By Jane Steen, Head of Engagement at ETL.
When you ask most aftersales teams how retention is doing, you’ll get a number. When you ask where the customers are going, everyone tends to go quiet.
That’s not a failing of the team; it’s a failing of the data they’re working from. Aftersales retention is often measured as a simple network-wide figure, reported monthly, and treated as a thing that simply rises or falls. What it rarely comes with is a map: you can see that customers stop coming back, but you can’t easily see which ones, after which visit, at which sites, or why. And without that, any retention initiatives are based on guesswork.
This is one of those problems that looks like a marketing challenge, but is actually a data one.
Retention isn’t one number, it’s a funnel
As with most industries, a customer doesn’t decide once whether to stay loyal. They decide repeatedly, at each touchpoint, to use your network for servicing, and they tend to leak out at predictable moments.
The first service after purchase is one. The end of the warranty or service-plan period is another, and is often the steepest. The third or fourth year, when the car is out of plan and the independent garage down the road quotes 40% less on a cambelt, is another again. Each of these is a distinct moment with its own drop-off rate, and each one responds to different interventions.
Reported as a single annual retention percentage, all of that detail disappears. You’re left knowing the network lost ground without knowing where to focus first. It’s the difference between being told your bucket is leaking and being shown exactly which joins have cracked.
The reason you can’t see the funnel
In our experience, the obstacle is almost never the willingness to look; it’s that the data needed to draw the funnel is spread across systems that were never designed to talk to each other.
Service history sits in the DMS. But there’s rarely just one DMS, there’s often as many as there’s dealers, each with its own way of recording a job, a customer and a vehicle. The same customer who bought from one site and had servicing at another can appear as two unrelated records. A “completed service” means one thing in one system and something subtly different in the next. Parts and labour are coded inconsistently. And warranty claims often live somewhere else entirely.
Pull a network-wide retention figure out of that environment and it isn’t wrong, exactly. It’s just an average of things that were measured differently, which isn’t ideal. We’ve seen teams spend more effort arguing about whose numbers to trust than acting on what the numbers said.
The unglamorous work, the part that gets skipped, is reconciling all of it: matching vehicles by VIN across sites, resolving the same customer into a single record, agreeing what a service event actually is, and making the consent flags travel with the data so anything you do next is on the right side of the GDPR. None of that looks particularly interesting in a monthly report, but all of it determines how well performance is measured.
What you can do once the data joins up
When service, sales, parts and warranty data are brought into a single, consistent view across the network, the questions change. Instead of “what’s our retention?”, which only ever produces a number, you can start asking the questions that lead somewhere:
At which point in the lifecycle are we losing people, and is that point the same across the network or specific to certain sites?
When a customer doesn’t return for a second service, what did the first one look like, in price, in turnaround, in which parts of the work were actually done?
Which dealers retain customers through the post-warranty cliff better than their peers, and what are they doing differently?
Are there parts and labour patterns that reliably precede a customer disappearing, so you can intervene before they’re gone rather than after?
These are ordinary aftersales questions that simply can’t be answered while the underlying data is fragmented. Once it’s joined up, they become routine.
Why this matters beyond aftersales
The same joined-up view is important for two neighbouring teams.
For network quality and development, it makes dealer benchmarking more accurate. You can only fairly compare sites on retention once “retention” means the same thing at each of them, and once you’re confident you’re not double-counting customers who move between sites. Consistent definitions have to come before the league table, not after.
For customer experience, it’s the difference between seeing a customer and seeing a scatter of disconnected transactions. A person who has bought, serviced and part-exchanged across three sites over six years is, to a fragmented system, several strangers. Stitch those records together and you can finally see the relationship the way the customer experiences it, which is the only vantage point from which measures like customer satisfaction or net promoter score mean very much.
The best question to ask
There’s a tempting version of this story where the answer is a new platform, or a layer of AI over the top. We’d love that to be true, but unfortunately it rarely is.
Instead, the useful question is: which of the aftersales decisions we’re making this quarter are being made on incomplete or inconsistent data, and what would it take to fix the join? It’s a less exciting question than “what should we buy?”. But in our experience it’s the one that tends to pay for itself, because it turns a limited number into something a team can actually act on.
Our DataHub system was built for exactly this: bringing data from across a dealer network’s disparate systems into one consistent, reliable view, so the people responsible for retention can finally see where it’s slipping away.
If any of this sounds familiar, get in touch with me on LinkedIn (click here) and we can compare notes.
