A brain dump is one pass in which everything currently open gets written down, in any order, without being sorted, judged, or turned into a plan. It takes about ten minutes. Done alone it is capture, not a solution; a second pass is what turns some of the list into specific plans with a plausible research basis.
The ten-minute procedure
Set a timer, because the point is to finish rather than to be thorough, and an untimed dump becomes an evening.
- One line per item, no punctuation standards, no complete sentences. ‘Nursery form’ is a valid entry.
- No categories, no priorities, no dates yet. Sorting during the dump is what stops the dump.
- Do not fix anything as it comes up. If an item takes under a minute you will be tempted to just do it; do not, because that is where the ten minutes goes.
- Include the trivial ones. The bulb in the hall, the unanswered message, the subscription you meant to cancel—small open items take up the same attention as large ones.
- Keep going past the point where it feels empty. The last two minutes routinely produce the items that were actually bothering you.
- Speak it if that is faster. At ordinary conversational and two-thumb typing speeds, ten minutes of speech can produce several times as many words as ten minutes of phone typing, and a dump is exactly the case where coverage matters more than polish.
Where the idea comes from, and the evidence for it
The practice is old—David Allen's Getting Things Done (2001) calls the exhaustive version a mind sweep—but the interesting evidence is more specific than ‘writing things down helps’.
Masicampo and Baumeister, in the Journal of Personality and Social Psychology in 2011, ran a series of experiments on unfulfilled goals. Unfinished commitments intruded on attention during unrelated tasks, as expected. The useful part is what removed the intrusion: not completing the goal, and not merely listing it, but making a specific plan for how and when it would be handled. The intrusive thoughts fell away once a plan existed, even though nothing had been done.
That result is the argument for a second pass. The experiments tested specific plans, not brain dumps, so a list by itself should not be presented as the intervention; the useful step is deciding what happens next and, where appropriate, when.
The second pass, which is where the value is
Go back through the list once and put every line into one of three buckets. This takes about five minutes and does not require you to do any of the work.
- Has a moment attached → it needs to interrupt you. Give it a time or a place, and let it leave the list.
- Needs a decision before anything can happen → rewrite it as the question. ‘Holiday’ becomes ‘which week in August, and do we book flights before the school dates are out’.
- Is just a record → leave it alone. Reference material, an address, a recommendation. It needs to be findable, not scheduled.
- Anything you cannot place in a bucket is usually two items stuck together. Split it and try again.
Doing it by voice, and why it changes the output
Typed dumps get edited as they are written, which slows them and filters out awkward items. Spoken dumps do not, because speech outruns self-editing. The result is messier but broader, and breadth is the point of the exercise.
The obvious objection is that ten minutes of unstructured speech produces a wall of text nobody will read. That is true if the transcript is all you get. It stops being true when the transcript is broken into individual items you can search by description later—which is the only form in which a dump has value the next day.
When to do one
Three moments repay it reliably: before a week that is already full, after a period where things have been slipping, and at the point where you notice you are rehearsing the same four items in your head on a loop.
A weekly cadence is commonly recommended and commonly abandoned. A more durable rule is to do a brain dump when you notice the same items looping in your head, rather than on a schedule that competes with everything else.
What a brain dump will not do
It does not prioritize. A dump plus a second pass gives you an accurate picture of what is open and what each item needs; deciding which three matter this week is separate work, and no procedure does it for you.
It also has an unpleasant first effect: seeing everything at once is heavier than carrying it vaguely, and the list is usually longer than expected. That passes, but it is worth expecting rather than being surprised by.
And it is a workflow, not a clinical tool. It is frequently recommended for people whose attention is inconsistent, and it may well help there, but nothing in this article is medical advice or a treatment for anything. Finally, a dump captures only what surfaced in those ten minutes—the item you were most successfully avoiding is often the one that does not appear.
Ten minutes, one line per item, no sorting, and keep going past the point where it feels finished—then spend five more putting every line into one of three buckets: needs a moment, needs a decision, or is just a record. The list alone is only capture; the second pass turns some of it into the kind of specific plan the research actually tested.