Capturing the Evidence a Warranty Approval Requires
Why this matters
The failed part is only half of a warranty claim. The other half is the evidence, and you get one chance to capture it: on the visit, before the old part hits the scrap bin and the unit is buttoned up and running. Thin evidence is the top reason a covered part gets denied and lands on the shop's invoice instead of the maker's. This is the field checklist for proof that survives a reviewer who was never on site.
What the reviewer is actually checking
A warranty reviewer approves three things, in order. Your evidence has to answer all three:
- Is it covered - right product, in warranty by date. Proven by the nameplate, serial, and date code.
- Did it actually fail - a measured fault, not an opinion. Proven by a reading.
- From a covered cause - an internal defect, not surge, water, neglect, or a prior bad repair. Proven by the failure's appearance and context.
Miss any one and the file is deniable no matter how obvious the failure was in person.
The evidence stack
| Capture | What it proves | Common miss |
|---|---|---|
| Nameplate, serial, date code (legible, one frame) | Coverage and age | Blurry or cropped so the serial is unreadable |
| Part installed, as found | Context and correct application | Photo taken only after removal |
| Meter reading with the part in frame | The actual fault | A number with nothing tying it to this part |
| The failure itself, close up | Covered cause | Only a wide shot that hides the burn, split, or leak |
| The retained failed part, tagged | Chain of custody | Part scrapped before authorization |
Capture the number, not the adjective
"Weak," "low," and "noisy" mean nothing to a reviewer. A microfarad value next to the nameplate rating, a winding resistance to each terminal and to ground, an element reading open, a pump pulling locked-rotor amps, a board with good input and no output - those are repeatable. Photograph the meter with the part identifiable in the same shot, or shoot the value and the part back to back so the pairing is obvious.
Retain the failed part like evidence
Many programs require the old part returned before they pay, and a scrapped part is an automatic denial. Bag or box it, tag it with the job or claim number, and note where it is. For a pressure or electrical part, that tag also protects you if the customer later asks what you pulled.
Timing is the whole game
Every item above is easy on the visit and nearly impossible after. Once the new part is in and the compressor, heater, pump, or panel is energized, you cannot re-read the dead part or re-photograph the burn in place. Build the capture into the diagnosis, not the cleanup: read, shoot, and tag before you remove, every time, so a later claim is never blocked by a photo you did not take.
Make it automatic
Standardize it so no tech has to remember: a short capture sequence saved on the job, the failed-part bin labeled, the claim number written on the tag at removal. The shops that rarely lose a claim are not luckier, they document the same way on every part whether or not they think it will need a claim, because the one you skip is the one that fails review.
References
- Manufacturer warranty documentation and return-authorization procedures (generic)
- Trade-standard practice for failed-part retention and photographic documentation
- See related: The Manufacturer Wants Proof Before Approving the Part (decision tree); Working a Tech Support Line Without Wasting the Visit