Warranty recovery as a line item: the money a small solar shop leaves on the table

Warranty recovery is the parts and labour a manufacturer pays back to you, against the cost of the visit that produced the claim. Almost no small solar company tracks it as a line, so the number is unknown rather than small. This post gives you the arithmetic. It does not give you the answer, because the answer is made of your rates, not ours.

We are not going to quote an industry average recovery figure, a typical claim value, or a percentage of revenue. Those numbers get invented constantly and none of them describe your business. What follows is the structure to put your own numbers into.

Why it is not a line item today

Warranty work does not arrive as a project. It arrives as a phone call in the middle of an install week, gets absorbed by whoever is nearest, and disappears into general overhead.

Three things follow.

The cost is invisible, because the truck was going out anyway and nobody costed the two hours it lost. The recovery is invisible, because a replacement part that arrives free feels like an absence of a bill rather than income. And the leakage is invisible, because a claim you never filed leaves no trace anywhere.

So the honest starting position for most three truck shops is not "we recover too little". It is "we do not know", and unknown is the state that lets the number stay bad indefinitely.

The four places recovery lives

1. Parts covered by the manufacturer. The largest bucket by value, and the one most companies do capture, because a free inverter is hard to miss.

2. Labour reimbursed by the manufacturer. Rare, and worth knowing exactly. Most brands pay nothing.

3. Term preserved by meeting a condition. The bucket nobody counts. When you reconnect a GoodWe inverter to SEMS, you moved that unit from a five year term back to a ten year one. No money changed hands today. You removed a future cost that would have landed on you, because the homeowner calls you, not the manufacturer.

4. Cost legitimately passed to the customer. Out of warranty work, priced as work. This one leaks when nobody can prove the term ended, so it gets absorbed to avoid an argument.

Who actually pays labour

This is the part worth knowing cold, because it decides whether a truck roll is a cost or a partial recovery.

Read the twelve month rows carefully. Twelve months of labour cover means year one. A failure in year six is a parts claim with the labour on you, which is the normal case across this catalogue.

Enphase is the one brand with a stated cash rate. In the United States, Enphase reimburses $200 per truck roll plus $25 per replaced microinverter, and a paid Labor Protection Program extension stretches labour cover to five or ten years from activation. Whether that extension pays depends entirely on your real truck roll cost against the reimbursement, which is a calculation and not a guess.

The per claim arithmetic

Write it as a worksheet and fill in your own values.

Cost of the claim to you

C = (drive time + on site time + return visit time) x loaded labour rate
  + vehicle cost for the distance
  + parts you fund yourself
  + admin time assembling and chasing the claim x office rate

The admin line is the one that gets left out and it is not small. A claim that goes quiet and needs three follow up calls has an hour of office time in it.

Recovery on the claim

R = value of parts supplied under warranty
  + labour reimbursement, where the brand pays one
  + any amount legitimately invoiced to the customer

Net position

N = R - C

For most brands, on most failures outside year one, the labour reimbursement term is zero. That is not a scandal, it is the market. It is also the reason the truck roll cost is the number that decides everything, and why guessing at it is expensive.

Work out your real per visit cost first, with your own drive times and loaded rate, using the truck roll cost calculator. Everything downstream depends on that one figure.

The per quarter arithmetic

Per claim numbers tell you whether a job was worth doing. Quarterly numbers tell you whether the process is working.

A = claims attempted in the quarter
P = claims paid
D = claims denied or abandoned
X = failures where a claim was never attempted

Recovery       = P x average R
Absorbed cost  = (D + X) x average C
Success rate   = P / A
Leakage        = (D + X) x average R

X is the number that matters most and the one nobody has. It is the count of failures where the claim was not attempted, because the serial was wrong, the nameplate photo was never taken, the install date could not be evidenced, or nobody could tell whether the brand's condition had ever been met.

Every one of those is a records failure rather than a technical one, and every one converts a recoverable cost into an absorbed cost silently.

The four numbers to start collecting on Monday

You cannot compute any of the above without these, and none of them require software to begin.

1. Cost per warranty visit. Even a rough figure, applied consistently, beats no figure. Use the calculator once and reuse the result.
2. Claims attempted, paid, denied. Three counters. A sheet is fine.
3. Denial reason, in the manufacturer's own words. Not "rejected". The actual reason. This is the field that tells you what to fix.
4. Failures where no claim was attempted, and why. The hardest to collect because there is no natural moment to record it. Ask the question at the end of every service visit: was there a claim here, and if not, why not?

Run those four for one quarter and you will have a real recovery number. It will also be the first time anyone in the business can say what warranty work costs.

What the denial reasons will tell you

Predictably, they will not mostly be about whether the fault was covered. They will be about proof: a serial that does not match a unit the manufacturer made, an install date that cannot be evidenced, a condition that was never confirmed.

That is good news, because those are all fixable on the day of installation at a cost of about a minute per unit. Photograph the nameplate. Store the serial against the unit. Record the confirmation number. See the warranty claim checklist for what each brand's claim actually asks for, and the claim documentation post for the short list.

Two decisions this number lets you make

Whether to buy a labour extension. The Enphase Labor Protection Program stretches labour cover to five or ten years from activation, at a price. With a real truck roll cost and a real claim rate on Enphase systems, that becomes an arithmetic question. Without them it is a vibe.

What to fit next. Labour terms differ enormously between brands, from REC's 20 years on the Alpha Pure-R line to nothing at all across most of the catalogue. Compare them on the inverter and battery pages. If two products are close on price and one of them puts every future service visit on your account, that belongs in the purchasing decision.

Where the honest limits are

Two of them, stated plainly.

The first is that recovery is not free money. Every claim costs you a visit and an hour of admin, and on most brands outside year one the labour is yours. A high recovery rate is worth having because it stops absorbed cost, not because it is a revenue line.

The second is about our own product. Claim Timer holds one record per piece of equipment, tracks each brand's conditions, and builds the evidence pack when a part fails, which is what closes the gap on X. It logs denial reasons on the plans that carry them. What it does not do today is report cost across a quarter, and it does not file the claim with the manufacturer for you. That first one is the reporting this post describes, and it is not built yet. When it is, we will say so here.

Until then the spreadsheet works, provided somebody actually keeps it. Start with the truck roll cost calculator, pick one brand you fit constantly, and count one quarter.