The Weekly Review Is the Only Habit That Survives Everything Else

Everyone who has used a task system has watched one decay. The list fills with items from a project that ended, dates slip until they stop meaning anything, and eventually you keep the real list in your head and the system becomes an archive of good intentions. This does not happen because the tool was wrong. It happens because nothing was scheduled to clean it.

Why the list rots

Capture is easy and low friction, which is by design. Deletion requires a judgement, and judgements are effortful, so nobody makes them in passing. Without a scheduled moment for judgement, the ratio of live items to dead items degrades until the list stops being trustworthy. An untrusted list is worse than no list, because you now maintain a second one mentally and feel guilty about the first.

The review that actually works

Thirty to forty-five minutes, once a week, at a fixed time. Four passes:

1. Empty the inboxes. Every capture location: notes, saved messages, the paper on the desk. Everything gets deleted, done immediately, or turned into a real item.
2. Close what is finished. Including the things you finished but never marked, which is more of them than you expect and is the part that makes the review feel good.
3. Kill what is dead. The honest pass. Anything you have deferred four times is a decision you are avoiding: either schedule it concretely or delete it and accept that it is not happening.
4. Choose the week. Pick three outcomes that would make the week a success. Not thirty tasks. Three outcomes, written where you will see them.

The part people skip

Step three is where the value is and where the resistance lives. Deleting a task feels like admitting failure, so items accumulate as a form of debt. Reframing helps: a task you have carried for two months without doing is already not happening, and the list is simply reflecting reality late. Deleting it changes the record, not the outcome.

A list you trust with twelve items beats a comprehensive list you avoid opening.

Adapting it to a team

The same structure works for a team backlog, with one addition: a column for the reason an item is still open. Items that cannot be justified in a sentence get closed. A backlog that has never been pruned is not a plan, it is a record of everything anyone has ever suggested, and treating it as a plan is how quarters get committed to work nobody wants.