What autonomy needs
Autonomy is permission, not instruction
Being told to decide for yourself is not the same as being able to. The permission is free to give; the context that makes it usable is what costs something.
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
Changing a database column type
1 → 1Small, reversible, 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 · not one of them gave you more room than you thought · two 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.
To keep measurement proportionate, teams can see the workflow explained and agree what is collected, who can see it, and how it will support decisions.
The wider organisational question can also be compared with resources from Change Management Institute.
"You own this. Run with it." Two sentences, warmly meant, and in most cases entirely sincere. The person saying them has handed over a real thing: the authority to decide. They have also, without noticing, kept the four things that make the authority usable, because they already have all four and cannot see them as separate from the work.
That is the whole of this site in a paragraph. Autonomy is the permission. Context is what turns the permission into something you can act on. Handing over the first without the second is not generosity and not neglect; it is a failure of imagination about what the other person can see.
What the person handing over can see
Everything. That is the problem.
They know the budget because they argued for it. They know the thing was tried in the spring because they watched it fail. They know the one person who will object, because that person objects to everything in this area and everyone has learned to check first. They know the work is wanted because a customer asked for it on a call they were on.
None of that is written anywhere. It is not withheld; it is simply not visible to the person holding it, in the way that your own accent is not audible to you. When they say "you own this", they hear themselves offering freedom. The person receiving it hears a sentence with no information in it.
What arrives instead
A task, a rough deadline, and permission. Which leaves the receiver doing four separate pieces of detective work before any of the actual work can start:
- Working out the constraint, usually by proposing something and seeing what the reaction is.
- Working out whether it has been tried, usually by finding a stale branch or an abandoned document six weeks in.
- Working out who can overrule them, usually by being overruled.
- Working out why it is wanted, which is the hardest, because the stated reason and the real one are often different and both sound plausible.
Each of those takes days and produces an answer that may be wrong. The stated reason it took so long is always "it turned out to be more complicated than we thought".
Why this is not a case against autonomy
It would be easy to read all of the above as an argument for closer direction, and it is the opposite.
Detailed instruction solves the information problem by removing the decision, which costs the thing autonomy was for: the person closest to the work deciding how it is done. The answer is not to take the permission back. It is to hand over the four things alongside it, each of which takes about a minute to state and a quarter to guess.
This also explains a pattern that otherwise looks like contradiction. The teams where people feel most free are usually the teams with the most explicit constraints — written spending limits, named decision owners, standing rules about what needs review. The explicitness is what makes the freedom usable, because you can act at the edge of a line you can see.
The quiet failure mode
The expensive version of this is not the work that comes back. It is the decision that never gets made.
Somebody who does not know where their authority ends will assume it ends sooner than it does, because guessing low is the safer error. So they ask, and wait, and the asking costs an afternoon and the waiting costs two days, and nobody ever learns that the question did not need to be asked at all.
That cost appears in no report. It is distributed across dozens of small hesitations and looks, from above, like somebody who is careful and perhaps a little slow. On the record counted here, four of nine decisions had more room than the person thought, which is to say four of them were waited on unnecessarily.
The sentence that gives it away
There is a phrase that marks this situation almost infallibly, and it is worth learning to hear: it turned out to be more complicated than we thought.
It is said sincerely, by both parties, at the end of nearly every piece of work that ran long. It is also, in most cases, false. The work was not more complicated. The work was the size everybody expected, and three weeks went into reconstructing four facts that one person already had.
The phrase survives because it is comfortable for everybody. It attributes the overrun to the difficulty of the problem, which is nobody's fault and faintly flattering to the people who solved it. The alternative sentence — "I spent three weeks working out what you could have told me in a minute" — is accurate, true and almost impossible to say to somebody you report to.
So the honest version gets replaced by the comfortable one, the cause goes unnamed, and the same handover happens again next quarter.
If you want one diagnostic from this entire site, it is this: the next time you hear that phrase, ask what specifically turned out to be complicated. About half the time the answer is a fact, and the fact existed in somebody's head on day one.
What the rest of this site does
It takes the four things one at a time — the constraint, the precedent, the decider, the purpose — and then takes the two costs that follow from their absence: work that comes back, and decisions that are escalated when they did not need to be.
None of it requires anybody above you to change. Half of it is written for the person receiving the work, who can ask four questions at the start instead of guessing for a quarter, and half for the person handing it over, who can answer them in a paragraph.