每周复盘与待办检查

The GTD Weekly Review Canon#

Why David Allen calls the weekly review the "critical success factor" of the whole method, what the three phases actually contain, and why the cadence matters more than the format.

The claim: the review is the keystone, not an accessory#

Getting Things Done rests on one psychological bargain: your mind will stop nagging you about commitments only if it trusts that the external system holds all of them and that the system gets looked at. Capture alone doesn't buy that trust — a list nobody reviews is just a graveyard with better formatting. Allen is explicit that the weekly review is "the critical success factor" and that people who fall off GTD almost always fall off here first: they keep capturing, stop reviewing, watch the lists rot, and then stop trusting the lists — at which point everything moves back into their head and the whole system collapses.

The corollary this skill enforces: a review that skims is not a review. Marking the ritual "done" while the core steps went unwalked actively damages trust in the system, because now the lists are stale and carry a false freshness stamp. Hence the hard gate: unaccounted-for GET CURRENT steps always force an INCOMPLETE verdict.

The three phases#

Allen's own material organizes the checklist into three movements, in a deliberate order:

1. GET CLEAR — return to ground zero#

• Collect loose papers and materials. Physical and digital strays: receipts, notes, downloads, screenshots, business cards. Everything into an inbox.
• Get "in" to zero. Process every inbox — email, notes app, task inbox, desk tray — by clarifying, not doing. Each item becomes a next action, a project, a waiting-for, a someday/maybe, reference material, or trash. The two-minute rule is the only exception: if acting takes under two minutes, do it now.
• Empty your head. A mind sweep: write down every new project, task, or nagging thought still riding in working memory. This is the step that converts psychic RAM into system entries.

2. GET CURRENT — make the system match reality (the mandatory core)#

• Review next-action lists. Mark off what's done, prune what's dead, notice what's stuck.
• Review the previous calendar. Look back over the past week or two for missed commitments, follow-ups spawned by meetings, promises made in passing. Each becomes a captured item.
• Review the upcoming calendar. Look ahead so preparation happens now instead of panic later.
• Review the waiting-for list. Everything delegated or pending from others: chase it, re-date it, or drop it. Unreviewed waiting-fors are where other people's slippage becomes your surprise.
• Review project lists. The load-bearing step: every active project must have exactly one next action on a list. A project with none is a wish; a project whose next action is buried is stalled. This step is why GTD projects keep moving without daily replanning.

3. GET CREATIVE — lift your eyes off the runway#

• Review someday/maybe. Activate what's become timely, keep what still sparks, kill what doesn't. This list only works if it's periodically culled — otherwise it becomes a guilt museum.
• Capture new ideas. With a clear head and a current system, ambition is safe again. Allen's point: creativity shows up after clearing, not instead of it.

Cadence discipline#

• Weekly is the resolution of most working commitments. Allen's observation is pragmatic, not mystical: most people's world shifts materially over a week — meetings compound, projects drift, promises accumulate. Daily is too granular to see project-level drift; monthly lets loops rot.
• Same time, same trigger. The canonical recommendation (echoed by every serious GTD guide) is a recurring, protected block — commonly Friday afternoon or Sunday evening — anchored to an existing rhythm so the review doesn't depend on motivation.
• Timeboxed. One to two hours. A review that regularly takes three hours gets skipped the moment a week is busy, and skipped reviews are how the habit dies. Better a 45-minute honest pass with steps explicitly skipped-with-reason than a marathon that happens twice and never again.
• The review is steering, not rowing. Forte's periodic-review framing and Allen's original agree: during the review you decide and route; you do not execute (two-minute items excepted). A review that turns into a work session never finishes.

What "done" means#

The review is complete when: every inbox is empty, every open loop found is on a list, every active project has one next action, both calendars have been swept, every waiting-for has been touched, and someday/maybe has been consciously kept or culled. That definition is checkable, which is why this skill gates it with a script instead of a feeling.

Sources#

1. David Allen, Getting Things Done: The Art of Stress-Free Productivity (Penguin, original edition 2001; revised edition 2015) — chapter on the power of the weekly review; the "critical success factor" claim; the two-minute rule; the project/next-action distinction.
2. David Allen, Ready for Anything: 52 Productivity Principles for Getting Things Done (Viking, 2003) — principles on review cadence and why unrenegotiated commitments erode self-trust.
3. GTD® methodology guides, gettingthingsdone.com (David Allen Company) — the official Weekly Review® checklist with the GET CLEAR / GET CURRENT / GET CREATIVE structure used here.
4. Tiago Forte, Building a Second Brain (Atria, 2022) and the PARA method essays (fortelabs.com) — periodic review as the maintenance loop of any personal knowledge system; weekly review as routing, not execution.
5. Doist, "The Complete Guide to Getting Things Done" (todoist.com/productivity-methods) — a widely used practitioner's walkthrough of the weekly review steps and common adaptations.
6. Cal Newport, Deep Work (Grand Central, 2016) and the weekly-planning essays at calnewport.com — weekly planning as the scale at which meaningful work is steered; the shutdown-ritual analogue of closing loops before rest.
7. Francis Wade, Perfect Time-Based Productivity (2014) — on why review cadences fail and how habit load, not intention, predicts whether a review sticks.