The cost of guessing
The guess that was right anyway
Most guesses about missing context turn out to be correct. That is the problem: a correct guess costs almost as much as a wrong one and teaches nobody anything.
Decisions this page is about, and where the line actually was
Changing the wording on a customer-facing error message
4 → 1It had been yours all along
Refunding a customer under fifty pounds
3 → 1A written policy you had never been shown
Telling a customer a date
3 → 6Dates are committed by one person here
Reordering your own week
4 → 1You asked twice before believing the answer
- 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
four decisions · three of them you had more room than you thought · one gave you less · the median gap was three steps on the scale
Four of the nine decisions counted on this site.
When autonomy has no visible boundary, a team can explore the available features and use project patterns to identify where clearer authority would remove rework.
For an independent perspective on management and working practice, consult Acas.
Of thirty-two slots of context on the record counted here, fifteen were guessed and never confirmed. Only a handful of those guesses turned out to be wrong.
That sounds like good news and it is the heart of the problem. A correct guess costs nearly as much as an incorrect one, and unlike an incorrect one it produces no evidence that anything was missing.
What a guess costs even when right
The time to make it. Working out a probable constraint from surrounding evidence takes hours and sometimes days. The output is a belief, not a fact.
The hedging it produces. Somebody working on a guess builds for more than one interpretation, because they cannot afford to be wrong. The work is bigger, slower and more general than it needed to be, and nobody can point at the excess afterwards because it is distributed through the design.
The attention it occupies. Running on an unconfirmed assumption means carrying a background process that monitors for contradiction. It is tiring and it is invisible, and it is why a thin brief leaves people more drained than a hard task.
The confidence it fails to produce. A person who was right because they guessed well does not know they were right. They know nothing contradicted them, which is weaker and does not accumulate into judgement.
Why right guesses are worse for the system
A wrong guess produces a return, a visible cost, and a conversation. Somebody notices, and there is at least a chance the cause gets named.
A right guess produces nothing. The work lands, it is fine, and the brief that caused all the hidden effort is confirmed as adequate. The person who handed it over has now received evidence that four lines of chat was enough, because it was: the work arrived.
Over a year that is how a team's briefing standard decays. Not through failures, but through successes achieved at an invisible cost by people who are good at guessing.
The confirmation that is nearly free
The fix is not to stop guessing. Guessing is how work proceeds while the person who knows is in meetings.
The fix is to say the guess out loud, in one line, where the person who knows will see it:
Proceeding on the assumption that the ten-minute downtime limit is contractual and not negotiable. Shout if not.
Three things happen. If the guess is wrong, it is corrected on day one instead of week three. If it is right, it is now a fact rather than a belief, and the hedging stops. And the person who handed the work over sees, in writing, one of the things they did not say — which is the only feedback they will ever get about the brief.
Doing it without sounding uncertain
The wording matters, and the failure mode is turning it into a question.
"Is the downtime limit negotiable?" invites an answer later and reads as not knowing how to proceed. "Proceeding on the assumption that it is not" reads as somebody who has made a call and is being transparent about it, which is exactly what is happening.
Put them in the same place every time — the top of the document, the first comment on the ticket, a standing line in the weekly note. Consistency is what makes them cheap to read, and cheap to read is what makes them get corrected.
Where to put the assumptions
The habit only works if the assumptions are somewhere they will be read, and the choice of place matters more than the wording.
At the top of the working document, if there is one. A short standing block — four or five lines, updated as answers arrive. Anybody who opens the document sees them, which includes the people most likely to know they are wrong.
In the thread where the work was assigned. Highest chance of reaching the one person who can correct them, lowest chance of reaching anybody else.
In the weekly note, as a standing section. This is the most reliable over a long piece of work, because the note is read by people who are not following the detail and are exactly the ones who notice a wrong assumption.
What does not work is putting them in your own notes, or in a comment somewhere in the work itself. Both feel like recording the assumption and neither exposes it, which is the entire function.
A useful shape for the block, whichever place you choose:
Working on: reader is the review group · downtime limit is a target, not contractual · no prior attempt found · the platform lead has not been consulted yet.
Four clauses, updated weekly, read in six seconds.
The number worth watching
Count the slots you are proceeding on without confirmation, once a week. If the number is zero, the briefs are good. If it is five, you are carrying five open risks, and the cost of confirming all five is about four minutes.