Where your say ends
Nobody ever draws the line
Everybody acts on a line nobody told them. The reason is not secrecy: the people who know where it runs cannot see that it needs saying.
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
Rescheduling the weekly review
2 → 2Hiring a contractor for two days
5 → 3Your own budget line covered it
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
five decisions · four of them you had more room than you thought · one you had exactly right · the median gap was two steps on the scale
Five of the nine decisions counted on this site.
Before adding another approval step, it helps to learn about the available capabilities and check whether unclear ownership or missing context is the real source of delay.
Teams can compare this approach with guidance available from HRZone.
Nine decisions, where you thought your say ended and where it actually ended
Changing the wording on a customer-facing error message
4 → 1It had been yours all along, and nobody had said so
Refunding a customer under fifty pounds
3 → 1There was a written policy you had never been shown
Picking the library for the new parser
1 → 4You chose and shipped; the architect had a standing view
Rescheduling the weekly review
2 → 2Dropping a requirement nobody had asked about in six weeks
2 → 5It was a commitment made to a customer before you joined
Hiring a contractor for two days
5 → 3Your own budget line covered it
Changing a database column type
1 → 1Telling a customer a date
3 → 6Dates are committed by one person here, and it is not you
Reordering your own week
4 → 1You asked about this 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
nine decisions · four of them you had more room than you thought · three gave you less · two you had exactly right · the median gap was three steps on the scale
A tenth line was given to this panel and is not drawn. Its real limit was never established — the question is still open, a year on. This sign is entirely about the gap between the two marks, and a line with one end would move the median without anybody being able to say in which direction.
Ask anybody what they can decide without asking and you will get an answer. Ask them how they know, and the answer is almost never "I was told". It is inferred — from what happened the last time, from how somebody reacted, from the seniority of the person who assigned the work, from a general sense of how much trouble one could be in.
The line is real. Somebody enforces it. It governs several decisions a week. And in most organisations nobody has ever written it down or said it aloud.
Why it does not get said
Not secrecy. Three more ordinary reasons.
- The people who know it have never had to find it. They arrived at their authority gradually, by doing things and not being stopped, and so they experience it as obvious rather than as information.
- It is not a single line. It varies by area, by amount, by how reversible the decision is and by who is on leave that week. Writing it down feels like promising more precision than exists.
- Saying it out loud sounds like a restriction. "You can spend up to two thousand without asking" is heard by some managers as a cap they are imposing, when to the receiver it is the first useful thing anybody has said.
The third is the most common and the most backwards. Naming a limit does not create it. The limit was already there; naming it just moves it from a thing you discover by crossing it to a thing you can work inside.
What people do instead
They guess, and they guess low.
Of nine decisions on the record above, four turned out to have more room than the person assumed and three had less. The four are the interesting ones, because nothing happened: the person checked when they did not need to, got a yes, and never learned that the checking was unnecessary. The permission was already theirs and asking confirmed it, which feels like confirmation that asking was right.
That is the mechanism by which an underestimate becomes permanent. A guess that is too cautious is never corrected, because the correction would require somebody to say "you did not have to ask me that", and almost nobody says it.
The two marks
The sign above puts both on one scale: a hollow ring where the person thought their say ended, a filled disc where it actually did. The bar between them is the gap, and the direction is the whole finding.
Towards the left means there was more room than assumed. These cost time and never get noticed.
Towards the right means there was less. These get noticed immediately, which is why everybody remembers them and why people calibrate downwards: the errors that produce feedback are all in one direction.
Two of the nine had no gap. Both were small and both had been tested before.
Why it is not one line
Part of the reason nobody writes it down is that it is not a single boundary, and anybody who sits down to write it discovers this immediately and gives up.
Authority varies along at least four dimensions at once:
- By amount. Fifty pounds is yours, five thousand is not, and the boundary is somewhere nobody has specified.
- By reversibility. The same change is yours if it can be undone this afternoon and not yours if it ships to customers.
- By area. You may own everything inside the import and nothing that touches billing, with no stated edge between them.
- By who is available. The line moves when the person who normally decides is on leave, and moves back without announcement.
Faced with that, writing a complete description does feel like promising more precision than exists, which is a fair objection to producing the complete version and no objection at all to producing a partial one.
The partial version — four or five concrete examples with a step against each — takes ten minutes, is obviously incomplete, and settles most of the recurring cases. Incomplete and written beats complete and imagined, and the imagined complete version is what everybody is waiting for.
What would fix it
One paragraph, written once, by whoever assigns the work.
Not a policy document and not an approval matrix. A paragraph naming three or four things the person can decide alone, one or two that need a word first, and one that is not theirs at all. It will be incomplete and slightly wrong, and it will still be better than the current state, which is nothing.
The paragraph has a second effect that is worth more than the first. It makes the line discussable. Somebody can read it and say "I think the refund one should be higher", which is a conversation that cannot happen about an unwritten rule, because there is nothing to point at.
The rest of this part is about what to do while waiting for that paragraph, which in most places means indefinitely: how to find the edge without testing it, what the six steps of deciding alone actually are, and why the cheapest version of this is to draw your own line and have it checked.