Rework, and Who It Belongs To
Hours spent doing something twice. Where they should be charged, why they usually are not, and what the figure is worth once it exists.
Rework is hours spent doing something that was already done, and it belongs on the job that generated it. In practice it is charged almost anywhere else — to a standing code, to the next job, to nothing at all — because charging it honestly makes a completed job look worse after it closed.
The commercial issue in “Rework, and Who It Belongs To” becomes easier to audit when estimated work, recorded hours and approved changes remain distinct. Teams considering read the complete workflow summary for how to detect mouse jigglers can strengthen the project-time evidence without treating a software report as a substitute for the quote, contract or accepted invoice.
The consequence is that the operation cannot say what rework costs, which means it cannot decide whether preventing it would be worth anything.
For a separate reference relevant to “Rework, and Who It Belongs To”, consult BLS productivity and costs data. Compare the general material with the organisation’s actual contracts, accounting policy and jurisdiction.
Why it hides
Three reasons. The job is closed and cannot take more hours. The person doing the rework did not cause it. And somebody's estimate will look wrong if the hours appear.
- The job is closed and cannot take more hours.
- The person doing the rework did not cause it.
- Somebody's estimate will look wrong if the hours appear.
All three are organisational rather than technical, and the third is the real one.
What it costs
In most operations with no measurement, between five and fifteen per cent of chargeable capacity. In operations that have measured it, the figure is usually at the upper end and surprises everybody.
It is the largest single recoverable inefficiency in most labour operations and the least visible.
Where it should go
On the original job, with a rework flag, even after that job has closed. The job's final cost is then its true cost.
This requires jobs to accept hours after closing, which most systems allow and most policies forbid.
The alternative that works
A rework code per job: hours go to the job, flagged, and the flag lets the job's cost be reported both ways — as delivered and as originally estimated.
Both figures are useful and they answer different questions.
Rework caused elsewhere
Where another team or a supplier caused it, the hours still belong to the job, and the cause belongs in a separate field.
Charging it to the causing department is a different mechanism and it turns a measurement into an internal dispute. Record the cause; do not cross-charge unless the organisation genuinely works that way.
Customer-caused rework
A change of mind, a late specification, a wrong input. These are chargeable in most commercial arrangements and frequently not charged.
Measuring them is the precondition for charging them, and in most operations the measurement alone recovers more than any pricing change.
The flag that makes it possible
One field on the hour entry: is this rework, and what caused it. A short list of causes: specification, our error, supplier, customer change, unknown.
Five values. It is the single most informative field a timesheet can carry after the job number.
What the data shows first
That rework is concentrated. A small number of job types, or a small number of causes, account for most of it.
Concentration is good news: it means the problem has an address.
Why it is not a performance measure
Because the person recording it is usually not the person who caused it, and because a rework flag that affects anybody's appraisal stops being used within a quarter.
It is a measure of a process, and saying so when it is introduced is what makes the data survive.
What the figure makes arguable
Not about blame. About whether a check earlier in the process would cost less than the rework it prevents, which is an arithmetic question once the rework figure exists.
Without the figure, the same conversation is a matter of opinion and the status quo wins.
Rework in the estimate
Once measured, it belongs in the estimate for the job types that reliably produce it. A class of work that runs at twelve per cent rework should be estimated with it rather than surprised by it.
That is not accepting the rework; it is pricing honestly while the cause is worked on. The two are separate activities with separate owners and separate timescales.
When rework is somebody else's revenue
In some arrangements the customer pays for it, which removes the financial pressure to reduce it and leaves only the capacity cost.
That capacity cost is real: hours spent on rework are hours not spent on new work. An operation that is comfortable with chargeable rework is usually one that has not counted what it displaces.
The word itself
Some operations avoid it, preferring "additional work" or "post- delivery support", because rework implies somebody got it wrong.
The euphemism costs the measurement: once the category is vague enough to include legitimate extra scope, the figure stops meaning anything. Keeping the word and keeping the cause field is what preserves both the accuracy and the distinction.
The rework table
Rework hours as a share of chargeable hours, split by cause, by period. One small table.
It is also the table that justifies a quality investment, which is otherwise argued for on principle and declined on cost.