Getting Things Done in 2026: what still works and what to drop

Getting Things Done (GTD) is David Allen's 2001 system for turning everything that has your attention into a trusted set of lists, so that your head stops doing the job of a filing cabinet ^[1]. It predates the smartphone and still sells books and app templates. Most pages about it summarise the five steps. This one asks which parts the research on memory and planning supports, and which parts fail in a week dominated by chat, meetings and AI drafts. The answer: the capture habit and the weekly review rest on reasonable evidence, the system as a whole has never been tested in a trial, and its weak points, priority and energy, are exactly where the Eisenhower matrix, time blocking and our guide to how to prioritise tasks take over.

The five steps

Step	What you do	The rule that makes it work	Where people stall
1. Capture	Write down every commitment, idea and "should" in as few inboxes as possible: a notebook, a notes app, an email folder	Nothing is held in your head; the inbox is for collection, not for sorting	Six inboxes instead of one; capturing without ever emptying
2. Clarify	For each item ask: is it actionable? If not, bin it, file it as reference or park it on a someday list. If yes, decide the very next physical action	The two-minute rule: if the next action takes under two minutes, do it now	Items that are projects in disguise ("sort out the website") never get a next action
3. Organise	Put next actions on context lists (at computer, calls, errands), multi-step outcomes on a projects list, dated items on the calendar, delegated items on a waiting-for list	The calendar holds only things that must happen at that time; everything else is a list	Over-engineered contexts; a projects list nobody reads
4. Reflect	A daily glance at the calendar and action lists; a weekly review that empties inboxes, checks every project for a next action and looks ahead two weeks	The weekly review is the step Allen calls non-negotiable	Skipped in the first busy week and never restarted
5. Engage	Choose what to do by context, time available, energy and, last, priority	You trust the lists, so you can act without re-deciding everything	Energy and priority are judged in the moment with no help from the system

Allen's phrase for the goal is "mind like water": a state in which nothing nags, because every open loop has a place and a next step ^[1]. That claim is psychological, and it can be checked against psychology.

What research says about capture and offloading

GTD's central bet is that unfinished commitments held in memory cost attention, and that writing them down with a plan removes the cost. There are no trials of GTD itself in our evidence database; what exists is a theoretical analysis of the method and a set of lab studies on its mechanism.

Finding	Evidence	What it means for GTD	Grade
Unfulfilled goals intrude on thought; making a specific plan removes the intrusion	In lab experiments, participants who wrote a concrete plan for an unfinished goal showed none of the distraction and intrusive thinking that the unplanned group showed, even though the goal remained unfinished [2]	This is the closest thing to a test of "capture and clarify". The relief comes from the plan (a next action), not from the list alone, which is why capture without clarifying does not calm anyone	B
Cognitive offloading is adaptive but has costs	A 2016 review found that using external tools to hold information is driven by sensible judgements about one's own memory and improves performance on the task at hand, while reducing unaided memory for the offloaded material [3]	Externalising your commitments is a reasonable strategy. You also become dependent on the lists; a system you stop trusting is worse than no system	A
Saving one thing can help you remember the next	Saving a file before studying a second one improved memory for the second file in a small lab study [4]	A candidate mechanism for why an empty inbox feels like clearer thinking; one study, one task, so suggestive only	C
GTD as distributed cognition	A 2008 theoretical paper argued that GTD works by moving memory and planning out of the head into the environment, in line with research on situated and distributed cognition; it presents no data [5]	The only academic treatment of the whole system. It explains why the method could work; it does not show that it does	C
Time management as a behaviour set	Moderately related to performance and more strongly to wellbeing in a 2021 meta-analysis [6]; the benefit runs through feeling in control of one's time rather than through any particular technique [7]	Lower strain is plausible and consistent with the literature; a productivity jump is not something the evidence promises	A

In short: the mechanism behind capture and clarify is supported, the system has never been tested as a package, and the benefit to expect is lower background strain rather than more output.

The weekly review problem

Allen is right that the review is the keystone, and the research on plans explains why: an action list only stops the nagging while you believe it is complete and current ^[2]. The review keeps that belief justified. It is also the step most people drop. Three findings help.

• Tie it to a cue, not an intention. If-then plans ("on Friday at 15:00 I open the review checklist") have a medium-to-large effect on follow-through across 94 studies ^[8]. A review that depends on finding time loses to whatever is urgent.
• Give it a definition of done. Specific goals outperform vague ones ^[9]; "inbox to zero, every project has a next action, two weeks of calendar checked" is a goal, "do my review" is not.
• Expect months before it is automatic. New daily behaviours reached automaticity after a median of about 66 days in one study, with a range of 18 to 254 days ^[10]. A weekly behaviour gets far fewer repetitions, so a review that still feels effortful after two months is normal.

Where GTD fails

GTD is a system for completeness. It is weak at two things a working week demands: choosing and pacing.

Prioritisation comes last and arrives unaided. Allen's four criteria for choosing an action put priority after context, time and energy, and the system gives no method for judging it. The gap matters because urgency distorts the judgement: in five experiments, people picked urgent tasks with smaller payoffs over important tasks with larger payoffs even when they had time for both, most strongly when they felt busiest B^[11]. A next-actions list sorted by context offers no defence; the Eisenhower matrix or a weekly big three has to sit above it.

Energy is a criterion with no data behind it. GTD asks you to match tasks to your energy, but the lists do not know what your energy is. Mental fatigue is better understood as a signal about the cost and benefit of continuing than as an empty tank B^[12], and a system that only ever adds items lowers the perceived benefit of each one.

Projects have no durations. GTD tracks outcomes and next actions, not how long things take, and people underestimate task duration even when they know their previous estimates were optimistic B^[13]. A projects list with forty entries and no estimates cannot be finished, and it says nothing about which three deserve the week.

Maintenance is a cost. Clarifying everything, every day, is work. For people with few inputs and one long project, the overhead exceeds the benefit; for people with many inputs from many channels, it pays. The method was designed for the second group.

What AI changes

Capture has become almost free: voice notes transcribe themselves, meeting assistants list action items, and a language model can turn "sort out the website" into five plausible next actions in seconds. Two of the five steps are easier than in 2001.

The other three are not. Deciding whether an item is actionable, whether it belongs on your list at all and what outcome you want remains a judgement, and the offloading research suggests that the more routinely you hand a judgement to a tool, the less practised it gets ^[3]. A 2025 cross-sectional study reported that heavier AI-tool use was associated with lower critical-thinking scores, with offloading as the proposed pathway; the design cannot show cause, so treat it as a caution C^[14]. The practical rule: let tools capture and draft, and keep clarifying and reviewing as the human steps, because that is where the benefit lives.

GTD in Todoist, TickTick, Things, Notion and Obsidian

GTD does not need software; Allen's original setup used paper folders. Software helps with capture from many devices and filtering by context; it also makes it easy to build a cathedral nobody maintains. Prices are from our catalogue, checked 2026-11-09, in US dollars before local tax.

Tool	Capture	Clarify and organise	Review support	Price	Fit for GTD
Todoist	Quick add with natural-language dates on every platform, email forwarding	Projects, labels as contexts, filters, sub-tasks	Saved filter views; no built-in review checklist	Free plan; Pro from $5 per month billed annually	The most direct fit: fast capture, labels as contexts, little temptation to over-build
TickTick	Quick add, voice input, email to task	Lists, tags, built-in calendar, Pomodoro timer	Calendar view for the look-ahead	Free plan; Premium $4.17 per month billed annually ($49.99 a year)	Good for people who want tasks and calendar in one place
Things (Cultured Code)	Quick entry on Apple devices, mail to Things	Areas, projects, headings, tags, Today and Anytime views	A clean Someday list; the design follows GTD closely	One-off licence per Apple platform (not checked for this page)	The most faithful GTD design; Apple only, no affiliate programme
Notion	Web clipper, mobile quick add, AI meeting notes	Databases with any property you want, including energy and context	Templates can hold a review checklist beside the lists	Free plan; Plus from $10 per month billed annually	Flexible to the point of risk; works when you copy one template and stop
Obsidian	Daily notes, mobile capture, plugins	Tasks plugin, tags, linked project notes	Templates for daily and weekly review in plain text	Free for personal use; Sync $4 per month billed annually	Local-first and private; best when your notes and your tasks belong together. No affiliate programme; recommended on merit

Whichever you choose, the setup should take an hour, not a weekend, and the first weekly review should happen before you customise anything.

GTD-lite for 2026

A version that keeps what the evidence supports.

1. One inbox. Everything goes to one place: a notes app, a Todoist inbox, a paper pad. Email stays in email.
2. A ten-minute daily clarify. Empty the inbox once a day. Every actionable item gets a next action written as an if-then: "when I am at my desk tomorrow at 9:00, I draft the first paragraph" ^[8].
3. Two lists, not seven. A projects list with an estimated size for each (small, medium, large) and one next-actions list. Add contexts only if you genuinely work in different places.
4. A priority layer above the lists. Each week, choose three outcomes. Each day, choose one highlight task. The lists feed the choice; they do not make it. Write next to each outcome why it matters, which weakens the pull of urgency ^[11].
5. A calendar slot for the big three. Put the highlight into a blocked morning slot; a next action without a time loses to whatever is urgent that day. The time blocking guide covers the mechanics.
6. A twenty-minute weekly review at a fixed time. Inboxes to zero, every project has a next action, two weeks of calendar checked, estimates compared with what happened.
7. Tools capture and draft, you decide. Let transcription and assistants fill the inbox; keep clarifying and reviewing as manual steps.

Demand note: who searches for this

In the United States "getting things done" draws about 9,900 searches a month with a keyword difficulty of 48, down 33 percent year on year and declining about 4 percent a year over three years (DataForSEO, retrieved 2026-11-09) ^[15]. The term is used untranslated across Europe: 2,900 monthly searches in Germany, 1,900 in the UK, 880 each in France and Spain and 210 in Czechia. German readers get a localised version of this page with local sources and tools rather than a translation.

What to do next

Decide the shape of the system before the software. The OS Builder asks about your role, inputs, meeting load and obstacles and returns a suggested priority system (a GTD-style capture, clarify and review loop is one of four it can recommend), a capture-to-review architecture mapped onto the tools you already use, a weekly rhythm with a review slot, and a four-week checklist that starts with the daily ritual rather than the app.

Frequently asked questions

What are the five steps of Getting Things Done?

Capture everything that has your attention into an inbox; clarify each item into a next action, reference, someday item or bin; organise actions onto context lists, projects onto a projects list and dated items onto the calendar; reflect with a daily look and a weekly review; and engage by choosing actions by context, time, energy and priority ^[1].

Is GTD backed by science?

Partly. The mechanism behind capture and clarify is supported: writing a concrete plan for an unfinished goal removed the intrusive thoughts it otherwise caused in lab experiments ^[2], and offloading memory to external tools is an adaptive strategy with known costs ^[3]. A 2008 paper analysed GTD as distributed cognition but presented no data ^[5]. No trial has tested the full system against a control group.

Why does the weekly review matter so much?

Because a list only relieves your mind while you believe it is complete and current ^[2]. The review restores that belief. To keep it, tie it to a fixed time and cue rather than an intention ^[8], give it a definition of done, and expect it to take months before it feels automatic ^[10].

What is the biggest weakness of GTD?

Prioritisation. GTD sorts work by context and asks you to judge priority last and unaided, while experiments show people choose urgent low-value tasks over important high-value ones even when there is time for both ^[11]. Pair the lists with a weekly big three or the Eisenhower matrix. The second weakness is that projects carry no durations, so the list cannot tell you what fits in a week.

Which app is best for GTD?

Todoist and Things follow the method most directly; TickTick adds a calendar and a Pomodoro timer; Notion and Obsidian let you build anything, which is a risk as much as a feature. Prices in the table above were checked on 2026-11-09. The setup should take an hour, and the first weekly review matters more than the choice of app.

Sources and evidence

1. Getting Things Done: The Art of Stress-Free Productivity — Allen, D. (2001), book C (retrieved 2026-11-09)
2. Consider it done! Plan making can eliminate the cognitive effects of unfulfilled goals — Masicampo, E. J.; Baumeister, R. F. (2011), RCT B doi.org/10.1037/a0024192 (retrieved 2026-11-09)
3. Cognitive Offloading — Risko, E. F.; Gilbert, S. J. (2016), review A doi.org/10.1016/j.tics.2016.07.002 (retrieved 2026-11-09)
4. Saving-Enhanced Memory — Storm, B. C.; Stone, S. M. (2015), RCT C doi.org/10.1177/0956797614559285 (retrieved 2026-11-09)
5. Getting Things Done: The Science behind Stress-Free Productivity — Heylighen, F.; Vidal, C. (2008), review C doi.org/10.1016/j.lrp.2008.09.004 (retrieved 2026-11-09)
6. Does time management work? A meta-analysis — Aeon, B.; Faber, A.; Panaccio, A. (2021), meta-analysis A doi.org/10.1371/journal.pone.0245066 (retrieved 2026-11-09)
7. Time management: Test of a process model — Macan, T. H. (1994), observational B doi.org/10.1037/0021-9010.79.3.381 (retrieved 2026-11-09)
8. Implementation Intentions and Goal Achievement: A Meta-analysis of Effects and Processes — Gollwitzer, P. M.; Sheeran, P. (2006), meta-analysis A doi.org/10.1016/S0065-2601(06)38002-1 (retrieved 2026-11-09)
9. Building a practically useful theory of goal setting and task motivation: A 35-year odyssey — Locke, E. A.; Latham, G. P. (2002), review A doi.org/10.1037/0003-066X.57.9.705 (retrieved 2026-11-09)
10. How are habits formed: Modelling habit formation in the real world — Lally, P.; van Jaarsveld, C. H. M.; Potts, H. W. W.; Wardle, J. (2010), observational B doi.org/10.1002/ejsp.674 (retrieved 2026-11-09)
11. The Mere Urgency Effect — Zhu, M.; Yang, Y.; Hsee, C. K. (2018), RCT B doi.org/10.1093/jcr/ucy008 (retrieved 2026-11-09)
12. Mental fatigue: Costs and benefits — Boksem, M. A. S.; Tops, M. (2008), review B doi.org/10.1016/j.brainresrev.2008.07.001 (retrieved 2026-11-09)
13. Exploring the "planning fallacy": Why people underestimate their task completion times — Buehler, R.; Griffin, D.; Ross, M. (1994), RCT B doi.org/10.1037/0022-3514.67.3.366 (retrieved 2026-11-09)
14. AI Tools in Society: Impacts on Cognitive Offloading and the Future of Critical Thinking — Gerlich, M. (2025), observational C doi.org/10.3390/soc15010006 (retrieved 2026-11-09)
15. Keyword database: DataForSEO Labs keyword overview and historical search volume, US/UK/DE/FR/ES/CZ (data/research/keywords) — Productivity Signal research (2026), dataset A dataforseo.com/apis/dataforseo-labs-api (retrieved 2026-11-09)