Three kinds of hours
Revenue per Billable Hour: The Column ServiceTitan Names Without a Formula, and the Definition We Use Instead
A technician is paid for nine hours, stands in customers' homes for six, and bills five. Which one is the hour in revenue per hour? The answer decides whether the number tells you about pricing, dispatch or neither, and ServiceTitan's help center does not settle it.
November 13, 2026 · 9 min read
The short answer
Revenue per billable hour divides revenue by the labor hours that produced it. ServiceTitan lists a Revenue Per Hour column in its Business Unit Performance template, but we found no published formula. Our definition, labeled as ours: completed revenue divided by sold hours, shown beside revenue per on-job hour and per paid hour, so the gaps between them point at pricing or dispatch.
On this page
- The hour on the timesheet is not the hour on the invoice
- A column name, a definition of sold hours, and one legacy formula
- Our three-hour definition, labeled as ours
- One technician's month, worked with invented numbers
- What the gaps say about pricing and dispatch
- Six ways an hourly number gets miscounted
- Monday with three hourly numbers, and what we offer
The hour on the timesheet is not the hour on the invoice
Follow one technician through a Wednesday. Clock in at 7:00. Load the truck. First driveway at 8:10. A two-hour repair the pricebook says should take ninety minutes. A detour to the supply house. Lunch in the cab. Two more calls, one of them a free return visit. Back at the shop by 4:30, paperwork until 5:15. The company paid for every minute of that. The customers were billed for some of it.
Revenue per hour sounds like one number. It is at least three, depending on which hours sit underneath: hours paid, hours spent on the job, or hours sold on the invoice. Each version answers a different question, and the gaps between them are often more useful than any single figure. This page sets out what ServiceTitan publishes, what it does not, and the definition we use when we build a screen.
It belongs with the other explainers in our ServiceTitan dashboards section. Average ticket tells you what a visit earns; this tells you what an hour earns. If the question is whether next week's hours are already full, that is the capacity dashboard, a different screen with a different job.
A column name, a definition of sold hours, and one legacy formula
ServiceTitan's Business Unit Performance template, as its help center described it in November 2026, lists "Item Billable Hours, Revenue Per Hour, Billable Efficiency" and "Job Billable Hours" in its performance section. We did not find the formula for Revenue Per Hour in the articles we read. So we will not tell you how ServiceTitan calculates it, and you should be wary of anyone who does without showing a source.
What ServiceTitan does define is the raw material. Its sold hours article says "Sold hours, also known as billable hours, are the estimated labor hours tied to a pricebook item. Use sold hours to calculate labor costs, technician efficiency, and Job Costing." Note the word estimated. A sold hour is the pricebook's opinion of how long the work takes, not a stopwatch reading.
One formula is published, in the legacy Daily Sold Hour Tracker Detail guide: "Efficiency: ([Total Sold Hours] x [Technician splits]) / [Hours the technician spent on a job] *100." The guide adds, "If the billable efficiency percentage is above 100%, the technician completed work faster than expected." That is efficiency, not revenue per hour, but it uses the same two kinds of hours we will.
Our three-hour definition, labeled as ours
When we put this number on a screen, we show three versions side by side and label all three as DataCram's definitions, not ServiceTitan's. The first two come from fields ServiceTitan's API documents, as its Developer Portal listed them in November 2026, and the third from your payroll records: invoice items in the Accounting API and job timesheets and splits in the Payroll API.
| Measure, our definition | Revenue divided by | Fields it reads | The question it answers |
|---|---|---|---|
| Revenue per sold hour | Sold hours on the same invoice items | Invoice item total, soldHours, generalLedgerAccount, technicianId | Are we charging what the pricebook says an hour is worth? |
| Revenue per on-job hour | Time from arrival to done on each appointment | Job timesheet arrivedOn and doneOn, job splits split | How much does an hour in the customer's home earn? |
| Revenue per paid hour | Hours on payroll for the period | Your payroll records, matched by technician | What does the business earn per hour it pays for? |

Revenue here means income items on completed jobs, the same basis ServiceTitan's dashboard uses for Completed Revenue, so the numerator does not wander between screens. We exclude items not mapped to an income GL account, as ServiceTitan's dashboard does. And we keep free return visits in the hours, because the company paid for that time whether or not anyone invoiced it. You may choose differently; the point is to choose once and print the choice under the tile.
One technician's month, worked with invented numbers
An illustration. These numbers are made up to show the arithmetic, not drawn from any company. A technician works 21 days at eight paid hours, so 168 paid hours. Timesheets show 112 hours between arrival and done. The invoice items on his completed jobs carry 96 sold hours. Completed revenue, after splits, is $26,880.
- 1.Revenue per paid hour: $26,880 divided by 168 is $160.
- 2.Revenue per on-job hour: $26,880 divided by 112 is $240.
- 3.Revenue per sold hour: $26,880 divided by 96 is $280.
- 4.Efficiency, using ServiceTitan's legacy formula with a 100% split: 96 divided by 112 is about 86%.
- 5.Paid hours not spent on a job: 56, a third of the month.
Nothing in that list says the technician is good or bad. It says the company paid for 56 hours that never touched a customer, and that jobs ran about 16 hours longer than the pricebook expected. Those are two separate problems with two separate owners. The first usually belongs to dispatch and territory. The second belongs to the pricebook, training, or both.
What the gaps say about pricing and dispatch
Paid hours far above on-job hours
Drive time, waiting on parts, shop time, meetings. Look at routing and zones before you look at the technician. The general manager dashboard is where this shows up daily.
On-job hours above sold hours
Efficiency under 100%. Either the pricebook underestimates the work, or the technician needs help on a job type. Check by job type first.
Efficiency well above 100%
ServiceTitan reads it as work finished faster than expected. It can also mean the pricebook's hours are generous. Both are worth knowing.
Revenue per sold hour falling
Discounts, stale prices, or more low-priced items per visit. The hours did not change; what each one earns did.
An hour the company pays for and nobody bills is still an hour the company pays for.
This is why we resist putting a single revenue per hour tile on an owner's screen. One number invites one reaction, usually aimed at the technician. Three numbers invite the right question: which gap moved? A technician scorecard can carry the per-person versions, as our technician scorecard page describes, while the owner sees the company line and the gaps.
Six ways an hourly number gets miscounted
- Splits left at default. ServiceTitan's glossary says "By default 100% of the technician split will go to the first technician assigned to the job." The helper on a big install then shows hours and no revenue.
- Pricebook items with no sold hours. ServiceTitan's dashboards setup article lists pricebook items linked to GL accounts with sold hours as a requirement. Missing hours inflate revenue per sold hour.
- Different business unit bases. ServiceTitan says revenue KPIs follow the invoice item's BU, while Business Unit Performance timesheet fields follow the job's BU. Filter by BU and the two halves of the fraction can disagree.
- Recalls dropped from the hours. Recall jobs are No Charge in ServiceTitan. Removing them flatters the number. Count them, and see our callback rate page.
- Adjustment invoices. Some ServiceTitan screens include them in revenue and some do not. Pick one and say which.
- Mixing job types. An hour of replacement work and an hour of maintenance earn different amounts by design. Compare like with like.
None of these says anyone's settings are wrong. They are choices, and ServiceTitan leaves most of them to the contractor. The mistake is making them silently and then arguing about the result. Our page on gross margin per job takes the same hours one step further, into what they cost.
Monday with three hourly numbers, and what we offer
- 1
Look up the column in your Reporting Dictionary
If ServiceTitan shows a calculation for Revenue Per Hour in your account, note it. Then decide whether to match it.
- 2
Audit sold hours on your top twenty pricebook items
Missing or stale hours poison every hourly number downstream.
- 3
Set splits deliberately
Agree how two-person jobs divide revenue before anyone compares technicians.
- 4
Read the gaps, not the totals
Paid to on-job is a dispatch question. On-job to sold is a pricebook question.
There is no target here. We found no sourced revenue per hour benchmark, and ServiceTitan's published API Terms, last updated April 15, 2026, restrict deriving benchmarks from API data, so our dashboards do not compare you with other companies. A custom ServiceTitan dashboard that shows all three measures is from $575 to build, depending on complexity, as of November 2026, with a fixed written estimate and hosting from $55 a month. The API integration goes live the same day.
The same terms limit storing API content to 24 hours except as they allow, and ServiceTitan's Developer Portal says customer-built apps need the "Works or Ent Plus package"; confirm both against your agreement. Our page on ServiceTitan API data explains the plumbing. For a standing conversation about what moved, the monthly numbers review is from $575 a month, as of November 2026, with no long contract. DataCram is a San Antonio web studio, building since 1999, not affiliated with ServiceTitan. The ServiceTitan alternative KPI page covers the same screen for companies leaving the platform.
Questions owners ask about ServiceTitan dashboards and numbers
We could not find out. ServiceTitan's Business Unit Performance template lists a Revenue Per Hour column, as read in November 2026, but the articles we read give no formula for it. Look it up in your account's Reporting Dictionary, which shows a calculation entry for each field. Until you have that, treat any revenue per hour number as its author's own definition.
ServiceTitan defines sold hours, also called billable hours, as the estimated labor hours tied to a pricebook item. Paid hours are the hours on payroll. A technician's paid hours include drive time, shop time and waiting, so they almost always run higher than sold hours. The gap between them is a dispatch and routing question as much as a performance one.
Show both. Revenue per sold hour tests pricing against the pricebook. Revenue per on-job hour, from arrival to done on the timesheet, tests how much an hour in a customer's home earns. Their ratio is close to ServiceTitan's legacy efficiency formula. One number alone tends to get blamed on the technician; two show which part moved.
ServiceTitan's legacy Daily Sold Hour Tracker guide says that if billable efficiency is above 100%, the technician completed work faster than expected. It can also mean the pricebook's sold hours are generous for that job type. Check a handful of jobs by type before deciding which explanation fits your company.
Yes, using fields ServiceTitan's API documents: invoice items carry soldHours, total and technicianId, job timesheets carry arrivedOn and doneOn, and job splits carry each technician's split. The result is your definition, labeled as such, not ServiceTitan's. The API connection goes live the same day, under ServiceTitan's package and storage rules.
Keep reading
- Dashboards by role: techniciansA Fair Technician Scorecard on ServiceTitan Data: Five Numbers and the Arguments They StartClose rate, average ticket, revenue per hour, recalls and memberships per technician, using ServiceTitan's definitions and comparing like with like.
- Revenue minus what it tookGross Margin per Job: What Belongs in Direct Costs, What the IRS Means by Gross Profit, and Why the Two DifferRevenue minus direct costs, job by job. ServiceTitan's job costing formulas, the IRS gross profit definition, and the costs that quietly go missing.
- Dashboards by role: dispatchA Capacity Dashboard on ServiceTitan Data: Booked Hours Against Open Hours, 7 and 30 Days OutBooked hours against available hours for the next 7 and 30 days, by skill, zone and business unit, read from your ServiceTitan account. From $575 to build.
- The build, in fullA Custom ServiceTitan Dashboard for One Role, With the List Behind Every NumberA custom ServiceTitan dashboard for one role: written definitions, a visible refresh time, lists one click away. From $575, hosting from $55 a month.
- Once a month, out loudThe Monthly ServiceTitan Numbers Review: What Moved, Why It Moved, and What Changes on MondayA monthly numbers review for ServiceTitan owners and managers: what moved, why, and what to do next. From $575 a month, no long contract, honest about fit.
- Dashboards by role: ownerA ServiceTitan Owner Dashboard for 7:45 a.m.: Eight Numbers, Each With Its List Behind ItThe eight numbers a home service owner checks before 8 a.m., read from your ServiceTitan account, each defined and one click from its list. From $575.
- Everything in ServiceTitan DashboardsSee the section
From the Insights desk
- The Estimate You Already Wrote Is the Cheapest Lead You Will Get This MonthAn unsold estimate is a lead you have already paid for in full. Why estimates go cold, what Jobber, Housecall Pro and ServiceTitan already do about it, what automation cannot do, and a follow-up rhythm that does not nag.
- The Leads Are Weak: What Glengarry Glen Ross Gets Backwards About Home Service CompaniesHome service owners quote the movie and complain about the leads, then close nearly every real homeowner who calls. The close was never the problem. The stack on the desk is. Where the good leads actually come from, why purchased ones feel weak, and how to own the board instead of racing for scraps.
See your three hourly numbers side by side
Tell us how your team counts revenue per hour today. Someone at DataCram replies with what that version measures, what it misses, and what a three-measure tile would take.
Which of these sounds like you? (pick any)
Rather talk now? Call (210) 346-0848.