Where your say ends
Draw your own line and check it
Do not wait to be told. Write the line down yourself, send it, and ask for corrections — which is a much easier thing for somebody to give than a map from nothing.
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.
When responsibility is handed over without enough context, managers can see whether it fits and compare project time and workload before treating delay as an individual failure.
A separate professional source for this subject is IPMA.
Three decisions, the smallest version of this exercise worth doing
Changing the wording on a customer-facing error message
4 → 1Checked afterwards: it had always been yours
Picking the library for the new parser
1 → 4Found out the hard way, in production
Changing a database column type
1 → 1Small, reversible, and tested once before
- 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 · one of them you had more room than you thought · one gave you less · one you had exactly right · the median gap was three steps on the scale
Three of the nine decisions counted on this site. Two were wrong in opposite directions and one was exact, which is the ordinary result.
Waiting to be told where your authority ends means waiting indefinitely, for the reasons in the rest of this part: the people who know cannot see that it needs saying, and producing the full map feels like a day's work to them.
So do not ask for the map. Draw it, wrongly, and ask for corrections. Correcting a draft is a five-minute job and a pleasant one; producing a document from nothing is neither.
The exercise
Twenty minutes, once.
- Write down ten decisions you have actually made or escalated in the last two months. Real ones, with specifics — not categories.
- Against each, put the step you used at the time, on the six-step scale: decide alone and say nothing, decide alone and mention it, check with a colleague, check with whoever assigned it, one level up, or not ours at all.
- Against each, put the step you now think it should have been.
- Send the list to whoever assigns you work, with one sentence: "this is how I have been treating these — tell me which ones are wrong."
That is the whole thing. The output is a corrected list, which is the map, and it arrives in one exchange rather than over three quarters of inference.
Why this works when asking does not
It gives the other person something to react to, and reacting is cheap. "No, the first one is yours" takes four seconds. "Write down everything I can decide" takes an afternoon that nobody has.
It is specific. The answers attach to real decisions, so they are unambiguous in a way that general guidance never is.
It is not a request for more authority. That matters more than it should. Asking for a bigger remit is a negotiation and people prepare for negotiations; asking whether your understanding is accurate is a factual question and people answer those immediately.
And it is evidence. The list shows how you have been operating, which is usually better than the other person assumed.
What comes back
Expect corrections in both directions, and expect the ratio to surprise you. On the record counted on this site, nine decisions produced four where there was more room than assumed, three where there was less, and two exact.
The four are the ones that change your week. The three are the ones that save you an argument later. The two exact ones are worth noting too — they tell you which kinds of decision you are already calibrated on, and those are usually the ones somebody was once explicit about.
Doing it from the other side
If you assign work to people, you can run the same exercise without waiting for anybody to ask.
Send three decisions you have seen them make and ask what step they used. The answers will tell you where they are guessing low, which is almost always somewhere, and you can correct it with a sentence.
It takes ten minutes and it is the only intervention on this site that produces an immediate, measurable change in how fast work moves — because every correction in the direction of more room removes a recurring wait from somebody's month.
The reaction to expect
Two reactions are common and neither is what people fear.
The first is mild surprise at the accuracy. Most people's lists come back with seven or eight of ten confirmed, which is a better result than the person expected and useful on its own: it tells you that your model is broadly right and that the hesitation is costing more than the uncertainty warrants.
The second is a correction in the generous direction. Managers reading a list of decisions somebody has been escalating frequently respond with some version of "you do not need to ask me about any of these". That sentence is almost never volunteered and reliably appears when a list is put in front of somebody.
What does not usually happen is the thing people are worried about: being told that several things were overstepped. If there were a pattern of overstepping, you would have heard about it by now — overstepping is the error that produces immediate feedback.
So the asymmetry that makes the line hard to find also makes this exercise safe. The errors you cannot see are almost all in the cautious direction, which means the corrections you get back will almost all be permissions.
And write it down afterwards
The corrected list is the artefact. Keep it where the work lives, not in a private note, and put the date on it.
Six months later, when something has moved, you will have a thing to compare against. And when somebody new arrives, you can hand them a list that took you a quarter to assemble and will take them four minutes to read — which is the entire argument of this site, applied to itself.