Skip to content
Left to Guess All fifty notes

Trust and autonomy, measured in what was not said

Nobody drew the line

You were trusted. You were not told where your say ends, what has been tried before, who can overrule you, or why the work is wanted. Four things make a handover usable, and on the quarter counted here the median arrived with two of them.

When responsibility is handed over without enough context, employee monitoring software can help a team compare project time and workload before treating delay as an individual failure.

For an independent perspective on management and working practice, consult SHRM.

123456
Where you thought it endedWhere it actually endedMore room than you thoughtLess room than you thoughtExactly right
50notes for somebody who was trusted and not told
6parts, from what autonomy needs to how to hand work over properly
32slots of context counted across the handovers on this site, four to each
12of those slots arrived before the work started, counted from the record

Nothing here is a benchmark. Every figure on this site is one person's record of one quarter, written down at the time, and the only claim made for it is that it happened. No person and no employer is named anywhere. Nothing in this collection is employment advice: what an employer may require, how work is assessed and what authority a role carries differ by jurisdiction and by contract, and they change.

Six parts, from what was missing to how to supply it

Every note belongs to one of them. The line at the foot of each card is what that part cost on the record counted here, stated as a thing that happened rather than as a principle.

What autonomy needs

What autonomy needs

The four facts that turn permission into something a person can act on, and what each one costs when it is missing.

one handover of eight that had all four

Where your say ends

Where your say ends

The edge of your own authority, which almost nobody is told and almost everybody guesses too low.

four decisions escalated that did not need to be

What was not said

What was not said

The context that never arrived, the context that arrived in week three, and why the people holding it cannot see that it is missing.

twenty of thirty-two slots guessed or late

The cost of guessing

The cost of guessing

Work that came back, decisions that were escalated for nothing, and the quarter that produced both.

six pieces of work that came back

Asking

Asking without losing face

Four questions, the wording that gets them answered, and the eleven days people spend working up to a four-minute reply.

eleven days of hesitation, four minutes of answer

Handing work over

How to hand work over

Four sentences in the message that assigns the work, and a fifth that costs fifteen seconds and saves more than the other four together.

ninety seconds against three weeks

One person's quarter, counted three ways

Three panels kept by the same person over the same three months: where they thought their say ended and where it did, what arrived with each piece of work, and how far the work went back when it came back. Nothing here is a benchmark. It is one record, and the only claim made for it is that it happened. No person and no employer is named anywhere on this site.

Nine decisions, where you thought your say ended and where it actually ended

What the decision wasWhere you thought it ended, and where it didSteps
123456

Changing the wording on a customer-facing error message

4 → 1

It had been yours all along, and nobody had said so

Refunding a customer under fifty pounds

3 → 1

There was a written policy you had never been shown

Picking the library for the new parser

1 → 4

You chose and shipped; the architect had a standing view

Rescheduling the weekly review

2 → 2

Dropping a requirement nobody had asked about in six weeks

2 → 5

It was a commitment made to a customer before you joined

Hiring a contractor for two days

5 → 3

Your own budget line covered it

Changing a database column type

1 → 1

Telling a customer a date

3 → 6

Dates are committed by one person here, and it is not you

Reordering your own week

4 → 1

You asked about this twice before believing the answer

Where you thought it endedWhere it actually endedMore room than you thoughtLess room than you thoughtYou had it exactly right
  1. 1Decide it alone and say nothing
  2. 2Decide it alone and mention it afterwards
  3. 3Check with one colleague first
  4. 4Check with the person who assigned it
  5. 5Needs a decision one level up
  6. 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.

Eight handovers, and which of the four things arrived with each

What was handed overThe constraintWhat was tried beforeWho can say noWhy it is wantedGiven

Rewrite the import so it does not time out

2 of 4

Produce the quarterly numbers for the board

1 of 4

The audience arrived in week three

Fix the signup flow

0 of 4

Four lines of chat and nothing else

Move the scheduler onto the new queue

4 of 4

The one complete handover of the quarter

Write the migration guide

2 of 4

Take over the vendor relationship

0 of 4

Decide what to do about the flaky test

1 of 4

Spec the reporting page

2 of 4
Given at the start: 12Guessed, and never confirmed: 15Arrived after it was needed: 5

eight handovers · 32 slots in all · the median arrived with two of the four · one was complete · two arrived with none of the four

A ninth line was given to this panel and is not drawn. It names three of the four slots. The strip compares handovers against each other across the same four parts, and a row of three would read as a shorter handover rather than as an incomplete record.

Six pieces of work that came back, and how far back they went

What the work wasHow far it got, and how far back it wentBack

The import rewrite

2 stages back

It went back because the constraint on downtime was never stated. It had been agreed with a customer two years earlier

The quarterly numbers

4 stages back

It went back because nobody had said who the audience was. Rebuilt from scratch for a different reader

The signup flow

1 stage back

It went back because a rule existed that nobody mentioned.

The migration guide

2 stages back

It went back because it was written for the wrong reader.

The reporting page spec

2 stages back

It went back because the purpose turned out to be a different one.

The vendor handover

1 stage back

It went back because a previous attempt had failed for a reason nobody passed on.

How far the work gotThe way back, and where it landedA stage it never reached
  1. 1the brief
  2. 2the plan
  3. 3the build
  4. 4review
  5. 5shipped

six pieces of work came back · the median went back two stages · the furthest went back four · two of them went all the way to the brief

Two further lines were given to this panel and are not drawn. One returns from a stage later than the one it reached, which cannot happen and would draw the arc forwards. The other has no readable return point, and the arc would have nowhere to land.

The two sentences

Almost every handover of work contains two sentences. The first is the task. The second is some version of "you own this — run with it."

The second sentence is meant warmly and it is usually sincere. It hands over something real: the authority to decide how the work gets done. What it does not hand over is any of the information that makes that authority exercisable, and the person saying it cannot see that there is anything else to hand over, because they already have all of it.

This site is about the gap between those two facts. It is not an argument against autonomy; it is an argument that autonomy arrives in two parts and that one of them routinely goes missing.

What the second part consists of

Four things, and they are the same four every time:

  • The constraint. What must not change, with a number on it, and whatever the number is attached to — a contract, a renewal, an outage somebody remembers.
  • The precedent. Whether this has been attempted before, and the reason it stopped. Not a verdict; a fact about the past.
  • The decider. Who holds a standing view on this area, whether or not they are in the room, and whether their view is a veto or an opinion.
  • The purpose. Who asked for the work, and what they will do with it. Not the version written in the plan — the specific one.

Each takes about a minute to say. Each takes between three days and three weeks to reconstruct from evidence. That ratio is the whole economics of this subject, and the reason it never corrects itself is that the minute is paid by one person and the three weeks by another.

What happens in the first hour

Watch somebody receive a thin brief and the first hour is not spent on the work. It is spent on four pieces of detective work running in parallel.

They look for the constraint by proposing something small and watching the reaction. They look for the precedent by searching for the nouns in the task and hoping a stale branch turns up. They look for the decider by reading who commented on the last change in this area. They look for the purpose by trying to reconstruct it from the shape of the request.

None of that is visible from outside. From outside it looks like somebody getting started slowly. And none of it produces facts — it produces four beliefs, which the work then gets built on.

The person is not being obstructed and is not being tested. They are doing archaeology on a decision that was made in a meeting they were not in, and the archaeology is the part that takes the three weeks.

The two costs, one visible

When the beliefs turn out to be wrong, the work comes back. That is the visible cost: it has a date, an artefact and a conversation, and everybody involved knows it happened. On the quarter counted on this site, six pieces of work came back, the median going back two stages and two of them going all the way to the brief.

The invisible cost is larger. It is every decision somebody escalated that they did not need to escalate — the question asked, answered yes, and forgotten by both parties within the hour. Four of nine decisions on this record had more room than the person assumed. Each one was an afternoon of waiting for permission that already existed, and not one of them was ever corrected, because correcting it would have required somebody to say "you did not have to ask me that".

Those two costs have the same cause, which is the most useful thing on this page: a fact existed, somebody held it, and it did not arrive. Whether that produces a delay at the start or a return at the end depends only on whether the person decided to ask or decided to proceed.

How this site is arranged

Six parts. The first sets out the four things and what each one does. The second is about the edge of your own authority, which almost nobody is told and almost everybody guesses too low. The third is about the context that did not arrive, and why the people holding it cannot see that it is missing. The fourth is the cost — rework, and the escalations that leave no trace. The fifth is about asking, which is the cheapest fix available and the one people avoid for reasons worth examining. The sixth is for whoever hands work over, and it is five sentences long.

Half of what follows can be done by the person receiving the work, without anybody above them agreeing to anything. Asking four questions at the start produces the same four facts as somebody writing four sentences, and it does not need their cooperation to begin.

05 / 06eleven days of hesitation, four minutes of answer

Asking without losing face

Four questions, the wording that gets them answered, and the eleven days people spend working up to a four-minute reply.

06 / 06ninety seconds against three weeks

How to hand work over

Four sentences in the message that assigns the work, and a fifth that costs fifteen seconds and saves more than the other four together.

The short version

Say where the line is before anybody has to guess

Four things make a handover usable: the constraint, what was tried before, who can say no, and why it is wanted. They cost a minute each to state and a quarter to guess. On the record counted here the median handover arrived with two of them, six pieces of work came back, and four decisions were escalated that did not need to be.

Practical comparisons

Tools for clearer handoffs

Three differently scoped guides for documentation, learning, and remote operations.