calendars swipe away their own reminders.

Next actions: context is more than a place

This is where the concrete physical actions live, sorted by what you need in order to carry them out. It is the list you keep open in working mode, and the only one holding something you can do right now.

Two things about it that most guides leave out.

The phrasing decides whether it is usable. "Quote for Miller" is not a next action, it is a topic. If a topic sits there, you redo the thinking every time you look at it — the thinking you should have done when you captured it. "Transfer Miller's costing from the sheet into the proposal template" costs twenty seconds more when capturing and saves five minutes of run-up at every glance.

The classic contexts no longer all work. @Computer made sense when there was one computer and it stood on a desk. Today @Computer is a list holding ninety per cent of all tasks, and therefore worthless.

What holds up in practice: @Calls works unchanged, because it has a mode of its own. @Out and @Errands work, because they are tied to a place. @Home works. And the most useful cut today is not the place but the capacity available. One list for things that need full concentration, and one for everything that also works with half a brain on a train. Tired in the afternoon, you open the second one and still get work done.

Waiting for: the list that is almost always missing

Everything you delegated or requested and are still accountable for. With a date from which you have been waiting.

This list is missing in roughly every second system I see, and it is the most common cause of unpleasant surprises. Without a waiting-for list you remember an open commitment only when a client asks. And you then delegate less, because experience tells you delegated things vanish.

The date is not decoration. It is the only piece of information that turns an entry into an action. Something open for four days you leave running. Something open for three weeks is a phone call.

Projects: anything needing more than one action

A project is an outcome requiring more than one action. In GTD that is a low threshold, which is why thirty to a hundred open projects is normal with a full working and private life.

Nothing on this list is something you can do. It answers not "what do I do now" but "do I have everything in view". You ask that question once a week, not at nine in the morning.

Phrase entries as an outcome, not an activity. "Antivirus rollout" is open-ended. "New antivirus solution active on all 240 clients" has a state you can recognise when it arrives.

Agenda lists: the underrated lever

One list per person you deal with regularly, and one per recurring meeting. Everything that is not worth a call of its own but should be raised the next time you speak lands there.

This is the list with the best effort-to-benefit ratio in the whole system, and the one kept least often. Two effects: you interrupt people considerably less, because small things can wait, and you never again leave a standing meeting realising afterwards what you had meant to ask.

With managers this is often the point at which the system changes something noticeable for the first time.

Someday/maybe: the valve

Everything you might want to do at some point but not in the coming weeks. The language course, the renovation, the book project.

This list has two functions. It keeps the active project list honest, because then only what is running is running. And it lets you write ideas down without committing to them. That permission is exactly why people start capturing everything in the first place.

The condition: skim it once a week. A list nobody looks at is a graveyard, and your head notices that soon enough and stops trusting it with anything.

Reference: everything with no action

Information where nothing needs doing but which you want to find again. Credentials, addresses, notes from meetings, contract documents.

This is also where my favourite among the minor lists belongs: the gift list. When somebody mentions a wish in passing, it goes straight in there, with the name. Before the next birthday that replaces half an hour of pondering with a glance. It is not a project and not a task, it is material that gets used exactly once a year and is then worth gold.

The boundary matters, because this is where everything tends to slide together. A list of films you want to watch is reference, as long as it is only a supply for the next film night. A particular book you want to have read by the summer is not reference but an intention, and belongs on someday/maybe or in the projects. The test is always the same: do I expect something to happen here.

The one test for everything

When you are unsure while processing where an entry belongs, three questions in this order will do:

Do I have to do something? If no, it is reference or someday/maybe or rubbish.

If yes: am I the one doing it? If no, it belongs on waiting for.

If yes: is one action enough? If no, it becomes a project, and the one action that gets it moving goes onto a context list.

That is the entire logic. Everything else is a tooling question.

Where it typically fails

Too many context lists. Create twelve contexts and you look at none. Four to six is almost always enough.

Lists in the tool, but the phrasings are topics rather than actions. The system then looks tidy and paralyses you anyway.

No weekly review. The lists drift so far from reality within two weeks that you no longer trust them. From that point you are carrying things in your head again.

Forcing everything into one tool. It is entirely fine for reference material to live in Obsidian or Drive and the next-actions lists in a task app. What has to hold together is the logic, not the software.

Common questions

Which lists do I need for GTD?

Calendar, next actions by context, waiting for, projects, agenda lists per person, someday/maybe and reference material. Each answers exactly one question, and that is precisely why they work better apart than together.

What is the difference between the project list and the next-actions list?

The project list holds desired outcomes and serves as a completeness check during the weekly review. The next-actions lists hold concrete actions and serve execution in daily work. Nothing on the project list is something you can do right away.

Do context lists based on places still make sense?

Partly. @Calls, @Out and @Home still work; @Computer is usually too big to help today. The more useful distinction is often the concentration available rather than the place.

How many context lists should I keep?

Four to six. Beyond that, the likelihood of looking at all of them regularly drops, and then you lose exactly the trust you were trying to build.

Count how many of the seven lists you actually keep. For most people waiting for and the agenda lists are missing, and those are precisely the two that change something fastest.

Writing lists down is easy; getting them into your working day is not. That is exactly what a workflow setup is for: at the end your material is in the system, not in a manual. Get in touch if you want someone beside you for it.

"Can you hear me?" is the worst opening line of any meeting

Mon, 18 May 2026 00:00:00 GMT

Next comes "hang on, let me make you a co-host" and then "can you all see my slides?". By that point two minutes are gone and every participant has started reading mail in another window. You never fully get them back.

I have seen this in trainings and in my own sessions often enough to say this: the lost minutes are not the damage. The damage happens somewhere else.

The arithmetic that is everywhere and still explains nothing

Five people times three minutes is fifteen minutes. That is in every article on the subject, and it convinces nobody who has twelve meetings a week. Fifteen minutes are recoverable somewhere.

What actually happens costs three times as much.

The start sets the norm for the whole meeting. A meeting that begins with fumbling stays unstructured. Anyone who spends the first two minutes settling roles that could have been settled beforehand will settle everything else on the fly for the next fifty too.

The participants learn. If nothing relevant ever happens in the first minutes, next time half of them arrive three minutes late. So you start later still. This is a loop that establishes itself in any team within a few weeks and that almost nobody consciously notices.

Attention is at its highest at the start. Burn the most expensive minute of the meeting on an audio problem and you do not get it back. People are in their mail by then, and the content starts against a resistance it did not have before.

The habit does not sit where you look for it

The reflex is to resolve to be better prepared next time. That does not work, because the resolution comes due at ten o'clock and the action would have to happen at five to.

A habit needs a trigger that fires at the right moment. "The meeting starts" is too late. The usable trigger lies before that, and it has to be in the calendar.

Concretely: five minutes before every meeting where you present or facilitate, there is a block. Not as an intention, as an entry. If you have recurring meetings, build the buffer into the series once and you are done. This is the same mechanism as the buffer for follow-up work after an appointment, just in the other direction.

The habit needs nothing more. The rest is procedure.

What happens in those five minutes

Check audio and output device. The most common failure is not the broken microphone but the wrong one. Bluetooth headphones that connected in the meantime, a docking station, a second monitor with its own speakers. The meeting app remembers the device from last time, and that is not always the one currently on your head.

Test screen sharing once, do not just assume the permission. On a Mac the meeting app needs screen recording permission under Privacy in System Settings. After an app update that permission can be reset, and you find out at exactly the moment you want to share. Also check which screen you are sharing if you work with two monitors.

Turn off notifications. Focus mode on a Mac, notification assistant on Windows. Not because of the distraction but because while sharing your screen private messages otherwise pop up for everyone to see. That is the mistake you make exactly once.

Force updates beforehand. Zoom and Teams always want to install their update at the moment you are in a hurry. Start the app once the day before and you do not have the problem.

Have the material open. Presentation loaded, browser tabs sorted, everything private closed. Do not go searching during the meeting.

Roles belong in the invitation, not in the first minutes

Co-host rights can be assigned in advance. In Zoom via the alternative host when you create the meeting, in Teams via the meeting options where you set who appears as presenter and who as attendee. Both take twenty seconds when you set them up and save a minute and a half in the meeting, plus the awkwardness.

Along with three questions that should be answered before the meeting:

Who facilitates. Who records the outcomes. Who decides at the end, if the group cannot agree.

The third point almost never gets settled and is the reason meetings end without a result and get repeated in two weeks.

As host you are in the room two minutes before the start. Then you are the one greeting people rather than the one still searching. As a side effect you notice whether somebody else's technology is playing up and can fix it before it affects everyone.

The opening that keeps the energy

When the clock hits the time, you start on content. No waiting for latecomers, no round of greetings with six people, no "let's give it another two minutes".

Two sentences are enough: why we are here and how we will know we are done. "Today we decide whether variant A or B goes into the proposal. At the end we have the decision and who implements it."

Wait for latecomers and you punish the punctual and reward the late. After three weeks everybody arrives late.

The bigger lever is elsewhere

So you can see the proportions: the zero-minute setup saves you two to three minutes. A meeting without a defined outcome burns sixty.

That is why the goal belongs in the invitation and not in the opening lines. One sentence in the invitation text about what should exist at the end. If that sentence will not come to you, that is the most useful information of the day, because the meeting can then usually be an email.

The clean start is the habit to begin with, because it is small and works immediately. The question of whether the meeting has to happen at all is the more uncomfortable one and the one with the greater return. Both are habits, and both only change if you tackle them one at a time instead of resolving to hold better meetings from now on.

Where it typically fails

The buffer does not get entered. Without a calendar entry the preparation does not happen on a full day, and full days are exactly the ones with the meetings on them.

The buffer gets used for one last task. Five minutes before the meeting you write one quick mail, and you are the person entering the room at 10:01 again.

Only the host prepares. If somebody else is presenting, they need the same run-up. Say so when you put someone on the agenda.

The punctual start happens once and is then abandoned, because half the group was missing the first time. That first time is exactly the point at which the team's habit changes, or does not.

Common questions

How do I avoid technical problems at the start of an online meeting?

Put a five-minute run-up in the calendar as a fixed entry and use it to test audio, output device and screen sharing, switch off notifications and open your material. The test has to happen before you open the room, not inside it.

How do I assign co-host rights in advance?

In Zoom via the alternative host when creating the meeting, in Microsoft Teams via the meeting options where the presenter role is set. Both are part of the invitation and take under a minute.

Should I wait for late participants?

No. Waiting permanently pushes back the effective start time, because everybody adjusts to it. Start on content on time and let latecomers join without repeating anything.

What belongs in the meeting invitation?

The intended outcome in one sentence, the roles for facilitation and note-taking, and who decides in case of doubt. If the outcome cannot be phrased, the meeting is probably unnecessary.

Look at your calendar for next week and count the meetings where you present. The five-minute block belongs right before those. Entering it takes longer than the intention, but it also lasts longer.

If more is jammed in the team than the meeting start, a digital workplace audit looks at the whole picture: which tools do the same thing and where information gets stuck. Write to me.

The most expensive click in Outlook is "mark as unread"

Tue, 12 May 2026 00:00:00 GMT

It costs half a second and you do it forty times a day. Here is what happens: you read the mail, you recognised that something needs doing, so you have done nearly all the thinking. And then you throw the result away and pretend you never saw it.

On the next pass you read the same mail again. And recognise again that something needs doing. And mark it unread again, because right now you have no time to do it.

For a mail that sits in the inbox for three days you do that thinking six or seven times. That is the real reason a full inbox is exhausting. Not the volume, the repetition.

The way out consists of a single principle: you make the decision once and you record it. In Microsoft 365 you can set that up with one Quick Step and three keys so that it is faster than marking things unread.

What you have to switch off first

Before you set anything up, the notifications go. All of them. Sound, desktop pop-up, envelope icon in the tray, badge on the phone.

The reason is not concentration in general but that this workflow only works in blocks. You process the entire inbox two or three times a day, and in between Outlook is closed. Anyone glancing over every eight minutes is not processing, they are reacting, and then the best Quick Step in the world will not help.

In Outlook for Windows this sits under File, Options, Mail, in the section on incoming messages. All four boxes off.

The Quick Step this is about

A Quick Step is a chain of actions on a single click. We need two of them.

Quick Step "Next action"

On the ribbon under Home you will find the Quick Steps area; from there choose Create New and then Custom.

Action: Flag Message, setting "No Date"
Action: Move to Folder, target "Archive"

The important third thing on the same dialog: assign the Quick Step a keyboard shortcut. Outlook offers Ctrl plus Shift plus 1 through 9. Without a shortcut you reach for the mouse, and then the workflow is slower than the thing you were replacing.

Quick Step "Waiting for"

The same sequence again, but with a category instead of the flag. Action one: Categorize Message, category "Waiting for". Action two: Move to Folder, Archive. A keyboard shortcut here too.

The usual guide leaves this second Quick Step out, and that is precisely why everything delegated slips through the net in such setups after two weeks. Everything you handed over and are still accountable for needs a place. Otherwise you remember it only when somebody asks.

The pass: three keys, top to bottom

When you process, you do not work. You only decide. Every mail gets exactly one of three treatments:

Delete for anything that can go. Newsletters, yesterday's meeting confirmations, thank-you notes from colleagues.

Backspace for anything you are merely taking note of. The mail lands in the archive. No folder structure, no sorting. One archive is enough; search finds it more reliably than any folder tree you invent.

Your Quick Step for anything where you have to do something. Mail gets flagged and moves to the archive.

One exception that comes before all three: if the reply can be written in less than two minutes, write it now. But hold yourself to the two minutes. In practice this is where processing blocks die, because two minutes become twelve and one pass becomes a whole morning. If you hesitate over whether it is two minutes: it is not.

The step where almost everyone fails

Flagged mails appear in Microsoft To Do in the "Flagged email" list. You have to switch that list on in the To Do settings first, otherwise you wait in vain. And the link only works with Exchange accounts, meaning Microsoft 365 or Outlook.com. Anyone who has hooked up an IMAP account from another provider does not get this bridge.

Now comes the part that decides between success and failure.

This list is not a task list. It is a second inbox. What sits there is "RE: query about quote week 34", not what you have to do. Anyone who only flags and never goes further has merely moved their unprocessed pile from Outlook to To Do, and gets to ask themselves in three weeks why the new system does not work either.

The second move is mandatory: rename the task into a concrete action, then push it into the right list.

"RE: query about quote week 34" becomes "Call Tobias about the price tiers on project X". Verb, object, clear enough that you can start in three days without thinking. The original mail stays attached to the task and is one click away — that is the real advantage of this coupling.

I do this straight after the processing pass, not later. First empty Outlook, then empty To Do. Both together take about fifteen minutes for fifty mails once you are practised.

What Microsoft To Do can do and where it stops

For next-actions lists it is usable. Separate lists per context, so Calls, Office, Errands, or per area. Hashtags in the task title work as a second sorting layer if lists alone are too coarse for you. The link back to the original mail is where it beats other tools.

Where it is weak: projects. A task with subtasks is not project management, and the project list in the GTD sense is not something To Do represents cleanly. People who work with it usually keep the project list elsewhere and use To Do only for next actions. That is not a flaw, as long as it is a conscious decision and not something you notice when half the list is missing during the weekly review.

"My Day" is optional. It helps people who stall under the weight of options, and it tempts others into building a wish list in the morning that is already wrong by eleven.

Inbox zero is not the goal

An empty inbox is a by-product, not an achievement. Make zero the goal and you start shoving mails aside to push the number down, and then you simply do not make the decisions. The result is an empty inbox and a full head, which is precisely the opposite.

The purpose is a different one: you have looked at every message once, decided what it means, and the result sits in a place you trust. That the inbox is empty afterwards follows by itself.

Where it typically fails

The flagged mails never get renamed. The most common mistake, see above. Two inboxes are worse than one.

No keyboard shortcut on the Quick Step. Sounds like a detail. Decides whether you keep the workflow up at eighty mails.

The two-minute rule gets stretched. The processing block becomes a working block, the inbox stays half full, and the next day it is worse.

Folder structures in the archive. If archiving still requires deciding where, you archive more slowly and still find nothing afterwards. One archive, use search.

The weekly review is missing. The "Waiting for" category is worthless if nobody goes through it once a week and checks what has been open for two weeks.

Common questions

How do I create a Quick Step in Outlook?

Under Home in the Quick Steps area choose Create New and then Custom. Add actions, in this case Flag Message with "No Date" plus Move to Folder into the archive. Assign a keyboard shortcut in the same dialog.

Why do my flagged mails not appear in Microsoft To Do?

The "Flagged email" list has to be enabled in the To Do settings. The link also only works with Exchange accounts, meaning Microsoft 365 or Outlook.com, not with IMAP accounts from other providers.

What is the difference between flagging and a task?

A flagged mail is a reminder that something here is still open. A task is a phrased action. The step from one to the other is the actual work and cannot be automated.

Do I need a folder structure for archived mail?

No. A single archive folder plus full-text search is faster than any folder hierarchy, because filing requires no decision and finding requires no memory.

Does this work in the new Outlook too?

The interface and the feature set differ by version and account. Check before you set up whether your Outlook version offers custom Quick Steps with keyboard shortcuts. The principle stays the same in any case: one action that flags and archives.

Set up the Quick Step and take it through your entire inbox once. The interesting moment comes afterwards, when you see in To Do how many subject lines are sitting there that nobody has turned into an action yet.

Outlook, Google Workspace or Apple — the logic is the same, the clicks are not. A workflow setup builds it in your environment. Tell me what you work with.

97 projects. And the sentence "I have no idea where to start."

Sun, 03 May 2026 00:00:00 GMT

The number comes from a conversation in spring. The client had opened his project list, scrolled a bit, and then said that sentence I hear in some form almost every week. He took the 97 for the diagnosis.

It is not. David Allen puts thirty to a hundred projects as the normal range for somebody with a full working life and a family. Anyone carrying responsibility at work and owning more than a houseplant at home gets there quickly. A short project list is no sign of clarity; usually it is a sign of incomplete capture.

The problem was a different one, and it is right there in the sentence. He opened the project list in order to decide what to do now.

Nothing on the project list is something you can do

A project is an outcome, not an action. "Launch the new product" is not something you can start at half past ten. If you look at that list in the morning and feel unable to act, that is not overwhelm, that is the correct response to a list built for another purpose.

The project list is a completeness check. It answers the question of whether I have everything in view that I have committed to. You ask that question once a week, in the weekly review.

The question "what do I do now" is answered by the next-actions lists. Those are sorted by context, hold concrete actions, and you can open them at half past ten and get going.

Mix those two lists, or use one for the other, and you get exactly the feeling described above. And no amount of re-sorting fixes it while the second list is missing.

What is actually on a list of 97

Before you sort, clear out. In practice a list like that consists in good part of things that have no business being there.

Areas of responsibility in disguise. "Bookkeeping", "Marketing", "Health". Those are not projects, because they are never finished. The test is simple: can you imagine ticking it off? For "Bookkeeping" never; for "Hand the 2025 annual accounts to the tax adviser" yes. Entries like that are particularly paralysing, because the brain senses there is no end to them.

Someday/maybe. The language course, the book project, the garage. Things where realistically nothing will happen in the coming weeks. They do not belong on the active list but on one of their own, which you skim during the weekly review. For most clients a third moves off here, and that is the single biggest effect of the whole clean-up.

Projects without a next action. They sit there, but nothing moves. They are the real reason for the bad conscience when you look at the list, because your brain notices at every scroll that a decision is open here. Every active project needs exactly one defined next action. If none comes to mind for a project, the next action is often "think through how this is even supposed to work", and that is a legitimate entry.

Of 97, experience says thirty to fifty active projects remain after this pass. Still a lot. But honest.

Areas of responsibility: what they can do, what they cannot

Only now is grouping worth it. Areas of responsibility are the level above projects, the areas that need ongoing attention and are never completed. For most people there are between ten and twenty, work and private life combined.

The benefit is real. When I open the Finance area on a Friday and four projects sit there, that is a quantity my head can handle. And looking at the areas shows immediately where nothing has happened for weeks. If there has not been a single project under "Health" for four months, that is no longer a sorting question, that is information.

There is, however, a methodical objection, and it is a fair one. A categorised project list tempts you into reviewing only the areas that are currently pressing. But the point of that pass is that once a week you see every open commitment, including the one you do not want to think about. Those are exactly the ones you reliably avoid when they are neatly folded away.

So my recommendation: group, yes, but the weekly review goes across every area. No exceptions, including the boring ones. Unfold everything, walk through it, fold it back. If you cannot keep that up, you will end up with more overview from the flat list than from the beautiful structure.

Doing it in practice

Trello: one board for projects, one list per area of responsibility, one card per project. Next actions do not belong in the card as a checklist but on a separate board or in your task app, otherwise you have to click through the project structure again while in working mode.

Obsidian or another markdown-based system: the PARA structure with separate folders for projects and areas works well, as long as every project note links to its area. The area then gets an overview of its projects by itself, through the backlinks. The advantage over folders: a project can belong to two areas without you having to choose.

Whatever the tool: the area structure must not slow down creating a project. If capturing means first working out where the card belongs, you will capture less. Better everything into one inbox and the assignment during the weekly review.

The thing about paper

Thinking a project through is not the same as storing it. For the thinking, paper is often better than any tool, because nothing about the structure is given. No fields, no hierarchy, no decision about which project the thought belongs to.

For larger things I do this in a café with a sheet of A4 turned sideways. When the plan holds, the results move into the system: the next actions onto the lists, the sheet photographed as reference material for the project. The photo is not decoration, it is the state of my thinking at that moment, and when I come back in three weeks I am inside it again in two minutes.

Where it typically fails

The structure gets built, the next-actions lists are still missing. The project list then looks lovely and the paralysis stays, because the underlying problem was never addressed.

Too many areas. Create thirty areas of responsibility and you have only pushed the clutter up one level. Ten to twenty is almost always enough.

Someday/maybe never gets reviewed. A list you never look at is a graveyard. Skim it once a week and it works as a valve.

Re-sorting replaces working. A tidy system feels like progress. It only is progress if somebody then does the next actions.

Common questions

How many projects are normal?

Thirty to a hundred open at once is usual with a full working and private life. What matters is not the count but that every active project has a defined next action and that the list is reviewed completely once a week.

What is the difference between a project and an area of responsibility?

A project has an outcome and therefore an end; you can tick it off. An area of responsibility runs continuously and is never finished. "Hand over the annual accounts" is a project, "Bookkeeping" is an area.

Why does my project list paralyse me?

Usually because it is being used for the wrong purpose. The project list answers whether you have everything in view, not what you should do now. The question about the next action is answered by context-sorted next-actions lists.

Should I group my project list by area?

Grouping helps with the overview and reveals neglected areas. It does carry the risk that uncomfortable areas stay folded shut during the weekly review. Group, yes, but open every area during the review.

Open your project list and mark everything where no next action comes to mind within five seconds. Those marks are your actual work for the next weekly review.

If you look away every time you open the list, it is rarely the list and usually a decision that is pending. Sparring exists for exactly that: one fixed appointment a month for the things that stay stuck. Write to me.

The fullest calendar in the room rarely belongs to the most productive person

Wed, 22 Apr 2026 00:00:00 GMT

The first thing I ask to see is always the calendar. Not the lists, the calendar. It tells me more about the state of a system in thirty seconds than half an hour of conversation.

What I often find: thirty, forty entries in one week, of which maybe eight are real appointments. The rest are tasks somebody squeezed into a time slot at some point. "Tuesday 9:00 to 9:30, prepare presentation." And when I ask what actually happened on Tuesday at nine, the answer is almost always: no idea, something came up.

That is not a lack of discipline. That is a tool built for something else.

Why we do it anyway

For most people the calendar is the only tool they genuinely trust. It reminds them by itself, it is on every device, it does not lie about appointments. So when the feeling of losing track creeps in, you reach for the thing that has always worked.

Only something happens in the process that you notice a few weeks later. The calendar knows nothing about backlog.

A task on a list stays there until you tick it off. It can turn a week old and still look at you during the weekly review. A calendar entry from yesterday disappears. Nobody reads yesterday's calendar. Whatever did not get done in a past time slot is not open — it is gone.

That is why the calendar as a task system is not merely impractical, it is unreliable in a way you do not notice. Things drop out quietly.

The price: you start distrusting your own calendar

The second effect is worse. When thirty out of forty entries are movable, every reminder becomes noise. You hit snooze, because experience tells you it is usually nothing hard. Two months later you hit snooze on the dentist too.

At that point you have devalued the one tool that used to be reliable. And then it gets genuinely expensive, because from there you no longer have any system you trust.

The hard landscape: three things, nothing else

The calendar takes only what has to happen on a particular day or at a particular time. Everything else is a task and belongs on a list.

Time-specific. The meeting on Tuesday at two. The train. The dentist. Test: if you turn up at four, it is over or you have a problem.

Day-specific. Something that becomes relevant on a particular day but has no time attached. The hotel address for Thursday. The note that the cancellation deadline expires on Friday. The reminder that your passport runs out in six months, entered on the right day next year.

This is the underrated category. Used as an all-day entry, it is a digital tickler file that needs no extra tool. It lets me park information on exactly the day I need it, rather than in a folder I never open again.

Blocks for focused work. Time reserved for you, with no content fixed.

The third one needs explaining, because at first glance it looks like a contradiction.

Time blocking, without setting the trap again

Time blocking has been popular for years, and in the strict reading of the hard landscape it has no business there. The difference lies in what is written on the block.

"Tuesday 9:00, prepare the presentation for the boss" is a task assignment. At 9:05 the emergency arrives, the block is dead, the task is gone.

"Tuesday 9:00 to 11:00, focus time" is a reservation. It protects two hours against the rest of the world but does not say what happens inside them. You decide that at nine, from your lists and from what has happened since. When the emergency arrives, you do not lose the task, only the block. The task stays on the list where it belongs.

The block reserves capacity. The list decides the content. The moment the block fixes the content, you have set the trap again, just with better vocabulary.

One practical side effect in companies with shared calendars: blocked time is visibly taken. An empty calendar is an invitation.

Doing it in Google Calendar

What has worked for me and for clients:

Separate calendars for appointments and for blocks, in different colours. You can then hide the blocks with one click when you want to see how full the day really is.
Notifications only for time-specific appointments. Blocks and all-day entries need no alarm, you see them in the daily overview. That is the lever against the snooze syndrome.
Day-specific things as all-day entries, so they sit in their own row at the top and do not get lost between meetings.
Name blocks generically. Mine are simply called Focus. Put a project name there and I start scheduling tasks again.
Enter recurring blocks, do not create them daily. A block you have to set every morning will not survive a full week.

What happens to the rest

Everything that needs no date moves onto context lists. That is the real work of switching over, and it is why the switch often fails: throw tasks out of the calendar and you need a list system you trust just as much. Without that, the empty calendar feels like losing control, and after ten days the tasks are back in it.

One deadline special case that always comes up: the submission on the 15th belongs in the calendar as day-specific information, because the date is hard. The work on it belongs on the project list with a defined next action. Enter only the deadline and on the 15th at nine in the morning you will discover you needed three days.

Where it typically fails

The blocks turn into a secret task list. After two weeks "Focus" has become "Focus: Miller quote". That is exactly where it started. Check for it now and then in the weekly review.

Colleagues book over the blocks. It happens, but it is rarely a permissions problem and usually a conversation problem. Say once that these two hours are fixed and you will mostly be respected.

Too many all-day entries. When five rows sit at the top, nobody looks any more. The day-specific category only works as long as it looks empty.

The calendar empties out and stays empty. If the clean-up leaves eight appointments a week and no blocks, the free space will be filled by other people within two weeks.

Common questions

Which three things belong in the calendar according to GTD?

Time-specific appointments, day-specific tasks and information, and reserved blocks of time with no content fixed. Everything else belongs on context lists.

Why should I not schedule tasks in my calendar?

Because the calendar knows nothing about backlog. Whatever did not get done in a past time slot disappears unnoticed instead of staying open. On top of that, every movable entry devalues the calendar's reminders as a whole.

Does time blocking contradict the hard landscape?

Only if the block fixes a particular task. A block that reserves time and leaves the content open is compatible with GTD. What happens in that time you decide when the block starts, from your lists.

Where do deadlines go?

The due date as day-specific information in the calendar, the work on it as a next action on the project list. The deadline alone tells you nothing about when you have to start.

Open your calendar and count the entries from last week that you could have moved. If it is more than half, you know why you swipe away your own reminders.

An overgrown calendar is hard to clear alone, because every decision hangs on a commitment you made. That is what a workflow setup and sparring are for. Write to me if you want to go through it.

The client had a perfect system. And still no day of his own.

Mon, 13 Apr 2026 00:00:00 GMT

I regularly sit across from someone whose method is flawless. Project list maintained, contexts clean, weekly review every Friday. And still the same sentence: "I just never get to my own things."

So we look at an ordinary Tuesday together. Almost always it goes like this: computer on, mail client open, and from there it is one after the next. One mail, one reply, the next mail. Something pops in that looks urgent, so deal with it now. By half past eleven the morning is gone and the client has not opened his own project list once that day.

The system is not the problem. On that day the system simply is not used.

Processing and working are two different modes

When you process, you decide. What is this, do I have to do something, and if so what exactly is the next physical action. That is thinking work at a rapid pace, many small decisions one after another.

When you work, you execute. One thing, a longer stretch, concentration.

These two modes demand completely different things of your head. Mix them and you do both badly. You decide sloppily, because you keep tipping over into execution, and you work without focus, because the next mail is waiting in the background. That is why a morning in the inbox feels like work and, by the evening, feels like nothing.

What David Allen means by Clarify is not "get through your mail". It is "decide what this stuff means". Doing comes later, in a different block, off a list.

Separate the two and you are not working harder. You have just stopped switching modes twenty times a day.

Why the inbox does not work as a task list

The inbox is sorted by arrival time. Your priorities are not.

On top of that: a mail never shows you the next action, it shows you a topic. Between "Quote for Miller" and what actually needs doing lies a decision that you make afresh every time you look at it. With one mail that does not matter. With eighty unread you make that decision eighty times without ever recording it. That is what makes you tired by the evening.

And the most expensive effect: whatever sits in the inbox appears on none of your lists. You now have two competing systems, and from that moment your head trusts neither.

The daily review: ten minutes before the day owns you

The weekly review stays the centrepiece, nothing changes there. The daily look is something else: not system maintenance but taking your bearings. It answers a single question. What does today realistically look like.

The calendar first. What are the hard edges, meaning the appointments I cannot move. How much free time is actually left. For most people that is the moment of sobering up: there are not six hours between the meetings, there are two.

Then the next-actions lists. Not all of them, only the contexts that come up today. What fits into those two hours.

Then the waiting-for list. The one place where something tips over daily. Something open for ten days becomes a phone call today.

Only after that, the inbox.

That takes ten minutes if your system is in order. If it takes longer, you do not have a daily-review problem, you have a weekly-review problem.

On the "five priorities for today"

A short daily list contradicts the pure doctrine. GTD has no daily to-do list, because it forces a decision in advance that can already be wrong by eleven.

It still works for a lot of people, and I recommend it when I see that someone stalls every morning under the weight of options. The condition is that the list comes off the context list and does not become a parallel system. And that three items are enough when the day only offers two free hours. Five items on a day with six meetings is not prioritising, it is self-deception with the character of a calendar.

What actually happens when you process

Two fixed blocks, for me late morning and early afternoon. In these blocks nothing gets worked on, things get sorted:

Rubbish goes out
Pure information gets archived, without folder archaeology, search will find it again
If something needs doing: the next action gets phrased, goes onto the right list, and the mail moves to the archive
If somebody else is up: out it goes, and onto waiting-for

The step where this fails is always the same. The next action does not get phrased, it only gets vaguely thought. "Quote for Miller" is not a next action. "Pull Miller's costing out of the sheet and transfer it into the template" is. The difference costs twenty seconds when processing and saves five minutes of run-up when working. At fifteen mails a day, that is the whole effect.

The two-minute rule is the biggest trap in mail processing. Two minutes means two minutes, not "that will be quick". I constantly see people write three replies inside a processing block, twelve minutes each, and then the block is over and the inbox is still half full. If you are unsure about a mail: it will take longer than two minutes.

When the inbox is done, it gets closed. Not minimised. Closed.

Expectations are negotiable, you just never negotiate

Almost everyone who believes they must reply within minutes has never discussed it with anybody. They trained themselves into it and then assume everyone else expects it.

The announcement is unspectacular: "I check mail twice a day. If it's urgent, call me." One line in the signature, said once in the team. In twenty years nobody has ever held it against me. What bothers people is not an answer after six hours, it is no answer and no idea where they stand.

One side effect that rarely gets mentioned: as soon as the phone is the escalation route, the number of urgent things drops noticeably. Calling costs the sender something. Marking a mail as high priority costs nothing.

Where it typically fails

The blocks are not in the calendar. A processing block that is only an intention does not happen on a full day.

The notifications keep running. If you set aside time twice a day to process and then glance over at every ping in between, you have gained nothing. The block replaces the interruption, it does not come on top of it.

The weekly review gets skipped and the daily review is supposed to cover for it. That goes wrong. The daily look only works when the lists are right, and Friday is what keeps them right.

The first block sits too early. Put your best hours of concentration into processing and you have the same order as before, just with a better name. For me the first real working session comes before the first mail block. Not possible for everyone, but check it honestly before you rule it out.

Common questions

What is the difference between processing email and working email?

When processing, you only decide what a message means and what the next action would be, and you record that on a list. When working it, you execute the tasks straight from the inbox. The first mode takes seconds per mail, the second consumes your morning.

How long should a daily review take?

Ten minutes with a maintained system. It covers the calendar, the next-actions lists relevant today, and the waiting-for list. If it regularly takes longer, the weekly review is missing.

How often should I check email?

Two fixed blocks a day are enough in most roles. What matters is not the number but that no notifications run between the blocks and the inbox stays closed.

Does the daily review replace the weekly review?

No. The daily look orients you inside a system that the weekly review keeps current. Without Friday you are looking at stale lists every day.

Where does your first glance land in the morning, the calendar or the inbox? And if it is the inbox: what would have to happen for that to be different next week?

If the inbox has been the same fight for years: in a workflow setup we build this with your real mail, not with an example. Drop me a line if you want to know whether it fits your situation.

I have not filed an invoice myself in a year

Sat, 04 Apr 2026 00:00:00 GMT

The trigger was a weekly review last winter. I went looking for an invoice from the month before, could not find it, and discovered while searching that my Downloads folder held over a thousand files. Invoices, quotes, screenshots, three versions of the same PDF. The review I had blocked ninety minutes for turned into a clean-up session.

This is not a time problem. Two minutes per invoice are two minutes. The problem is what those two minutes do to the system I am supposed to trust.

Why digital clutter is a GTD problem, not an efficiency problem

Any task I do by hand is a task I can also forget. And any filing habit I sometimes let slide becomes a filing system I no longer trust. If I am not sure whether the invoice ended up in the right folder, next time I search in three places instead of one. That costs more than time, it costs trust. And a system I do not trust is one I have to carry in my head — exactly what GTD is supposed to end.

The two-minute rule works for one invoice. For forty a month it is a bad idea. Forty times two minutes is over an hour, and more importantly it is forty interruptions. Filing is not the work. Filing is the condition under which the work stays findable.

Tasks like that do not belong on a list. They belong gone.

What changed: the programmer now sits in your browser

I have been building software for over thirty years, professionally since 1998. I can write scripts. I still did not do it for years, because the maths did not work: two evenings of work for a script that saves me ten minutes a week pays for itself somewhere inside a year in which I no longer need the script.

That calculation no longer holds. I describe to an AI what should happen, it writes me the code for Google Apps Script, I test it and adjust two small things. Two evenings become forty minutes. At forty minutes, almost anything is worth it.

If you have never written a line of code you can get there too, but you will need patience while testing. That is the honest part most articles on this leave out. More on it below.

Six automations running here

1. Invoice filing with no clicks at all

Trigger: an email carrying the label Invoice, or a PDF landing in a particular cloud folder.

What happens: the script pulls the attachment, sends it to an AI model for extraction, gets back supplier, invoice date, number and gross amount, renames the file to the pattern 2026-05-15_Supplier_InvoiceNumber.pdf and files it under year and month. In parallel it appends a row to a Google Sheet.

Result: a filing system with consistent names that I can actually search, plus a summary table I can export for my accountant. The effort is zero, as long as nothing breaks.

Extraction is the part that did not work before. OCR on its own gives you text mush. A model that understands which of the four numbers on the document is the gross total turns that mush into structured data. That is the actual leap.

2. The project folder appears when the card does

Trigger: a new card in a particular Trello list.

What happens: the automation creates a Drive folder with the same name and writes the link back onto the card.

Result: I never again have to decide where the files for this project go. Sounds trivial. It is, however, the decision I used to postpone every single time, which is why project files ended up in Downloads. See above.

3. The weekly review gets its data served

This is the automation that helps me most in methodical terms.

Trigger: Friday morning, on a schedule.

What happens: a script collects every Trello card that moved to Done this week, every calendar entry from the past two weeks and the next two, every project without a next action, and every waiting-for entry older than ten days. All of it lands as a single document in my vault.

Result: I start the review by thinking rather than by gathering. The thirty minutes that used to go into scraping the facts together now go into the question of whether these are even the right projects.

If your review keeps slipping, start here. The most common reason reviews get skipped is not a lack of time, it is the hurdle of the first fifteen minutes.

4. A waiting-for list that keeps itself

Trigger: I apply the label Waiting for when I send a mail.

What happens: once a day a script scans every thread marked that way, checks whether a reply came in, and maintains a list with recipient, subject and the number of days since sending. Threads that got an answer disappear on their own.

Result: the list is always current without me tending it. And I can see at a glance what has gone unanswered for three weeks. That is the point at which a waiting-for turns into a phone call.

5. Meeting notes to next actions

Trigger: a document in a folder called Meeting notes.

What happens: a model reads the note and pulls out everything that sounds like a commitment, a task or a date, each with an owner. The result lands as a list in my inbox, not straight onto a project list.

That last point matters. I do not let the AI decide what a next action is. It delivers raw material into the inbox; the clarifying is mine. Hand that step to the machine and within two weeks you have lists full of things you never agreed to.

6. Downloads tidies itself

Trigger: daily at six in the morning.

What happens: anything older than seven days moves into a folder Archive/Year-Month. Images, installers and zip files older than thirty days are deleted.

Result: the folder is what it is meant to be again, a place things pass through rather than a warehouse. This is the simplest automation of the lot and it cost me the least. Ten minutes.

Where it goes wrong

Four things I learned, some of them the hard way.

Automated mess is still mess. If your folder structure did not work before, a script just breaks it faster. Sort out the process first, then automate. That is the same mistake as buying a new tool instead of repairing your own system.

The code runs until it doesn't. AI-generated scripts typically have no proper error handling. When Google has a moment or a PDF is corrupt, the thing fails silently. I have it email me on every error. That is the one line of code you should not skip.

Watch out for idempotence. A script that runs over the same data twice must not create the same folder three times. Ask the AI for it explicitly; it rarely offers on its own.

Not everything should be automated. Clarifying and deciding are the heart of the matter, and no machine can take those off me. What can be automated is transport: moving files, renaming them, gathering data, keeping lists current. The moment the question becomes what something means to me and what the next action is, I am up.

How to start

Take the most annoying recurring task of the past week. Not the biggest — the most annoying. Then describe it to an AI like this:

Write me a Google Apps Script. Every day at 6 a.m. it should move every file in my Drive folder "Downloads" that is older than 7 days into a subfolder "Archive/YYYY-MM", matching the file's modification date. Create the target folder if it does not exist. If an error occurs, email me the error message. Explain step by step where I paste the script and how I set up the trigger.

That last sentence is the important one if you have never worked with Apps Script. Test with a handful of copies first, not with your real data. And never let a script delete anything permanently until you have watched it for two weeks.

My first automation was the Downloads one. Not the most elegant, but the one that made me want to carry on.

Common questions

Do I need programming skills to automate routine tasks?

No. For scripts in Google Apps Script, or automations between Trello and Drive, describing the desired sequence precisely is enough. What you need is patience while testing and a willingness to feed error messages back to the AI instead of giving up.

Which tasks are worth automating?

Anything that comes up regularly, follows clear rules and requires no judgement: renaming and filing documents, creating folders, extracting data from documents, keeping lists current, gathering information for the weekly review. What is not suitable is anything that requires a decision about meaning and priority.

How much time does this actually save?

For me it is around three hours a month on invoice filing and about two hours on preparing the review. The bigger effect lies elsewhere: the filing is reliable, and I no longer have to think about it.

What is the most common mistake when automating?

Automating a broken process. If your filing structure did not work before, a script only speeds up the chaos. Settle the structure first, then automate.

If the same thing happens by hand every day at your place: automation and AI is exactly that — analysis and implementation in production, not a prototype. Tell me what your recurring handwork is.