Ask the internet what a good customer retention rate is for a home service business and you’ll get an answer in about four seconds. Forty to sixty percent. Or thirty-eight percent. Or sixty-six. Or seventy-four to ninety-one percent for maintenance plan members, ninety-seven if they’re really good.
We went looking for where those numbers come from.
They don’t come from anywhere. Not from ServiceTitan, Housecall Pro or Jobber. Not from ACCA, PHCC or NECA. Not from the Census Bureau or the Bureau of Labor Statistics. Not from a peer-reviewed journal. Every figure in circulation traces back to a contractor marketing blog with no citation, and the figures disagree with each other by more than double.
That’s a frustrating thing to hear when you came here for a number. It’s also the most useful thing in this article, because it means anyone who hands you a benchmark with a straight face is selling something.
Here’s what we checked, what survived, and what we’d track instead.
What a good retention rate is
There isn’t a published one. Here’s the search.
| Source | Sample | Publishes a retention benchmark? |
|---|---|---|
| ServiceTitan 2026 Residential State of the Trades | 1,000 residential contractors, via Thrive Analytics | No. Priorities and sentiment only |
| ServiceTitan’s own HVAC retention guide | n/a | No. Gives the formula, offers no benchmark |
| Housecall Pro Customer Service Survey | 1,040 US homeowners | No. Stated intent, not behavior |
| Jobber 2026 Home Service Trends | 1,050 US owners, ±3 points at 90% confidence | No. Lead source mix only |
| ACCA, PHCC, NECA | n/a | Nothing we could find published |
| Census, BLS | n/a | No such measure is collected |
| Academic literature | n/a | Nothing for residential home services |
The most telling line in that table is the second one. The company with the largest dataset in residential home services will show you how to do the arithmetic and won’t tell you what a good answer looks like.
What the vendor research does show is what contractors say they care about. In ServiceTitan’s 2026 survey of 1,000 residential contractors, customer retention came in at 53 percent as a business priority against 31 percent for new customer acquisition. That’s stated priority, not performance. It tells you where people’s heads are, not what their books say.
The numbers you’ll see that aren’t real
| The claim | What’s behind it |
|---|---|
| ”HVAC retention runs 40 to 60 percent” | An agency blog with zero citations on the page |
| ”The industry average is 38 percent” | A benchmarks page laid out like real research, with no sample, method or source on any figure in it |
| ”Plan members renew at 80 to 95 percent, some reach 97” | No source anywhere |
| ”89 percent for plan customers versus 42 percent for non-plan” | Same unsourced benchmarks page |
| ”Minimum 25 percent conversion per ACCA benchmark” | Attributed to ACCA. We couldn’t find it in any ACCA publication |
| ”30 to 35 percent of contracts operate at a net loss (PHCC/Service Roundtable, 2024)“ | Attributed to PHCC. Not traceable to anything PHCC has published |
| ”Repeat customers spend 67 percent more” | Recycled e-commerce folklore with no traceable origin |
| ”A 5 percent lift in retention raises profit 25 to 95 percent” | Real underneath, badly mangled on the way here. More on this below |
The two we’d flag hardest are the ones wearing association names. An unsourced number is a guess. An unsourced number attributed to ACCA or PHCC is a guess wearing a badge, and it will get repeated by people who assume somebody checked.
Why the standard formula doesn’t fit the trades
The formula everyone uses looks like this:
Retention rate = ((customers at the end of the period − new customers acquired during it) ÷ customers at the start) × 100
It’s conventional arithmetic, it’s in the marketing textbooks, and ServiceTitan publishes it with a worked example. It’s also built for a different kind of business than yours.
Peter Fader and Bruce Hardie draw the line in the Journal of Interactive Marketing: what defines a subscription business is that when a customer leaves, you find out. Your gym cancels. Your software doesn’t renew. The departure is an event you can see. In a business without contracts, it isn’t. Their phrase for the problem is telling a customer who’s gone from one “merely in the midst of a long hiatus between transactions.” (Fader & Hardie, 2007, 21(1), 76 to 90.)
That’s your exact situation. A homeowner you saw in 2023 hasn’t quit you. She also hasn’t come back. She’s in a state the formula has no box for, and forcing her into one gives you an answer that swings wildly depending on the window you picked. Run it over twelve months on a plumbing book and you’ll look terrible. Run it over five years and you’ll look great. Neither number means anything.
There’s a second trap worth naming. Housecall Pro surveyed 1,040 US homeowners in October 2025 and found 68 percent said they’d rehire the same business after excellent service, 73 percent said they’d refer. Good methodology, census-balanced, real report. But it measures what people say they’ll do. Keiningham and colleagues, reviewing roughly 8,000 customer responses across banking, retail and internet providers, found repurchase intentions predict actual repurchase only moderately to weakly. Treat 68 percent as a ceiling on goodwill, not a forecast of bookings.
What your software will actually tell you
We read both vendors’ help documentation instead of the blogs about them. The answer is thinner than you’d expect.
ServiceTitan
- No retention report. Their built-in reports index has nothing named for retention, repeat customers, or new versus returning revenue
- The customer-related set is Customer List, New Customers, and AR Transactions by Customer
- The Customer List report does carry Last Job Completed and Lifetime Jobs Completed
- Export it, count everyone with two or more completed jobs, do the division in a spreadsheet
Housecall Pro
- One named Repeat Customer report, on the Dashboard
- Shows the percentage of repeat customers and the revenue from them
- Essentials and MAX plans
- Two caveats straight from their help center: it’s current month only, and the legacy Dashboard it sits on is “being phased out and will eventually be deprecated”
Between the two dominant platforms in this market, there is one retention metric. It covers thirty days, on a page marked for deletion.
That’s worth sitting with. Contractors aren’t measuring retention because their software largely doesn’t measure it for them, and the vendors selling you retention advice are not shipping the report that would let you check their advice. The workaround is unglamorous and it works:
What to measure instead
Four things, in order of how much they’ll tell you.
Repeat customer rate over a window you chose on purpose. Count customers with two or more completed jobs inside a period that matches how often your work actually comes back around. This sidesteps the whole problem, because it never claims anybody left. It only counts who came back.
Revenue from returning customers as a share of total revenue. This is the one Housecall Pro chose to ship, which is a decent signal that it’s the number the industry operates on. It also answers the question an owner actually has, which is how much of the year depends on people who already know you.
Time between jobs. Not a retention metric on its own, but without it the first two are uninterpretable. If you don’t know your own repurchase interval, you can’t tell a dead customer from an early one.
Lead source mix. Jobber’s 2026 report, surveying 1,050 US owners, found 59 percent name referrals and repeat work as their top source of leads. You can talk honestly about retention’s share of your business without ever computing a retention rate. (Jobber also suggests 15 to 35 percent of new work should come from referrals and repeat customers, though they don’t publish the data behind that range.)
The one piece of real data on follow-up
There is exactly one published, platform-scale dataset in this industry that isolates follow-up as a variable, and almost nobody uses it.
ServiceTitan looked at commercial service agreements on its platform as of January 2026. Agreements that reached end of term renewed at 37 percent overall. Contractors running fully manual renewal workflows renewed 29.8 percent. Turning on auto-renew took it to 42.4 percent. Layering email reminders on top of auto-renew took it to 49.1 percent.
Three honest caveats. This is commercial work, not residential, so don’t read it straight across to your membership base. ServiceTitan doesn’t disclose how many agreements or contractors are in it. And they attach their own note that correlation does not imply causation, which is fair, since the contractors who turn on automation probably differ from the ones who don’t in ways that also affect renewals.
With all of that said: a 6.7 point difference associated with adding email reminders, and a 19 point spread from the manual floor, is the strongest trade-specific evidence anyone has published that systematic follow-up and retained revenue travel together. We’d rather hand you that with its caveats attached than hand you a clean number that isn’t real.
What retention is actually worth
The famous claim is that a 5 percent lift in retention raises profits 25 to 95 percent. That’s a paraphrase of Frederick Reichheld and W. Earl Sasser Jr. in Harvard Business Review, September to October 1990, and the original is more specific and more useful than the version that reaches you. What they reported was that cutting the defection rate by five points produced 85 percent more profit in one bank’s branch system, 50 percent more in an insurance brokerage, and 30 percent more in an auto-service company.
Thirty percent, in the auto-service company. That’s the closest thing in the original research to your business: episodic repair work, no contract, a customer who shows up when something breaks. It’s a smaller number than the one everybody quotes and it’s the one worth quoting.
Reichheld’s own later restatement for Bain, in the 2001 brief Prescription for Cutting Costs, is narrower still: in financial services, a 5 percent increase in retention produces more than a 25 percent increase in profit. One industry, stated plainly, no methodology given.
On the cost side, the “five to 25 times more expensive to acquire” line is Amy Gallo’s, and she hedged it herself: it depends on the industry and which study you reference. Nothing measures the cost of retention in home services. What is measured is what a new lead costs. LocaliQ’s 2025 benchmarks, drawn from 3,211 US home service search campaigns run between April 2024 and March 2025, put the median cost per lead at $90.92 across home services: $127.74 for HVAC, $129.02 for plumbing, $93.69 for electrical, $228.15 for roofing and gutters.
Per lead. Not per booked job. Whatever your close rate is, divide by it, and that’s the real cost of the stranger you’re about to buy. Then compare it to the cost of emailing somebody who already paid you.
Where that leaves you
You can’t benchmark your retention against the industry, because the industry hasn’t measured itself. You can measure your own, if you pick a window with a reason behind it and export the report your software does have.
And you can do the thing that every honest source here points at from a different angle. Reichheld found that service businesses profit disproportionately from customers who stay. Fader and Hardie explain why you can’t tell who’s staying. ServiceTitan’s renewal data shows that reminders and renewals move together. Jobber’s owners say most of their work already comes from people who know them.
None of that requires a benchmark. It requires that the people who already paid you hear from you before the next thing breaks, which is the entire argument for staying in touch, and it’s the one part of this nobody has to guess about.