The GTD weekly review checklist

The weekly review is the habit that makes every other part of GTD work, and it is the one people skip first. David Allen's version has eleven steps in three phases — get clear, get current, get creative — and takes about an hour once you are practised. Here is the whole thing, in order.

Why this is the step that matters

A task system decays quietly. Things get captured and never processed, projects finish without being closed, and a list you have stopped trusting is a list you have stopped reading — usually without noticing the moment it happened.

The weekly review is the maintenance that stops that. Allen's own framing is that it is the critical success factor for the whole method: not because reviewing is virtuous, but because a system you can trust is the only kind that lets you stop carrying things in your head. He suggests running it every seven to ten days.

Everything below assumes nothing about which app you use. It works on paper.

Get clear

The point of this phase is that nothing is left floating — not in your bag, not in your inbox, not in your head.

1. Collect loose papers and materials. Business cards, receipts, notes to self, anything on the desk. Physically pile it in one place. You are not processing yet, only gathering.

2. Get "in" to zero. Process every note, email and voicemail. Processing does not mean doing — it means deciding what each thing is and where it goes. Two minutes or less, do it now; otherwise it becomes a next action, a project, a someday item, a reference note, or rubbish.

3. Empty your head. Write down anything still rattling around. New projects, action items, things you have been meaning to say to someone. Keep going until it stops coming; the last three things out are usually the ones that were actually bothering you.

Get current

Now the system gets brought up to date with reality, which is the part that takes the longest and is worth the most.

4. Review your next-action lists. Mark off anything already done. Look for items that have quietly stopped being the next action — those are projects wearing a task's clothes.

5. Review previous calendar data. Go back two or three weeks. Meetings generate follow-ups that nobody wrote down, and this is where you find them.

6. Review the upcoming calendar. Look forward far enough to see what needs preparing — usually two to four weeks. Anything that needs work before the date becomes a task with its own date, not a thing you hope to remember.

7. Review your waiting-for list. What have you delegated or ordered that has not come back? Chase what is overdue, and note the date you asked.

8. Review your projects list. Every project, one at a time. For each one: is it still alive, and does it have a next action? A project without a next action is stalled, and stalled projects are what make a system feel heavy.

9. Review any relevant checklists. Job responsibilities, recurring routines, areas of focus. This is where you notice the whole area you have not thought about in a month.

Get creative

The phase everyone drops, and the reason the review is worth an hour rather than being a chore to survive.

10. Review your someday/maybe list. Anything ready to become a real project? Anything you can honestly delete? A someday list you never review becomes a graveyard, and deleting things from it is a feature.

11. Be creative and courageous. Allen's own words invite "any new, wonderful, hare-brained, creative, thought-provoking, risk-taking ideas" into the system. This is the step that turns a maintenance chore into the most useful hour of the week.

Where people actually give up

Having watched this fail in a few different ways:

• It is too long the first time. The first review is not an hour, it is three, because it is really a backlog cleanup wearing a review's clothes. Do the cleanup once and separately, then start the weekly habit.
• No fixed slot. A review that happens "when there is time" happens four times and stops. Friday afternoon works for many people because the week is still in memory; Monday morning works if you would rather plan than close.
• Skipping get creative. If you only ever do steps 1 to 9, the review is pure admin and you will resent it. Step 11 is the payoff.
• Reviewing in the app you are reviewing. Being inside the tool invites fiddling with the tool. Printing this and working through it on paper is faster, which is why the page is built to print.

Print it

Use your browser's print command. The header, footer and everything else is stripped, leaving the eleven steps on a clean sheet you can tick with a pen and keep on the desk.

Allen's own checklist is a free PDF from the source, linked below, and it is worth having too — this page is a longer-form companion, not a replacement for his.

If you keep the list in an app

Any app that can hold a projects list, a next-actions list, a waiting-for list and a someday list will do. Nothing above needs a particular product, and the steps have outlasted every tool they have been done in.

One structural thing worth checking, whatever you use: whether someday is a place or a label. If it is a tag, a someday item still sits in its project and still catches your eye in every list, which is how someday lists turn into graveyards. In TaskPocket it is a location — moving something there takes it out of its project entirely, and moving it back is one drag. That is the only sentence about our product on this page, and the checklist above works exactly as well without it.