The cost of guessing
What rework actually costs
Six pieces of work came back in a quarter, and every one of them went back because of something that could have been said in a sentence at the start.
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.
Where ownership crosses several roles, teams can review the platform in detail to make time and recurring bottlenecks visible enough for a useful operational conversation.
Teams can compare this approach with guidance available from APQC.
Six pieces of work that came back, and how far back they went
The import rewrite
2 stages backIt 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 backIt went back because nobody had said who the audience was. Rebuilt from scratch for a different reader
The signup flow
1 stage backIt went back because a rule existed that nobody mentioned.
The migration guide
2 stages backIt went back because it was written for the wrong reader.
The reporting page spec
2 stages backIt went back because the purpose turned out to be a different one.
The vendor handover
1 stage backIt went back because a previous attempt had failed for a reason nobody passed on.
- 1the brief
- 2the plan
- 3the build
- 4review
- 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.
Six pieces of work came back during the quarter counted here. The causes, in the words of the person who recorded them at the time:
- The constraint on downtime was never stated.
- Nobody had said who the audience was.
- A rule existed that nobody mentioned.
- It was written for the wrong reader.
- The purpose turned out to be a different one.
- A previous attempt had failed for a reason nobody passed on.
Six returns, six causes, and every cause is one of the four things from the first part of this site. Not one of them is a mistake in the work.
Reading the picture
Each row shows how far the work got and how far back it went. The lime bar is progress; the arc is the return.
The median went back two stages. The furthest went back four — from shipped to the brief, which is the whole thing again. Two of the six went all the way to the brief, which is the expensive case, because a return to the brief discards every decision made after it.
A return of one stage is ordinary and often healthy: work goes to review, review finds something, it goes back to build. That is review functioning. The returns worth counting are the ones that cross more than one stage, because those are not about the quality of the work at all.
The cost is not the redoing
The obvious cost is the time spent doing it twice. That is the smaller half.
The larger half is everything built on top of the wrong assumption. The quarterly numbers were not simply rewritten; a hundred small decisions about level of detail, framing and structure had been made for an analyst and had to be remade for a board. None of those decisions were visible as decisions. They were just how the document had come out.
Then there is the cost that never appears anywhere: the person's estimate of their own judgement. Work that comes back twice in a quarter teaches somebody that their reading of a brief is unreliable, and the behaviour that follows is caution — more checking, more proposals, less deciding. That is the mechanism that turns one bad handover into a permanently slower person.
What the picture does not say
Three honest limits on this sign, worth stating on the page that introduces it.
It records that work came back, not whether coming back was right. Some returns are good: the review caught something real. A high count is not automatically a bad quarter.
It does not distinguish a return caused by missing context from one caused by a genuine change of mind upstream. The reason field carries that, and the reason is a judgement made by the person recording it.
And it cannot see the work that should have come back and did not — which is a different and worse failure, and invisible to any instrument of this kind.
Reading the distance
The distance a piece of work travels back is more informative than whether it came back at all, and the five stages map onto five different kinds of problem.
| Went back to | What it means |
|---|---|
| Review | The work was wrong in a detail. Review working. |
| Build | Something in the implementation was wrong. |
| Plan | The approach was wrong — usually a missing constraint. |
| Brief | The work was about the wrong thing. |
| Nowhere | It was abandoned, which is a different record entirely. |
The boundary that matters is between plan and brief. A return to the plan discards an approach; a return to the brief discards the understanding of what was wanted, and everything downstream of it.
On the record here, two went to the brief, two to the plan and two to the build. The two that reached the brief account for considerably more than half the lost time, which is the ordinary distribution: the rare deep returns cost more than the common shallow ones.
Which gives a priority. Preventing single-stage returns means better review, better testing, more care. Preventing returns to the brief means one sentence at the start, and it is where almost all of the available saving is.
Why count it at all
Because it is the only cost of bad handovers that leaves a trace.
The unnecessary escalations leave none. The three weeks of reconstruction leave none; they look like the work taking a while. A return is the one failure that is visible to everybody, has a date, and can be asked about afterwards.
So it is the lever. One question, asked each time something comes back — "what would have had to be said at the start for this not to happen?" — turns the most visible cost into the most useful piece of feedback the person handing over will ever get.