Your On-Call Rotation Is a Retention Problem in Disguise
Your On-Call Rotation Is a Retention Problem in Disguise
In 2022, I lost two techs in twelve months and told myself it was the market. Good guys, both of them. One had three years in, one had about eighteen months. Both left for competitors. I gave myself the standard story: other shops are paying more, the labor market is crazy, nothing I could have done differently.
Then I ran the actual numbers on what replacing them cost me — recruiting time, training hours, the jobs I couldn't take because I was short a truck — and landed somewhere around $34,000. Not a survey number. My number, built from my calendar and my P&L.
The thing those two techs had in common: they were both on what I generously called an "on-call rotation." Which meant, in practice, that I called whoever picked up.
The On-Call Tax Nobody Is Calculating
On-call isn't a scheduling problem. It shows up in your schedule, but it lives in your P&L — and if you're not running a monthly close, it never surfaces as a line item. You see "labor ran high again" and you move on. The actual driver stays invisible until one of your better techs puts in notice.
Here's how I know. At launch in early 2022, I priced my service calls without any after-hours premium. Flat rate, same price regardless of when the call came in. I figured I'd stay competitive. I figured the volume would make up for it.
The after-hours call that goes sideways at 10:45pm doesn't just cost the labor. It costs the next morning's decision quality — the service call I quoted badly on four hours of sleep, the tech I burned out by calling him on a Sunday three weeks running because the "rotation" was really just me knowing which guy was least likely to complain.
Run the close. Every month. If you don't know what after-hours coverage cost you last quarter, you don't know your real margins.
The Tech Who Doesn't Quit — He Gets Quiet First
First-to-third year techs are not fragile. They can handle the 11pm call. They can handle the Saturday panel swap. What they can't handle is not knowing when the next one is coming.
Unpredictability burns people out. Not the work itself. A tech who knows he's carrying the week of the 14th can make plans — tell his girlfriend what's happening, mentally prepare, book his camping trip for the week after. A tech who might get called any night this week, or possibly next week, depending on who's available, can't do any of that. He just waits. Waiting is exhausting in a way that working isn't.
The shops not putting on-call expectations in the job posting are setting this up wrong from day one. The tech accepted a job. He didn't sign up for undefined terms. When those undefined terms materialize — 11pm Sundays, no premium, no warning — he starts updating his resume. Quietly, so you don't see it coming.
There's a comp dimension here too. A journeyman at $55K eating unpredictable after-hours calls is running a worse deal than that number looks. A journeyman at $85K with a defined rotation and an after-hours premium is a different job entirely. The shops paying the higher number and publishing a clear rotation are winning on two dimensions at once. The $55K shops are competing on neither.
What a Real Rotation Looks Like (And Why "We All Just Help Out" Isn't One)
"We all just help out" is not a rotation. It's an anxiety distribution system.
A real rotation is written down. Every tech knows their week before the month starts. And the techs were in the room when it was built — that last part is the one most shops skip.
At five trucks, the math is clean. One tech carries on-call for one week, then it rotates. Everyone carries roughly one week in five. That's a known quantity. A person can plan around one week in five. One week in five is a job condition. Random availability forever is a different animal.
A rotation handed down from the owner is a grievance waiting to happen. The tech didn't author it, didn't raise concerns, didn't negotiate any of it. When it creates friction — and it will — he has no investment in making it work.
Bring your techs into the room. Literally. One meeting, 45 minutes: here's the problem we're solving, here's the structure I'm proposing, what am I missing? You will hear things you didn't expect. Someone has a standing Tuesday conflict. Someone just had a kid and has a strong preference about Fridays. You can accommodate most of this and still have a functional rotation. What you can't do is hand down the schedule and act surprised when resentment builds.
Write it down after the meeting. Put it somewhere everyone can see it. Review it every quarter.
After-Hours Pricing Is Not Optional — It's the Subsidy You Didn't Know You Were Running
The "we charge the same rate to stay competitive" logic is self-defeating. I know because I used it.
Pricing flat across all hours is the same mistake as pricing day rates. I priced day rates my first three months in business because they felt simple, and I deserved every dollar I lost on them. You cannot know if you're making money if you don't know what any given job actually cost you. A day rate collapses the variation into a number that feels clean but isn't. After-hours flat pricing does the same thing.
That 10pm call isn't costing you what the 2pm call cost you. The tech got pulled off a family evening, drove across town, lost sleep. You're compensating him — you have to, or he quits — but you're not recovering it from the customer. You're subsidizing it. You've just made the subsidy invisible by calling it your standard rate.
You anchor to the standard rate because that's what you've always charged. The retention problem builds quietly. Then a good tech leaves, and you reconstruct the damage, and the numbers are worse than you expected because you were never seeing the real cost.
Price after-hours calls at a premium. In my market — East Austin — I landed at 1.5x standard labor for evenings, 2x for middle-of-the-night, flat trip charge on top regardless. Your market may be different. The specific numbers matter less than having numbers at all. Once you have a rate, you can evaluate whether it's working. Flat rates can't be evaluated — they just exist until they don't.
What February 2022 Actually Taught Me About After-Hours Calls
When I launched Reeves Electric, I was the on-call rotation. One person, two trucks, no dispatcher, no CSR. I answered everything.
I had not fully prepared for what that meant at 9:30pm when an irritated customer called because her breaker kept tripping and she had a house full of guests. I answered it. I took it poorly. I was tired, and the part of me running the business and the part of me that needed to sleep were in direct conflict — and at 9:30pm exhausted, one of them was always going to lose.
My NPS in month two was a 4. The number is embarrassing but accurate. What it told me was that intake was broken — people were calling with problems and getting a version of me that was depleted and moving too fast.
What fixed it wasn't marketing. It was ninety days of answering every call myself with a notepad and the specific intention of hearing what was actually wrong. That discipline — sitting with the thing that's broken long enough to understand it — is what took the NPS to 81 by month nine.
The on-call problem responds to the same approach. Pull your after-hours calls. Listen to them. Not to evaluate the tech's phone manner — to understand what the call actually was. Was it a real emergency or a scheduling failure that could have been handled next morning? Did you bill a premium? If not, why not?
The recording is the diagnostic. Fix what you find.
What to Actually Do Before Friday
First: pull your last 30 days of after-hours calls in CallRail. Tag each one by source, call duration, whether it booked, and whether it billed at a premium rate. If you don't have a premium rate, that last column will be empty — and that's your answer. You've been running after-hours coverage at standard rate and now you have the data to prove it. Not on CallRail yet? Base tier runs around $50 to $150 a month depending on your volume. Worth it for this exercise alone, and worth it every month after.
Second: have the rotation conversation with your techs this week. One meeting. Come in with a proposed structure — one week in five at five trucks, one week in four at four trucks. Ask what you're missing. Write down the output. The owner needs to be in this meeting, not delegating it, because your techs need to know the person who built the rotation is the same person they'll call when it isn't working.
Third: wait 30 days before changing the pricing. Do not fix the rotation and launch a new rate structure the same week. Your techs need to understand the rotation before you add a premium structure on top of it. Your customers need to hear a consistent answer before your team even knows what to say. Get the rotation stable, run it a month, then have the pricing conversation with a clear script for whoever answers the phone.
The rotation conversation is Monday. The pricing change is 30 days later. That gap is what makes both changes stick.
FAQ
My guys have always just covered for each other — why would I formalize it now and risk making it weird?
Because "always just covered for each other" works until it doesn't, and when it stops working, you find out at the worst moment — usually when your best tech is fed up and your second-best is already halfway out the door. The formalization gives every person on your crew a known quantity they can plan around. The adjustment period lasts about a week. The resentment from not formalizing it lasts until they quit.
How do I figure out what to charge for after-hours calls if I've never separated the rate before?
Start with what the call actually costs you. The tech's on-call premium, the trip charge, the time-of-night disruption. Build from the cost up, not from a competitor's price down. In my market I landed at 1.5x standard labor for evenings and 2x for middle-of-the-night, plus a flat trip charge. Your number will be different. The point is to have a number you built deliberately rather than a flat rate you inherited from a pricing decision you made at launch and never revisited.
What's the minimum team size where a real on-call rotation makes sense?
Three trucks. Below three, the rotation is too thin — one week in two is still a lot of exposure per person, and you'd be better off pricing for it and handling it yourself while you grow. At three trucks, one week in three is manageable with a premium structure. At four or five trucks, the tech genuinely has four out of five weeks free. That's a real benefit, not a marginal one.
One of my best techs told me he'll quit if I put him on formal on-call — how do I handle that?
Have the actual conversation. Not a performance management conversation — a real one. Ask what specifically about formal on-call is the problem. In my experience it's usually one of these: he's been burned by an informal rotation that wasn't actually fair, he has a life circumstance the current structure doesn't accommodate, or he's never seen a well-run rotation and assumes it'll be as bad as what he's seen. Address the real issue. If after a full conversation he genuinely won't participate in any coverage structure on any terms, that tells you something about the match — but don't assume that's where it lands before you've had the real conversation.
How do I sell after-hours rates to customers who are used to paying the same price regardless of when they call?
Tell them directly: after-hours work requires your tech to be available specifically for that call, and that availability costs money we now price for explicitly. Most customers understand this. The ones who push back have been trained to push back because shops never held the rate — those first few conversations are harder, but they get easier fast. Post the after-hours rate on your website. Put it in the booking confirmation. The more visible it is before the call happens, the less friction at the invoice.
Should the on-call tech get a flat weekly stipend just for being available, even if no calls come in?
Yes. The tech is giving up real flexibility even when the phone doesn't ring. In my shop we pay a flat weekly amount for carrying the rotation regardless of call volume, then the per-call premium on top. The stipend doesn't have to be large. But paying nothing for availability and expecting no resentment means you've priced the rotation wrong — same as pricing a service call without accounting for the drive time.
Enjoyed this article?
Get articles like this in your inbox every Monday. Free, no spam.