Every sold unit in your group cost you acquisition money — ad spend, floor time, BDC (business development center) labor, gross you gave up to close. That cost is sunk. The customer record it produced is not. A group’s sold, service, trade and lead records are the cheapest audience it will ever have, and most groups treat them as an archive: rows in the DMS (dealer management system) that get touched when a recall notice goes out and ignored the rest of the year.
The math on ignoring them is getting worse. Cox Automotive’s service study found only 54% of owners of vehicles two years old or newer returned to their purchasing dealership for service in 2025, down from 72% in 2023 (Digital Dealer). That is an eighteen-point slide in two years, among the owners most likely to still be under warranty. Those customers didn’t stop needing oil changes. They went somewhere else for them.
What first-party data does the group already have?
Every rooftop in the group holds four record types — sold, service, trade and lead — spread across each store’s DMS and CRM (customer relationship management system). Together they describe who bought, what they drive, what work they’ve declined, and when they’re likely to transact again.
Walk the inventory of what sits in your systems right now:
- Sold records. Every delivery, with the vehicle, the term, the payment, the F&I products. The DMS has it.
- Service records. Every repair order (RO), including declined work — the brake job the customer said no to in March is a marketing event in September.
- Trade records. What customers brought in, and what that says about what they drive next.
- Lead records. Everyone who raised a hand and didn’t buy, sitting in the dealer group CRM — often in five different CRMs, because stores were acquired at different times with different vendor contracts.
None of this cost you a dollar of new media. It’s the residue of deals you already paid to acquire. The question is why it produces so little.
Why isn’t it working?
Three failures, and they compound.
Duplicates across stores. The customer who bought a truck at your Chevy store and services a crossover at your Honda store exists twice, as two strangers. Neither store sees the household. Mail the same family three conquest offers and they conclude the group doesn’t know them — because, in the data, it doesn’t.
Unclear consent. Records imported from an acquired store, leads bought from a third party years ago, service customers who never opted into marketing: nobody can say which contacts are safe to email, which are safe to text, and which should never be touched. So either nobody sends anything, or somebody sends everything. Both are failures.
No single owner. Each store’s GM owns their P&L. The BDC owns appointments. Fixed ops owns the service lane. Nobody owns the customer across the lifecycle — which means the second-year service drop-off, the leak the Cox data points at, is nobody’s job to plug.
This is an architecture problem before it is a marketing problem. You can’t buy your way out of it with another vendor tool bolted onto the stack. The work is deduplication, consent classification, and ownership — in that order.
Which lifecycle moments actually matter?
Five moments carry most of the value: first service, the second-year drop-off, declined work, equity and trade timing, and lease end. Each has a trigger already sitting in the DMS.
Not every contact is worth automating. These five are.
First service. The delivery-to-first-RO window sets the habit, and it is where fixed ops marketing earns its budget. An owner who services with you once is a retention candidate; an owner who never comes in is a Cox statistic.
The second-year drop-off. The Cox study’s decline — 72% returning in 2023 down to 54% in 2025 among owners of vehicles two years old or newer (Digital Dealer) — lands right where warranty comfort fades and the independent shop down the street starts looking reasonable. If your service marketing calendar doesn’t have a specific play for the 18-to-30-month owner, you’re running the default: silence.
Declined work. Every declined line on an RO is a pre-qualified, pre-diagnosed offer waiting for a follow-up. Most groups log it and never read it again.
Equity and trade timing. The owner whose payment on a new unit would roughly match their current one doesn’t know that. Your DMS does. Equity mining is the rare play where the data literally names the customer and the offer.
Lease end. The most predictable transaction in retail automotive, with a date stamped on it at signing. If lease-end contact starts when the OEM’s letter arrives, you’re already sharing the customer.
How do you activate it safely?
Consent first. Then channels.
Before any campaign, every record needs a consent status: email-safe, text-safe, or do-not-contact. For email, the FTC’s CAN-SPAM guide requires a clear way to opt out of marketing email and honoring an opt-out within 10 business days (FTC). That’s the floor, not the ceiling. Marketing texts carry stricter consent rules under the TCPA (Telephone Consumer Protection Act) — have counsel review your texting consent before the BDC sends a single marketing SMS. Don’t take a vendor’s word that their platform “handles compliance.” Vendors don’t sit in depositions; you do.
With consent classified, three channels do the work:
- Email to owners, triggered by the lifecycle moments above — not a monthly “newsletter” nobody asked for, but a declined-work reminder six months after the RO, a service offer timed to the second-year window, an equity letter when the numbers pencil.
- Text to owners who have opted in properly, reserved for the highest-value, most time-sensitive moments. Scarcity protects the channel.
- Customer-list audiences on ad platforms. Upload consented owner lists — the same customer-data boundaries apply — and you’re paying to reach people with a known relationship to your stores instead of renting cold audiences at conquest prices. Suppression matters as much as targeting: stop showing acquisition ads to people who bought from you in April.
This is where first-party data beats buying third-party lists on cost alone. You already paid for these names. Everyone else is still bidding on strangers.
How do you measure it?
Not with platform dashboards. With the DMS.
Three numbers, pulled from your own books, tell you whether the program works:
- Repair orders from existing owners — RO count and fixed ops gross from the sold customer base, by store and by cohort. Is the 18-to-30-month cohort coming back at a better rate than it did last year?
- Repeat sales — sold units where the buyer already existed in the group’s records. Track it as a percentage of total units; it should climb.
- Trade-ins from existing owners — the direct output of equity and lease-end plays, and the inventory acquisition channel every used-car manager wishes were bigger.
Ad platforms will claim credit for everything; sending sold units back helps them optimize, but it doesn’t prove cause. The DMS doesn’t flatter. If repeat sales and owner ROs aren’t moving, the program isn’t working, no matter what the dashboards say. This is the honest version of customer retention measurement: transactions, not impressions.
Who owns it?
Group-level. Not per store.
A store-level owner will optimize for their rooftop, and the duplicate-household problem comes right back. The lifecycle view only exists above the stores — someone who can see that the Chevy buyer and the Honda service customer are one family, set the group-wide consent standard, run the deduplication, and hold each rooftop to the same lifecycle calendar while the GMs keep owning their month.
That role doesn’t need to be a full-time executive hire at a 3–15 rooftop group. It needs to be a defined seat with authority over the data, the consent policy, and the marketing calendar that sits on top of both. I spent 3.5 years running marketing for La Mesa RV, an 11-store RV dealer group, where I moved $500K a month from broadcast that could not be measured to digital that could. Different inventory, same architecture: multi-rooftop records, one customer base, and a budget that had to answer for itself.
Where a fractional CMO fits
The customers you already paid for are the one audience where you control the data, the message and the timing. A fractional CMO is the natural owner of this work at a group your size: auditing the records across every rooftop, setting the consent standard with counsel, building the lifecycle calendar, holding vendors to DMS-verified numbers, and giving the GMs back their month. The data is already sitting in your stores. What it needs is an owner.