/ The cliff

72% in 2018. 54% in 2025.

Cox Automotive's 2025 Service Industry Study tracks the percentage of dealership service customers who return to the dealership of purchase. Among under-2-year vehicles — the group most loyal to dealers — that share has fallen 18 points in seven years.

2018
72%

Of under-2-year customers returned to dealer of purchase. The benchmark.

2025
54%

Returning today. Nearly half of your newest customers are servicing somewhere else.

Cox cuts the same data three more ways and the pattern holds: on the broader question of dealer-vs-alternative-provider share for <2-year vehicles, dealerships lost 13 points (68% → 55%) in the same window. The decline shows up whichever angle you measure.

/ Where they end up

It's not one competitor. It's all of them.

When the <2-year customer leaves, no single provider captures them. Cox's 2025 share-by-vehicle-age breakdown for under-2-year vehicles:

12%

General Repair · up from 9% in 2018

8%

Tire Stores · up from 5% in 2018

8%

Quick Lube · up from 5% in 2018

8%

Mobile + OEM Mobile · new in 2025, didn't exist as a category in 2018.

Each individual gain looks small. Stack them, and they account for the whole 13-point dealer loss. Which means a retention strategy that focuses on out-competing one provider misses the picture — the leak is everywhere.

/ The actual reasons

It's never the work. It's the friction.

Cox asks the customers who didn't return to their dealership of purchase why. The top reason for the second study running is convenience — not a convenient location. Reasons two through five are all about price perception: "they'll overcharge me," "total cost not reasonable," "labor charges," "parts charges."

Both of those reasons collapse on contact. Dealerships are actually $14 cheaper than general repair on average ($261 versus $275 per visit, per the same Cox study). And when Cox holds price equal between dealer and general repair, 45% prefer the dealer versus 32% for general repair — a clear preference that just doesn't get surfaced.

The customer leaving in year one isn't leaving because the dealer is bad at the work. They're leaving because nobody's proactively in conversation with them at the moment they start considering an alternative. By the time they call the Quick Lube down the road, the decision is already made.

/ The compounding tail

The customer you keep is the next car you sell.

This is the part most fixed-ops conversations miss. Cox's 2025 study isolates the link between service retention and vehicle repurchase, and the gap is enormous:

74%

Of customers who returned for service in the past 12 months are likely to repurchase from that same dealer.

44%

Of customers who didn't return for service are likely to repurchase. A 30-point spread on the same scale.

The customer who keeps coming back for service is the customer who buys their next vehicle from you. The one who drifts to general repair is the one another store is about to win.

Cox Automotive 2025 Service Industry Study

Which means the <2-year defection isn't just $1,440 of fixed-ops revenue per lost customer (using Cox's 2.4 visits/yr × 2 years × $300 RO). It's also a 30-percentage-point hit to the next sale. Plug the under-2-year leak and the compounding effect runs in both directions: incremental service revenue this year, increased repurchase rate two to three years out.

/ What it costs to do nothing

Run the math on your group.

We built a calculator for exactly this. It uses the Cox 46% defection baseline and 2.4 visits/yr average, lets you input rooftops, average customer-pay RO, and new-vehicle volume per rooftop, and shows the size of the under-2-year recapture opportunity for your group. The math runs at AutoEngage's $1,895/mo per-rooftop subscription with no add-on modules.

Plug in your numbers in the first-2-year recapture calculator. The defaults assume a 12% recapture rate against the at-risk pool — conservative — and the slider lets you push it as far as 40%.

Related
Study breakdownCox 2025: where the dealer share went →Case mathWhat 11× ROI actually looks like →Case studiesSee the lifecycle motions in real conversations →