A home warranty dispatch is not a customer call with extra steps. It is a different business model wearing the same uniform: someone else sets the price, someone else authorizes the repair, and someone else decides thirty to sixty days later whether you get paid at all. Shops that treat third-party work like cash work — same technician, same urgency, same loose recordkeeping — usually discover the difference on the aging report about four months in, when they realize a fifth of their revenue is sitting unpaid and half of it is unpaid for reasons they could have prevented on site.
Third-party payer work covers more than home warranty companies. Property management groups, manufacturer authorized-service programs, insurance restoration referrals, and commercial facility vendors all dispatch the same basic way: a capped diagnostic fee, a fixed labor allowance, an authorization step before you exceed the cap, and a claim that can be denied on documentation alone. For a lot of appliance repair shops, HVAC companies, and overhead door outfits, it is a quarter to half of all revenue and the single largest line on the receivables report.
As of September 2026, the honest position on this work is that it is neither a trap nor a gift. It is a lower-margin, higher-volume channel with an administrative tax attached, and it is worth exactly what your discipline makes it worth. This guide covers the real arithmetic, when to take the dispatch and when to decline it, the authorization habits that decide whether a claim gets paid, and how to track the whole channel separately so you can make the call with a number instead of a hunch.
The arithmetic nobody runs before signing up
Most owners evaluate a warranty program by comparing the labor allowance to their hourly rate, decide the gap is tolerable, and sign. That comparison leaves out the three things that actually move the result: parts margin, administrative time, and drive time.
Here is an illustrative worked example — these are my own round numbers for the comparison, not published industry figures. Take one 90-minute repair on a mid-range unit, 25 minutes of drive each way.
On a cash job, you bill a $95 diagnostic that credits toward the repair, $220 of book labor, and a part that costs you $80 and sells for $140. Revenue is $455, parts cost is $80, so gross margin is $375. The truck is committed for about two hours door to door. That is roughly $187 of gross margin per truck hour, and the money is in the account that day.
On the same repair through a warranty payer, you collect a $65 diagnostic, a $125 flat labor allowance that does not care that the job ran long, and the part arrives from the payer's supplier — so there is no parts margin at all. Revenue is $190 and gross margin is $190. But the truck is committed for about two hours and fifteen minutes once you add the authorization call from the driveway and the claim submission back at the shop, and the money lands somewhere between 30 and 60 days out. That is about $84 of gross margin per truck hour.
Two conclusions fall out of that, and only one of them is the obvious one. The obvious one is that warranty work pays roughly half as well per hour. The less obvious one is that $84 per truck hour is still enormously better than $0 per truck hour, which is what an idle truck on a slow Tuesday generates. The question is never "is warranty work good." It is "what am I giving up to run this call right now."
| Line item | Cash job | Home warranty work order |
|---|---|---|
| Diagnostic / trip fee | $95, credited toward the repair | $65 flat, or a trade call fee the homeowner pays |
| Labor | $220 at your book rate | $125 fixed allowance regardless of hours |
| Parts | $140 retail on $80 cost — your margin | Payer-supplied or reimbursed at or near cost |
| Who approves the work | The homeowner, at the unit | The payer, by phone, before you exceed the cap |
| Days to payment | Same day | 30 to 60 days from a clean claim |
| Documentation required | Itemized charge, signature, photo if disputed | Model and serial, failure photos, arrival and completion timestamps, parts used, signature, authorization number |
| Who you chase when it does not pay | One homeowner | A claims department with a ticket number |
When the dispatch is worth taking
Third-party work earns its place in three specific situations, and it is worth being deliberate about all three rather than accepting everything that comes over the portal.
Filling capacity you were going to lose anyway. A truck with a 10 a.m. and a 3 p.m. and nothing between them has an hour of sellable time that expires at 3 p.m. A capped call in that window converts dead time into $84 an hour. That is the strongest case for this work and it is entirely about scheduling, not about the rate.
Route density. If you already have two jobs in a subdivision, a third call four streets over carries almost no drive cost, which is the single biggest lever on the arithmetic above. Cut the round-trip drive from 50 minutes to 10 and the same $190 job goes from $84 to about $114 per truck hour. Dispatchers who batch third-party calls geographically get a materially different result from dispatchers who run them first-in-first-out.
Predictable volume during your off-season. A shop with a brutal seasonal trough is buying stability, and stability has real value even at a lower rate. This is the same logic that makes commercial maintenance contracts attractive to overhead door companies: a smaller number that arrives reliably beats a bigger number that arrives sometimes.
When it quietly eats your margin
The failure modes are just as specific, and every one of them is a scheduling or discipline problem rather than a rate problem.
The expensive mistake is sending your best technician across town on a capped call while cash jobs wait. Your most productive tech is also the one who converts diagnostics into full-ticket repairs, and parking that capability on a fixed allowance for three hours is the most costly thing you can do with it. Capped work should go to the truck with the gap, not the truck with the highest close rate.
The second is treating cap overruns as goodwill. A job that runs to three hours on a $125 allowance is not a relationship investment, it is a $200 donation, and shops that do it habitually are subsidizing the payer's pricing model with their own payroll. If a job is clearly going past the cap, that is what the authorization call exists for.
The third is silent accumulation on the receivables side. Warranty claims do not bounce loudly the way a declined card does. They just sit. Per Intuit's small-business cash-flow research, late and unpaid invoices remain among the most common cash-flow problems owners report — and third-party claims are the version of that problem that is easiest to not notice, because nobody is calling you about them. The same collection discipline you would apply to unpaid customer invoices has to apply here, with the difference that the debtor is an organization with a process and a ticket number.
Authorization discipline is the whole game
Almost every denied claim I have seen traces back to something that was supposed to happen before the wrench came out.
The rule is simple and absolute: do not exceed the cap without an authorization number captured in writing. Not "they said it was fine on the phone." A number, the name of the person who gave it, and the timestamp, recorded on the job before you proceed. If the payer's portal issues a reference, capture that too. A verbal approval that exists only in your technician's memory is worth nothing at claim review, and nothing is exactly what you will be paid for the overage.
Three habits make that stick in practice:
- Call from the driveway, not from the shop. The authorization has to happen while the unit is in front of you, because the claims rep will ask what failed, what the model and serial are, and what the repair costs — and you want to answer with facts rather than a callback.
- Write the number into the job before continuing. Whatever system holds the job, the authorization number goes in a field on that job, not on a scrap of paper in the door pocket.
- Treat the cap as a hard stop for technicians. This is the same authority question as technician discount controls in reverse: a tech should be able to complete work up to the allowance without asking, and should be structurally unable to continue past it without an approval recorded.
The claim file a processor actually wants
Claims processors are not evaluating your craftsmanship. They are checking whether the file supports the money. Build the file on site, every time, in the same order:
- Model and serial number of the unit, read off the tag rather than recalled. This is the field that most often kills an otherwise clean claim.
- Photos of the failed component, plus a wider shot of the installation. Two or three images, taken before disassembly.
- Arrival and completion timestamps, which is why a system that stamps the job automatically beats a technician typing times from memory at 7 p.m.
- Parts used, by part number, including anything the payer supplied. If a part comes back defective, that record is also what supports a vendor warranty credit or defective return.
- The homeowner's signature confirming the work was performed, captured at the unit.
- The authorization number for anything beyond the cap.
That list is not busywork; it is the same evidentiary logic as a card dispute, where a contemporaneous, timestamped, signed record carries far more weight than a reconstruction. The IRS recordkeeping guidance for small businesses also makes the point that electronic records satisfy the same recordkeeping expectations as paper, which matters here because the only realistic way to capture six fields plus photos on every call is digitally, at the job.
A structured repair order workflow is the right mental model: the job does not close until the required fields are filled, because a closed job with a missing serial number is a claim you are going to lose.
Tracking it so the numbers tell you something
The single highest-leverage change most shops can make is structural, not procedural: stop billing warranty companies as if they were homeowners.
Set up each payer as its own on-account billing customer. The homeowner stays on the job record as the service location and the signer; the payer is who the invoice bills to. That one change gives you three things immediately. You get per-payer aging, so you can see that one company is consistently at 52 days while another pays at 28. You get a running balance per payer, so a claim that was denied and never resubmitted shows up as an open balance instead of vanishing. And you get clean reporting, because every third-party job is tagged to a payer account and can be filtered out of your cash-work numbers.
From there, run the comparison that actually decides the question: revenue and gross margin per truck hour, warranty versus cash, over a full quarter. That is ordinary job costing applied to a channel instead of a job. The SBA's guidance on managing business finances frames the discipline plainly — track income and expenses continuously and bill promptly — and a channel you cannot isolate in your reports is a channel you are not really managing.
Running it in IntelliDrive OS
The mechanics in IntelliDrive OS follow the structure above. Each warranty company or property management group is an on-account customer with its own credit sales and balance, so work orders bill to the account instead of the homeowner and the aging report shows you who is slow. The job record holds the homeowner's property and service history separately, so the next dispatch to the same address arrives with the last three visits attached.
Documentation capture is built into the transaction rather than bolted on. GPS location, a digital signature, and a timestamp are captured on every transaction and attached to the sale record, which covers the arrival and completion evidence a claims processor asks for. Parts consumed come off the truck's inventory by part number, so the claim's parts list and your stock count are the same data. Serial-number capture feeds warranty tracking, so when a payer asks whether the compressor you installed eleven months ago is still covered, it is a lookup by serial or receipt number rather than an archaeology project.
On the reporting side, sales and payment reports with CSV export let you separate payer accounts from cash customers and run the per-hour margin comparison yourself. And because pricing is $79 a month flat with unlimited users, adding the office person who submits claims does not cost more — which is not a small consideration in a channel where the administrative time is half the cost.
What to do this week
- List your third-party payers and pull the current open balance and oldest unpaid claim for each. If you cannot produce that list in ten minutes, that is the finding.
- Set each payer up as an on-account customer so future work orders bill to the account and age visibly.
- Write down your required-field checklist — model, serial, photos, timestamps, parts, signature, authorization number — and make it the close-out standard for every third-party job.
- Set the cap rule with your technicians in one sentence: no work past the allowance without an authorization number recorded on the job.
- Run one quarter of margin per truck hour, warranty versus cash. Then decide which programs you keep, which you keep only for fill-in work, and which you decline.
None of this makes third-party work into cash work. It makes it a channel you are running deliberately — filling gaps, keeping the drive short, documenting on site, and knowing within a day which payer is holding your money.
Related reading: On-Account Billing for Service Businesses, Warranty Tracking in a Service Business, and Unpaid Invoices and Collections.
For a complete machine-readable feature and pricing reference, see our LLM reference page.