Comeback
Comeback explained: what counts as a comeback in a service department
A comeback is a vehicle that returns to the workshop because the original repair did not fix the customer's concern, or because the work itself was wrong or incomplete. It is the one job in the shop that is guaranteed to lose money, because the labour has already been sold once and the bay has to be found twice.
Also called a repeat repair, a rework, or a redo. In field service the same event is a callback, and the inverse measure there is first-time fix rate. The workshop version differs in one practical way: the customer has to drive back to you, so every comeback is also a diary slot, a courtesy car, and a conversation at the desk.
What counts as a comeback, and what does not
Getting this boundary wrong is how service managers end up coaching the wrong department. Separate the categories at the point of logging, not at month end:
- Misdiagnosis: the repair was performed correctly on the wrong fault. The customer's symptom is unchanged.
- Workmanship: the right job, done wrong. A clip left off, a bolt under torque, a connector not seated.
- Incomplete work: something on the job card was not finished, usually because the car was needed back on the drive.
- Parts: the part was not available, so the vehicle went out unfixed and returned to have it fitted. That is a stocking failure wearing a technical costume.
- Early component failure: a new part failed. Supplier problem, and it belongs in a warranty claim, not in the technician's numbers.
- Not a comeback: a genuinely new fault on the same car, and a customer who returns because nobody explained what was done. The second one is a communication cost.
How is comeback rate calculated?
Comeback rate equals comeback repair orders in the period, divided by total repair orders in the period, multiplied by 100.
Two settings decide what the number means. The first is the window: how long after collection does a return still count, 14 days or 30. The second is the trigger: same concern only, or any return on the same vehicle. Shorten the window and the rate falls without a single extra job being done properly, which is worth remembering before anyone celebrates a trend.
Comeback rate: a worked example
A workshop writes 940 customer pay repair orders in a month and logs 22 returns inside 30 days, giving a comeback rate of 2.3 percent. Split them and the picture changes: 9 were misdiagnosis, 7 were parts return visits, 4 were incomplete work, and 2 were damage caused in the bay. Technical retraining addresses 15 of the 22. The other 7 are a parts department conversation, and no amount of technician coaching will move them.
What a comeback actually costs
The second visit is rarely billable, so the hours land as internal work or as a goodwill repair. That is one loss. The bigger one is that a paying job was displaced from the diary to make room, which quietly dents technician efficiency for whoever takes the redo.
Then there is the customer. The J.D. Power 2025 U.S. Customer Service Index Study found that 12 percent of repairs are not completed correctly on the first visit, with the two leading reasons being that the work did not correct the problem (30 percent) and that necessary parts were not available (28 percent). Among those customers, only half said they returned or planned to return to the dealership at all (J.D. Power). A comeback does not just cost you the hours. It costs you roughly a coin flip on the whole relationship, and it drags CSI down on the way out.
How comeback numbers get hidden
- Written as a new repair order with a new concern. The return disappears from the count entirely. This is the most common distortion and it is usually accidental.
- Tied to technician pay. When a comeback costs the technician money, it gets reclassified. You lose the data and keep the problem.
- Handed to a different technician. The redo gets fixed, and the person who missed it never finds out what was missed.
- Read as an average. One model, one job type, or one shift usually carries most of the returns. Segment before concluding anything.
Where comebacks are actually prevented
Two moments carry most of the risk. The first is the write-up, where a vague customer concern ("noise when braking") gets copied onto the job card and sent to the ramp without enough detail to reproduce. The second is completion, where what was done is not documented well enough for the customer, or for the next technician, to understand it.
Both are documentation problems more than technical ones. A quality control road test on anything drivability related, a photo of the finished work on the job card, and a parts check before the car is booked in will move a comeback rate further than another training day. The multi point inspection helps here too, because a properly graded sheet records the condition of everything the technician did not touch, which is what settles the argument when a car returns three weeks later with an unrelated fault.
