Getting Things Done: a practical GTD guide for people with actual work

The most common to-do list is not a list of actions. It is a list of mildly threatening nouns: budget, dentist, launch, taxes, kitchen.

David Allen’s Getting Things Done fixes that by separating two jobs we usually mix together: deciding what something means and doing the action.

GTD is not mainly about getting more done. It is about making commitments clear enough that your brain no longer has to keep reopening them at random.

The GTD workflow

Stage	Question
Capture	What has my attention?
Clarify	What is it, and is action required?
Organize	Where should the result live?
Reflect	Is the system current and complete enough to trust?
Engage	What is the right action now?

The sequence matters. If you organize before clarifying, you get beautifully labeled confusion.

The official GTD overview describes the same five steps. The method is a practical framework, not a single scientifically validated intervention. Its parts overlap with research on external memory, implementation planning, and situated cognition, but the complete branded system does not come with one universal effect size.

Why an ordinary list stays mentally noisy

A mixed list asks you to interpret each item whenever you see it. “Taxes” might be a project, a deadline, a missing document, or a five-minute download. The list stores the reminder but not the decision.

GTD separates raw inputs from clarified commitments. That separation is why capture alone is not enough. A full inbox can reduce fear of forgetting while still leaving every item ambiguous.

Capture: stop using attention as storage

Write down commitments, ideas, requests, and loose ends in a small number of inboxes. An inbox can be email, a notebook, a voice-capture shortcut, or a physical tray.

Capture is not planning. “Mum’s birthday” may be enough to get the thought out of your head. The decision comes later.

Too many inboxes defeat the point. A note in a forgotten app is not external memory. It is archaeological material.

Researchers use the term cognitive offloading for moving memory or computation into the environment. A review by Risko and Gilbert describes how people use notes, reminders, physical actions, and tools to reduce internal cognitive demands.

Offloading works only when the external system remains available and trustworthy. Choose a small number of capture points and define when each is processed. Email can remain an inbox; it should not also be the only project list, reference archive, and calendar.

Clarify: turn stuff into decisions

For every item, ask:

1. What is it?
2. Is any action required?
3. What is the next physical action?
4. If one action is not enough, what is the desired project outcome?

Examples:

Stuff	Next action
Taxes	Download last year’s income statement
Launch	Ask Maya to confirm the release date
Dentist	Call the clinic and request the next available check-up
Kitchen	Measure the space for a new shelf

Clarity does not make an unpleasant task fun. It removes the extra task of figuring out how to begin.

Projects and actions are different objects

“Publish the annual report” is an outcome requiring several actions. “Email Finance for the final revenue table” is an action. Keeping both visible prevents the action list from losing direction and the project list from becoming a museum of nouns.

A concrete plan can reduce intrusive unfinished-goal thoughts

In experiments by Masicampo and Baumeister, unfinished goals were more likely to intrude during another task. Forming a concrete plan reduced that interference. The authors emphasized that the effect came from plans that helped execution, not from writing anything at all. The study was published in the Journal of Personality and Social Psychology.

This supports a narrow GTD principle: “deal with report” is weaker than a credible next action connected to a situation. It does not prove that every item needs a detailed project plan.

The two-minute rule

If an item takes roughly two minutes, Allen recommends doing it while processing.

Useful:

• confirm an appointment;
• forward a document;
• add a date to the calendar.

Dangerous interpretation: spending the whole morning completing tiny requests because each one is “only two minutes.”

The rule belongs inside inbox processing. Important work still needs protection from a thousand efficient interruptions.

Organize by meaning, not by anxiety

A basic system needs:

• Next actions
• Projects
• Calendar
• Waiting for
• Someday/maybe
• Reference

The calendar is for things tied to a real date or time, not a motivational display of everything you hope to do.

A project is any outcome requiring more than one action. The project list keeps the outcome visible; the next-action list keeps movement possible.

The David Allen Company advises against hiding all current next actions inside project notes. Otherwise, a person with twenty free minutes must open every project to rediscover what can move. The connection is restored during review. See the official explanation of linking next actions and projects.

Calendar

Use it for actions that must happen at a particular time or date, plus information that is useful on that date. Do not schedule every wish; false appointments make the calendar less trustworthy.

Waiting for

Record what you requested, from whom, and when you will follow up. “Waiting” should remain a monitored state rather than a reason the project silently disappears.

Someday/maybe and reference

Someday/maybe holds options you are not committed to now. Reference holds information that requires no action. Review the first at an appropriate rhythm; search the second only when needed.

The weekly review is the engine

Lists become useless when they stop matching reality. The weekly review updates the map:

• clear inboxes;
• scan upcoming and recent calendar entries;
• review projects;
• add missing next actions;
• remove obsolete commitments;
• check waiting-for items.

The goal is trust, not perfect cleanliness. A 30-minute review that happens is better than a two-hour ceremony you avoid.

During review, ask two separate questions: “Is the outcome still active?” and “Does it have a current next action?” A project can remain valuable while its old action becomes obsolete. It can also contain well-written actions that no longer serve a useful result.

The review interval should match change. Weekly is a practical default for personal systems, not a law. A fast-moving launch may need a brief daily scan; a someday list may need monthly attention.

Choosing what to do: context is not a command

GTD traditionally considers context, time available, energy, and priority. Those filters help narrow options, but they do not create capacity or resolve organizational conflict.

If two deadlines cannot fit, communicate the trade-off. If low energy is persistent or health-related, do not make “energy context” a way to normalize overload. If every action is available everywhere, context tags add maintenance without improving choice.

Use the smallest set of filters that changes actual decisions. Location and tool access may matter; fifteen mood categories probably do not.

What research can and cannot say about GTD

Francis Heylighen connected GTD with distributed and situated cognition, explaining how external memory, environmental triggers, and predefined actions can support its logic. The analysis appeared in Long Range Planning.

That is a theoretical account, not proof that GTD outperforms every task system. Research supports some mechanisms—offloading, specific planning, environmental cues—under particular conditions. The effectiveness of a personal setup still depends on review, workload, task quality, and whether the user can act on the commitments.

How GTD becomes productive procrastination

Common failure modes:

• rebuilding the app instead of doing the work;
• using tags that never affect a decision;
• collecting faster than reviewing;
• writing outcomes as actions;
• keeping commitments you no longer intend to honor;
• treating the weekly review as a moral audit.

The smallest viable GTD system may be one inbox, one next-action list, a project list, and a calendar.

Other failure modes deserve attention:

• keeping hundreds of projects active because archiving feels like failure;
• processing every input immediately and never protecting important work;
• treating capture as permission to accept unlimited commitments;
• choosing a concrete action that moves efficiently toward the wrong outcome;
• turning review into a backward-looking score of personal worth.

The repair is usually subtraction: fewer inboxes, fewer tags, fewer active projects, and a shorter review with explicit decisions.

A 30-minute reset

1. Capture every loose end for ten minutes.
2. Choose ten current items.
3. Clarify each one.
4. Put dates only where dates are real.
5. Choose three actions you can do this week.
6. Schedule a short review.

Do not process your entire digital history. GTD is allowed to begin today.

Minutes 0–10: capture

Collect current open loops without organizing them. Stop when the timer ends; the goal is a usable sample, not total memory extraction.

Minutes 10–20: clarify

Process the first ten items. Delete, archive, delegate, do the genuine two-minute action, or define the next action and outcome.

Minutes 20–25: organize

Create only the lists required for those items: actions, projects, waiting, and real calendar entries.

Minutes 25–30: leave a return point

Choose three actions for the coming week and schedule the next short review. A system becomes trusted through repeated maintenance, not one heroic cleanup.

Using GTD with Flammy

Flammy is not designed to be a full reference archive or email-processing system. Use your task manager for the complete inventory.

Bring selected items into Flammy when the gaming layer helps:

• a recurring next action;
• a larger outcome as a quest;
• a difficult action with a small minimum;
• a personal reward for meaningful effort.

GTD clarifies what the work is. Flammy can make doing the chosen work more visible.

The verdict

GTD works when it lowers uncertainty:

1. capture what has your attention;
2. decide what it means;
3. name the next action;
4. keep outcomes separate from actions;
5. review often enough to trust the system.

The goal is not an empty list. It is opening the list and knowing what to do next.

Related reading

• Continue with our Jedi Techniques guide.
• Continue with making a realistic to-do list.