Getting Things Done

The problem: open loops

David Allen, a management consultant and coach, opens with a claim about stress. Most of the pressure people feel does not come from having too much to do. It comes from the gap between what they have agreed to do and what they have actually tracked. He calls each untracked commitment an open loop: a promise to yourself that your mind keeps repeating because it does not trust that you will remember it. Open loops range from "buy cat food" to "decide whether to change careers." Your mind does not distinguish by importance. It nags about all of them, often at 3 a.m.

Allen's goal is a state he borrows from karate: mind like water. Water responds to a thrown pebble with a reaction exactly proportional to the force, then returns to calm. A person with mind like water handles what comes in and does not overreact or underreact. The book argues that you reach that state by getting everything out of your head and into a system you trust, so that your mind can stop holding it.

The book has three parts. The first explains the method. The second walks through setting it up over a full weekend. The third explains why it works and what changes when you use it. The 2015 revised edition updates the tools, adds a chapter on the cognitive science behind the method and addresses digital workflows, but the method itself is the same as in 2001.

The five steps

Allen's method is a workflow with five steps. The 2001 edition called them collect, process, organize, review and do. The revised edition calls them capture, clarify, organize, reflect and engage.

1. Capture: collect everything that has your attention into a small number of inboxes. A physical in-tray, a notebook, an email inbox, a voice memo app. The rule is that every open loop goes into an inbox, not into your head. Allen suggests a full mind sweep at the start, using a list of triggers (projects, family, finances, health, home repairs, career) to pull out everything you are carrying.
2. Clarify: process each item in the inbox, one at a time, from the top, and never put anything back. For each item ask: what is it, and is it actionable?
3. Organize: put the result where it belongs, in one of a small set of lists.
4. Reflect: review the lists often enough that you trust them, with a full weekly review as the anchor.
5. Engage: choose what to do in the moment with confidence, because you can see all your options.

Allen stresses that most people fail because they mix the steps. They try to organize while capturing, or decide what to do while processing. Keeping the steps separate is what makes the system work.

Clarifying: the workflow diagram

The core of the method is a decision tree for each inbox item.

If the item is not actionable, there are three options. Trash it. File it as reference material you may want later. Or put it on a Someday/Maybe list of things you might do but have not committed to, such as learning Italian or building a deck.

If the item is actionable, ask two questions. First, what outcome am I committed to? If that outcome takes more than one action step, it is a project, and it goes on the Projects list. Allen's definition of a project is deliberately broad: "get new tires" is a project because it needs a call, a visit and a payment. Second, what is the next action? The next action is the next visible, physical thing you would do to move the outcome forward. Not "plan the offsite" but "email Sarah to ask for three possible dates."

Once you have the next action, there are three ways to handle it:

• Do it if it takes less than two minutes. This is the two-minute rule. It takes longer to track the item than to finish it.
• Delegate it if someone else should do it. Put it on a Waiting For list with the date and who owes you.
• Defer it if it takes longer and it is yours. If it must happen on a specific day or at a specific time, put it on the calendar. Otherwise put it on a Next Actions list.

Allen is strict about the calendar. It holds only three things: time-specific actions (appointments), day-specific actions (something that must happen on Tuesday) and day-specific information (a reminder that a colleague is out). A daily to-do list on the calendar is forbidden, because items that do not get done have to be rewritten, and the calendar loses its authority as the hard landscape of the day.

Organizing and reflecting: lists, contexts and the weekly review

Allen's system uses a small number of lists. Projects. Next Actions. Waiting For. Someday/Maybe. Calendar. Reference. Project support material (notes and plans tied to a project). That is the whole system.

Next Actions are sorted by context, the tool, place or person needed to do them. Typical contexts are @Calls, @Computer, @Errands, @Office, @Home, @Agendas (things to raise with a specific person at the next meeting) and @Read/Review. Context sorting means that when you are at the hardware store with 20 minutes, you look at @Errands and nothing else. The 2015 edition notes that smartphones have blurred some contexts, but Allen argues that the principle of seeing only what you can do right now still holds.

For reference material he recommends a simple A to Z filing system that you can file into in under a minute, with a labeler so that folders look neat enough that you use them. He also describes the tickler file: 43 folders, 31 for the days of the month and 12 for the months, that let you drop a paper or a note into the future so it shows up on the day you need it.

Allen calls the weekly review the critical success factor of the method. Once a week, you block one to two hours to get clear, get current and get creative. You empty every inbox. You process loose papers and notes. You review the calendar for the past week (anything to capture?) and the coming weeks (anything to prepare?). You walk through every project and make sure each one has a next action. You review Waiting For and chase what is overdue. You look at Someday/Maybe and move items into Projects or delete them. Without this review, lists go stale, trust collapses and the mind starts holding things again.

Engaging and planning: choosing what to do

Allen offers three models for choosing in the moment.

The four-criteria model says you choose an action by context (what can I do here?), time available (how long until the next commitment?), energy available (am I sharp or drained?) and priority (given the first three, what matters most?). Priority comes last because it only applies to the options the first three leave open.

The threefold model describes the three kinds of work: doing predefined work from your lists, doing work as it shows up and defining your work (processing inboxes and planning). Most people spend their days on the second and never get to the third, which is why their lists are never current.

The six-level model connects actions to purpose through what Allen calls horizons of focus. The 2001 edition used an airplane metaphor: runway for current actions, 10,000 feet for projects, 20,000 for areas of responsibility, 30,000 for one- to two-year goals, 40,000 for a three- to five-year vision and 50,000 feet for purpose and principles. The revised edition numbers them Ground and Horizons 1 to 5. Allen argues that most productivity books start at the top, with purpose. He starts at the bottom, because a person who cannot find their cat food list cannot think clearly about their life purpose.

For projects, Allen describes how the brain plans when it works well, such as deciding where to go for dinner. The natural planning model has five phases: define purpose and principles (why do this, and what standards apply?), outcome visioning (what does done look like?), brainstorming (how might we get there, with no judgment yet), organizing (sort the ideas into components, sequences and priorities) and identifying next actions. He contrasts this with the unnatural planning model used in most organizations, which starts with a good idea, then jumps to organizing in a meeting before anyone has agreed on purpose or outcome. His advice for a stuck project is to move down the list: if you lack action, do more organizing; if organizing is chaotic, brainstorm more; if brainstorming is dry, revisit the outcome.

Setting it up and why it works

Part two is a step-by-step guide to a two-day setup: block the time, set up a workspace with an in-tray and filing, gather every physical item into the in-tray, do the mind sweep, process everything to zero and build the lists. Allen insists on physical tools even for digital workers, because paper still arrives.

Part three explains the effects. The capturing habit reduces anxiety because you no longer have to trust your memory. The next-action decision turns vague worries into concrete steps and ends the procrastination that comes from undefined work. Outcome focusing links daily actions to results. The 2015 edition adds research support, including work on distributed cognition and a study showing that writing a plan for an unfinished task reduces the intrusive thoughts the task produces. Allen also argues that the system is for anyone, from executives to students to parents, because the problem it solves is universal.

What the critics say

The most common complaint is that GTD takes a lot of effort to set up and maintain, and that many readers abandon it after a few weeks when the weekly review slips. Writers such as Cal Newport argue that the method handles small tasks well but says little about deep, focused work or about choosing which commitments to make in the first place. Others note that the book is dense and repetitive, and that its emphasis on contexts and paper tools dates it even after the 2015 revision. Fans answer that any part of the system, such as the two-minute rule or the next-action question, works on its own.

Who should read it

Read it if you carry a long list of undone things in your head and feel vague stress about work you cannot name. It suits managers, freelancers and anyone whose job arrives by email and meetings. Readers who want a philosophy of what to work on should pair it with Deep Work or Essentialism.