What was not said
Reading a brief for what is missing
Ten minutes with a new brief, looking for the four absences rather than reading what is there. A procedure and the questions each gap produces.
Decisions this page is about, and where the line actually was
Picking the library for the new parser
1 → 4The architect had a standing view
Dropping a requirement nobody had asked about in six weeks
2 → 5A commitment made before you joined
Telling a customer a date
3 → 6Dates are committed by one person here
- 1Decide it alone and say nothing
- 2Decide it alone and mention it afterwards
- 3Check with one colleague first
- 4Check with the person who assigned it
- 5Needs a decision one level up
- 6Cannot be decided here at all
three decisions · not one of them gave you more room than you thought · three gave you less · the median gap was three steps on the scale
Three of the nine decisions counted on this site, all going the same way.
For distributed work, teams can read the complete overview to add evidence about capacity, attendance, and project effort without creating another status ritual.
The wider organisational question can also be compared with resources from IESE Business School.
Four handovers read against the same four slots
Move the scheduler onto the new queue
4 of 4Nothing to ask
Rewrite the import so it does not time out
2 of 4Two gaps, both found on the first reading
Write the migration guide
2 of 4One gap and one that arrived at the draft
Fix the signup flow
0 of 4Four gaps, and the brief was four lines
four handovers · 16 slots in all · the median arrived with two of the four · one was complete · one arrived with none of the four
Four of the eight handovers counted on this site, read in order of how much was there.
Reading a brief for what it says takes a minute. Reading it for what it does not say takes ten, and it is the highest-return ten minutes available at the start of any piece of work.
The procedure is mechanical, which is the point: you are looking for four specific absences, and absences are hard to notice without a list to check against.
The procedure
Print it, or copy it somewhere you can mark up. Then four passes, in this order.
- Constraint. Underline every number, date and limit. For each, write next to it what it is attached to. Any number you cannot attach to a contract, a commitment or a person is a number somebody estimated, and it may be negotiable.
- Precedent. Search the words of the brief in whatever holds your team's history — the repository, the document store, the channel archive. You are looking for the same nouns eighteen months ago.
- Decider. List everybody named in the brief. Then ask who is not named who will be affected. The second list is where the vetoes live.
- Purpose. Write one sentence beginning "this is wanted because". If you cannot finish it from the brief alone, you have the most important gap.
Ten minutes. The output is a list of between zero and four questions.
What each gap sounds like
| Gap | How the brief reads | The question it produces |
|---|---|---|
| Constraint | A date with nothing attached | What is the date attached to? |
| Precedent | A familiar noun, no history | Has anyone tried this before? |
| Decider | Only the assigner is named | Who else will have a view? |
| Purpose | A goal that could mean anything | Who asked, and what will they do with it? |
Four questions, asked in one message, at the start. Each is cheap to answer and specific enough that the answer is usable.
Scoring a brief before you start
The panel above reads four of the quarter's handovers against the same four slots, in order of completeness.
The top one needed no questions. The bottom one was four lines of chat and needed all four, which is the ordinary relationship: the shorter the brief, the more of the four are missing, because short briefs are short precisely when the work sounded self-explanatory.
Scoring it takes thirty seconds and it tells you something practical: how much of your first week is reconstruction rather than work. A brief with one gap is a normal start. A brief with four is a week of detective work that nobody has budgeted, and saying so at the beginning is far easier than explaining it at the end.
The one thing not to do
Do not fill the gaps silently.
The temptation is strong, especially for anybody who wants to appear self-sufficient: you can infer a plausible constraint, assume a reasonable purpose, and proceed. Everything then looks fine until the work is finished and the inference turns out to be wrong, at which point the cost is maximal and the responsibility is ambiguous.
If you must proceed on an assumption — and often you must, because the person who can answer is away — write the assumption down where they will see it, in the form of a statement rather than a question. There is a note on exactly this, and it is the single most useful habit in this part of the site.
Reading what is there for what it implies
The four passes look for absences. A fifth, quicker pass is worth doing on what the brief does contain, because the wording frequently carries information nobody meant to put there.
Hedged verbs. "See if we can", "have a look at", "explore" usually mean the purpose is not settled. Work described tentatively is work whose shape will change, and that is worth knowing before committing to an approach.
Named people in passing. Anybody mentioned by name in a brief is there for a reason, even when the sentence seems incidental. "You might want to talk to the platform team at some point" is almost always a decider being half-named by somebody uncomfortable naming them.
A date with no object. "By the end of the month" with nothing attached is either a real date whose attachment was assumed, or an invented one. Both are worth one question.
Comparison to something else. "Like the thing we did for billing" is a precedent being offered without being identified as one. Go and look at the billing thing; half the context is in it.
None of this is reliable. It is a set of guesses that cost thirty seconds and tell you which of the four questions to prioritise when you only get to ask one.
And when the answers never come
Some briefs cannot be completed, because the person who could complete them does not know either. That is worth establishing early, because it changes what you are doing: you are not reconstructing a brief, you are writing one.
Say so explicitly. "Nobody seems to have decided the audience, so I am going to decide it and say what I decided" is a reasonable sentence and it moves a hidden risk into the open, where somebody can object to it while objecting is still cheap.