Service customer churn is usually silent: a customer who has decided to leave rarely tells you. They pay, they drive off, and their next oil change happens at an independent shop or a quick-lube down the road. You find them by looking for who is overdue and calling the ones whose last visit showed warning signs, and you keep them by fixing what went wrong while it's still fresh.
How big is the problem?
Cox Automotive, whose businesses include service software, tracks where owners of newer cars take them for service. Its finding: only 54% of owners of cars under two years old went back to the selling dealer for service, down from 72% (Cox Automotive). "Selling dealer" here means the store where they bought the car, so these are customers you sold a car to.
The service relationship is also tied to the sales floor. Cox reports that 74% of buyers who returned to the dealer for service in the past year are likely to buy from it again, against 44% of those who don't (Cox Automotive). Put plainly, buyers who return for service are 30 points more likely to buy again (Cox Automotive).
On the service side alone, Cox estimates that losing a service customer can mean more than $12,000 in potential lifetime service spend (Cox Automotive estimate). It's a modelled figure across the life of the vehicle.
J.D. Power looks at it from the satisfaction end. When dealer service satisfaction is 950 or higher on a 1,000-point scale, 86% of mass-market and 88% of premium customers who own 1–3-year-old vehicles say they "definitely will" return to the dealer for paid service (J.D. Power 2026 U.S. Study).
These studies measure different things, so quote them separately and don't add them together.
Takeaway
Quote each figure with its study name, and it holds up when someone asks where it came from.
Why do service customers leave without saying anything?
In the 2025 Cox Automotive Service Industry Study, 45% of dealership service customers reported a frustration during their visit.
That finding doesn't say how serious those frustrations were. From the counter, the reasons people drift away often look ordinary, and many never get raised at the cashier.
- Price surprise. The number at the cashier didn't match the estimate, or the advisor recommended work the customer didn't expect. They paid, and decided the dealer is "expensive".
- Time. The car was promised for noon and ready at closing time, and nobody called.
- Trust in the advisor. A customer who felt sold to won't argue at the counter. They'll book the next visit somewhere else.
- Convenience. The independent shop is close by and takes walk-ins. Once the factory warranty is gone, that's a strong pull.
- A problem nobody asked about. A rattle that's still there, or a loaner that smelled of smoke. The customer never mentioned it, so nobody fixed it.
Takeaway
Ask your advisors at Monday's drive meeting which of these they saw last week, and write down the names.
The warning signs at the visit
Your advisors see much of the risk before the car leaves the lot.
How to read it: each row is a moment at the counter, what to do about it before close, and who owns it.
| Warning sign | What it looks like at the counter | What to do the same day | Who owns it |
|---|---|---|---|
| Declined recommended work with a price reaction | "That much?" A long pause, a phone out to check prices, a flat "no, just the oil change." | Note the declined items on the or inspection with the reason. Offer to break the work into what's urgent and what can wait, in writing. Don't push again at pickup. | Advisor |
| Came in for warranty or recall only | No lines on the RO. The customer may not think of you as their shop at all. | Walk the car with them at pickup, explain what's due next and when, and give them a direct line to their advisor. | Advisor |
| Waited past the promised time | Promise time on the RO has passed and the car isn't ready, or the customer is in the lounge asking. | Call or walk over before they ask. Give the real time and the reason. At pickup, the advisor or manager acknowledges the delay in person. | Advisor, then service manager |
| Same concern, same vehicle, back again. | Service manager speaks to the customer before the car goes back in. After the repair, confirm the fix by road test with the customer if they're willing. | Service manager, shop foreman | |
| "I'll think about it" | A soft no on a diagnosis or a major repair. | Send the written estimate with the advisor's direct number. Don't chase it as a sales lead; ask whether they have questions. | Advisor |
| Complaint at the counter | Anything said out loud: price, cleanliness, wait, attitude. | Log it where the manager will see it today. Manager follows up before close of business. | Advisor logs, service manager follows up |
What it tells you: the advisor starts almost every fix, and the service manager steps in on comebacks, missed promise times and complaints.
Write each sign down on the RO, in the inspection tool or in your , with the reason.
Takeaway
Add a reason field for declined work and a flag for missed promise times, and the report below can find those customers for you.
Finding the ones who've already drifted: the report
This report catches the customers who already slipped through: everyone who should have been back by now and hasn't, sorted by what their last visit looked like. Your reporting tool can likely export the history, and the work is in the logic.
The fields to pull
From your DMS RO history (plus your , inspection tool and scheduler, where noted), pull a row for each customer-vehicle pair.
How to read it: each field, where it lives, and why the report needs it.
| Field | Where it comes from | Why it matters |
|---|---|---|
| Customer name, phone, email | DMS customer record | Who to contact |
| Do-not-contact flag | DMS or CRM contact preferences | Keeps people who asked not to be contacted off the call list |
| , year, make, model | DMS vehicle record | Segments by vehicle age |
| In-service (warranty start) date | DMS vehicle record or warranty lookup | With last odometer, shows who has aged out of factory coverage |
| Last RO number and open date | RO history | Anchor for "how overdue" (close date can lag, especially on warranty ROs) |
| Odometer at last RO, and at earlier ROs | RO history | Estimates how many miles a day this car drives |
| Pay types on last RO (, warranty, internal) | RO labor lines | Flags warranty-only customers |
| Declined recommendations on last RO | Declined-services report, RO notes, or your /inspection tool | Flags price reactions and deferred work |
| flag | Repeat-repair flag, or same VIN and same concern within your comeback window | Flags comebacks |
| Promise time and "ready" time | Promise time from the RO header; ready time from the last tech flag time or your scheduler's status | Flags long waits (pickup time often isn't in the DMS at all; use it only if you capture it) |
| Complaint or customer-relations note | CRM or customer-relations log | Flags logged complaints |
| Advisor on last RO | RO header | Who should make the call |
| Future appointment | Your scheduler | Excludes customers already booked |
| Expected next visit date | Calculated (below) | The core of the report |
| Days past expected | Calculated | Sort order |
What it tells you: the DMS holds the bulk of it.
Working out "expected next visit"
From the customer's recent ROs, work out the average miles per day the car drives. Project forward from the last RO's open date to the next maintenance interval your OEM recommends for that model (for an oil-life monitor, the interval your store schedules). That's the expected next visit date. If there's no earlier RO on file, use the OEM's time-based interval.
Then set a grace window past that date before a customer counts as drifting, write it down, and keep it fixed so the lists compare week to week.
Exclude internal ROs, contract fleet accounts, vehicles sold or traded in, customers who moved, anyone with a future appointment, customers serviced at a sister store, and anyone flagged do-not-contact. Leave them off the list: calling a booked customer to ask why they left costs the list its credibility with advisors.
Sorting the list
Split the overdue customers into groups by their last visit's signals:
- Comeback on last visit. Start here. You know why they'd leave.
- Complaint logged. Same priority.
- Long wait on last visit.
- Declined recommended work. Especially safety items.
- Warranty or recall only. They may never have thought of you as their shop.
- No signal. A clean last visit. A service reminder through your usual channel is enough.
A customer can sit in several groups. Put them in the top-priority one.
The weekly routine
- Monday morning: the service manager (or a lead the manager trusts) runs the report.
- Monday: assign groups 1–3 to named people, ideally the advisor who wrote the last RO or the manager for comebacks. Groups 4–6 go to the BDC or your regular reminder process.
- Through the week: every call gets an outcome logged: reached, voicemail, wrong number, moved, sold the car, using another shop and why, booked, asked not to be contacted.
- Friday: the manager reads the outcomes, especially the "why" answers.
- Monthly: if the same advisor or op code keeps showing up in groups 1–3, look upstream for the cause.
Check your store's do-not-call list before dialing, and call within your store's calling hours. Text only customers who agreed to texts. Which customers your store may call, and how, is a question for your store's own counsel.
Takeaway
Run the report on Monday and give groups 1–3 to named people.
Win-back: the sequence
A win-back campaign that opens with a coupon tells the customer you think they left over price. Often they didn't, and the coupon tells them nobody listened. So ask what happened before you offer anything.
Step 1. Who calls. A person with a name and authority: the advisor who knows the customer, or the service manager for and complaints. The customer should hear from someone who can do something about it.
Step 2. What to say. Open by asking. Something like:
"Hi, this is [name], the service manager at [store]. I was looking through our records and noticed we haven't seen your [model] since [month]. I wanted to ask how that last visit went for you, honestly. If something wasn't right, I'd like to know."
Then stop talking. If they tell you what went wrong, don't defend it. Repeat it back so they know you heard it, and ask what would make it right.
Step 3. What to offer. Fix what they told you. If the rattle is still there, offer to look at it. If the price stung, have the manager review the and decide on goodwill for that job. If the wait was the problem, offer a time you know you can hit, with a loaner or a ride. Keep a general discount for when nothing else fits.
Never tie anything to a review or a survey. No "we'll take care of this if you update your rating," no discounts for responses, no asking only the customers you've won back to post a review.
Step 4. If you don't reach them. Leave a short voicemail with your name, direct line and the same question. Follow with an email, or a text if the customer agreed to texts. Try again on another day that week, at a different time.
Step 5. When to stop. When they tell you they've moved on, thank them, ask what made them switch if they're willing to say, and log it. Decide how many attempts you make and write it down. Stop immediately if anyone asks not to be contacted, and set the do-not-contact flag so the next list respects it.
For what to do with a customer who left unhappy and does tell you, see service recovery in the dealership. If unhappy customers are showing up online before they show up in your drive, see handling bad Google reviews of your service department.
Takeaway
Hand each caller the script and your written stop rule before Monday's list goes out.
The warranty cliff
While the car is under factory warranty, the customer has a built-in reason to come to you. When the warranty ends, that reason goes away, and in many stores nobody notices the drop until it has happened.
Before the warranty ends: the visits during warranty are your chance to become their shop by choice. Explain what's coming due after the warranty ends and what it will involve. If your store offers extended service contracts or maintenance plans, the advisor can raise them plainly and drop it if the customer isn't interested.
After it ends: use the in-service date and last odometer in the "who's drifting" report to find vehicles that recently aged out of coverage with no visit since, and put them at the top of group 5.
Takeaway
Give group 5 to the advisor who knows those customers, so the call comes from a voice they recognize.
Measuring retention without fooling yourself
What is a customer retention rate?
For a service department, a workable definition is: of the customers who had a visit in period A, the share who came back for any visit in period B. The hard part is consistency.
Decisions to write down before you run it:
- What counts as a customer. Per or per customer record. Per VIN counts each car in a household separately.
- What counts as a visit. All , or only. Including warranty and recall visits will flatter the number while cars are young.
- The periods. Pick what fits your reporting and keep it.
- Exclusions. Internal, fleet, sold or traded vehicles.
What is customer churn rate?
The other side of the same coin: of the customers active in period A, the share with no visit in period B. If your retention and churn don't add up to the whole, your definitions don't match.
Segment by vehicle age
A store-wide retention number hides where the losses are. Split it by vehicle age (and by warranty status if you can), because a store full of new cars under warranty will look healthy right up until those cars age out.
Don't compare your number to someone else's unless you know they define it the same way.
Takeaway
Put the written definition at the top of the retention report, and you'll see the warranty cliff in your own numbers while you can still call those customers.
Where Lokam fits
Hear from unhappy service customers starting the day after the closes, while there's still time to fix what went wrong. Lokam is post-visit follow-up for US dealerships. Your store calls service customers the day after the repair order closes, using Lokam's AI assistant. The assistant asks how the visit went and flags unhappy customers from what they say, and the people on your store's escalation list get the customer, a short summary and the main issue.
FAQ
What is a customer retention rate?
For a service department, it's the share of customers with a visit in one period who came back in the next, measured the same way every time.
What is meant by customer churn rate?
The share of customers active in one period who didn't come back in the next. In a service drive it's usually silent, because nobody cancels anything.
What's a good customer retention rate?
It depends on how you define it and on the age of the vehicles you service, so compare yourself to your own past numbers, segmented by vehicle age.
How do you win back a lost service customer?
Call them personally, ask how their last visit went before offering anything, fix what they tell you, and stop when they've moved on or ask not to be contacted.
What is a win-back strategy?
A planned way to reach customers who have stopped coming back: a list of who's overdue, sorted by their last visit, named callers, a script that opens with a question, and a clear point where you stop.
What is the word for losing customers?
Churn, or attrition. In you'll also hear "defection" or just "lost customers."
Before you go
What you can do with this
- Run the "who's drifting" report every Monday and you'll know who to call that week.
- Act on warning signs at the counter while the customer is still easy to reach.
- Track retention on a written definition by vehicle age, so you see the warranty cliff coming.
Where Lokam helps
- Cut churn at the source: fix the problems before customers leave.
- Free your team from the follow-up list: they call back the customers who need it.
- Every open issue has an owner, and a reminder when it sits.