83% of HVAC businesses raised prices in the past 12 months, according to Jobber’s 2026 Home Service Trends Report. Inflation and material costs drove 68% of those increases, labor costs drove 52%. Here is the uncomfortable follow-up question: after all those increases, which of last week’s jobs actually made money? Most shops with 3 to 30 trucks cannot answer it. They know revenue. They know the bank balance. They find out margin in April, when the accountant closes the year.
That gap has a name: job costing. Big commercial outfits have done it forever. Residential service shops mostly have not, because stitching timesheets, material receipts, and invoices together by hand takes hours nobody has. In July, that excuse got weaker. Two of the three big field service platforms shipped job-level cost tracking aimed squarely at shops your size. The workflow is now a settings toggle and a weekly 20-minute habit, not a spreadsheet project.
Price Hikes Papered Over the Margin Leaks
A price increase raises every ticket. It does not tell you which tickets still lose money. The leaks hide in specific places: a water heater swap quoted at four hours that always takes six, a maintenance visit priced in 2024 against 2026 material costs, one tech who burns an extra 40 minutes per call, callbacks that get logged as new jobs instead of warranty work. Averages bury all of it.
The math is not small. On $1.5 million in annual revenue, a single point of gross margin is $15,000. Shops that start tracking costs per job routinely find a job type running 10 to 15 points below the house average. That is not a rounding error. That is a truck payment, or the recruiting budget you said you could not afford.
Service Nation’s 2026 Trend Report, covered by ACHR News in late July, lists rising operational complexity and economic pressure among the seven forces squeezing home service contractors this year. Translation for the field: your costs move faster than your price book, and the only defense is seeing margin while the job is still warm, not eight months later. Weekly beats quarterly. Quarterly beats April.
July’s Releases Put Job Costing Within Reach
Jobber shipped job costing updates on July 7, letting you track what you spend on each job, labor and materials against the quoted price, inside the same app your techs already use for scheduling. On July 24 it followed with schedule updates that make rework visible: sortable unscheduled appointments, bulk shifts, and client notifications when visits move. Job costing lives on Jobber’s Grow tier, which runs about $199 a month. If your crews already run Jobber, the feature is sitting there unturned-on.
Housecall Pro went further on July 15 and launched trade-specific packages for HVAC, plumbing, and electrical shops. Each package ships preconfigured with industry workflows plus three tools that matter for this conversation: pricing benchmarks built from over 100 million jobs, technician performance reporting, and membership management. The benchmarks give you an outside reference for what a job should cost. The tech reporting shows you who beats the estimate and who blows it. Mid-tier Housecall Pro plans run roughly $150 a month, with the trade packages priced by tier.
ServiceTitan remains the heavy option for shops at the top of the 3-to-30-truck range. Its job costing is mature and its dispatch AI is real, but expect a per-tech price that operators peg in the $250 to $400 a month range and an implementation measured in months, not days. The honest read: if you are under 10 trucks and not already on it, you do not need ServiceTitan to start costing jobs this month. You need the toggle in the software you already pay for.
The Weekly Margin Review, Step by Step
The stack only works if the data goes in, so start with capture. Techs clock in and out of each job on their phone, in the app, not on a whiteboard back at the shop. Materials get attached to the job the day they are bought: a purchase order if you have supplier integration, a photo of the receipt if you do not. Both Jobber and Housecall Pro handle receipt capture on mobile. Enforce it for two weeks and it becomes muscle memory.
Then run the review. Every Monday, 20 minutes, before the trucks roll. Pull every job closed in the last seven days. Sort by gross margin percent. Ignore the winners. Look at the bottom five and ask one question per job: was it the estimate, the tech, the materials, or the price book? Write the answer down. Patterns show up by week three, and they are rarely what you guessed.
Last, close the loop. A price book line that keeps losing gets repriced or retired. A job type that always runs over gets its standard hours changed in the estimate template. A tech who consistently beats book time gets asked how, and the answer gets taught at the next meeting. This is the part most shops skip. The review is diagnosis. The price book edit is the cure.
Here is what the review looks like in practice. A five-truck plumbing shop sorts last week’s 41 closed jobs and finds water heater replacements sitting at 31% gross margin against a 48% house average. The estimate template books four hours. The timesheets say the real number is five and a half, because haul-away, permit paperwork, and the supply house run were never in the book. One edit to the template and a $40 price book bump later, the next ten water heaters come back at 46%. Twenty minutes of review, one fix, roughly $170 recovered per job. That is the whole method.
One warning from the field: do not weaponize the numbers in week one. Early data is dirty. Techs forget to clock out, receipts land on the wrong job, and a witch hunt over bad data will kill the capture habit you just built. Spend the first month fixing data quality and price book lines. Talk to techs about patterns in month two.
The Decision to Make Before Labor Day
August is the right month for this. Peak cooling season winds down, the fall shoulder is coming, and shoulder season is when thin margins turn negative. A shop that starts capture this week has four to six weeks of clean data before the slow stretch, enough to reprice the worst offenders before October instead of discovering them in April.
So make one decision this month: pick the system where your job data already lives and turn on job costing there. Jobber shop, upgrade to Grow and switch it on. Housecall Pro shop, look hard at your trade’s new package and its pricing benchmarks. ServiceTitan shop, you already own the tools, so schedule the Monday review and actually run it. Switching platforms is not the assignment. Using the margin features your current platform shipped this summer is.
The signal to watch over the next 30 days: whether the platforms push benchmark data further into estimates themselves. Housecall Pro’s 100-million-job dataset pointed at pricing is the start of software that flags a losing quote before you send it. The shops feeding clean cost data into these systems now will be the ones whose estimates get smarter first. The shops still running on gut feel will keep raising prices and keep wondering where the money went.
Atlas Unchained tracks the operational side of AI for the trades every Thursday. If a weekly margin review sounds better than an April surprise, subscribe and get the next play in your inbox.
