Everybody Added Capacity. Nobody Added Output.

What Q2's fixed ops stall means for dealership operations.

In four business days at the end of July, six of the largest public dealership groups in North America told the same story with slightly different numbers. Service is now roughly half of their gross profit. And in the second quarter, it barely grew.

What Happened

Asbury reported on July 28. Penske and Lithia on July 29. Group 1 and Sonic on July 30. AutoNation on July 31. Read the six releases side by side and one line of the P&L does all the work.

Fixed operations has quietly become the profit center these companies live on. Parts and service produced 44.7% of Penske's retail automotive gross profit, 49.3% of AutoNation's total gross profit, 49.7% of Asbury's, and 42.2% of Lithia's. At Penske, service and parts generated under 12% of revenue and nearly 45% of gross profit. That is the modern dealership in one ratio.

The problem is what happened to the growth rate. On a same-store basis, second-quarter fixed operations gross profit came in at roughly +3.1% (Lithia), +3.4% (Penske), +2% (Sonic franchised), +0.1% (AutoNation), −0.2% (Group 1) and −1% (Asbury). Six groups. A spread of about four points. And a midpoint that rounds to flat.

For context on why that matters: same-store new-vehicle gross profit per unit fell somewhere between roughly 9% and 18% across these same six companies in the quarter. Fixed ops was supposed to be the shock absorber. In Q2 it absorbed almost nothing.

Sources: Penske Automotive Group Q2 2026 results (July 29, 2026) · AutoNation Q2 2026 results (July 31, 2026) · Lithia & Driveway Q2 2026 results, 8-K Ex-99.1 (July 29, 2026) · Group 1 Automotive Q2 2026 results (July 30, 2026) · Asbury Automotive Q2 2026 results, 8-K Ex-99.1 (July 28, 2026) · Sonic Automotive Q2 2026 results, 8-K Ex-99.1 (July 30, 2026)

The Bigger Trend

Zoom out and the structural case for fixed ops has never been stronger. Cox Automotive's 2026 Fixed Operations and Ownership Study found that nearly two-thirds of consumers now keep a vehicle five years or more, up from 54% in 2024, and the average vehicle being disposed of is ten years old. Average dealer service and parts revenue reached roughly $9.23 million in 2025, up 33% over eight years.

The car park is aging in exactly the direction that should print money for a franchise service department. Cost per mile runs about 20 cents in the first five years of ownership and roughly $1.10 per mile past ten years. Sonic's investor relations team noted on its call that the average vehicle coming through its service lanes is about five years old and that the returning-to-dealer cohort should keep growing for two to three more years.

And yet, in the same study, dealer share of service visits fell from 33% to 29%. There are now nearly 299,000 auto mechanic businesses operating in the U.S., up 12% since 2018, plus a mobile-service category that did not meaningfully exist a decade ago. Average consumer spend at a dealership was $261 against $275 at general repair — dealers are already cheaper on average, and still lose the perception battle.

So the demand is growing, the dealer share is shrinking, and the groups that depend most on this line of business just posted their softest quarter of growth in it. That is not a demand story. That is a capacity-conversion story.

Source: Cox Automotive 2026 Fixed Operations and Ownership Study (April 9, 2026)

Why Dealers Should Care

Here is the part that should stop an operator cold. Every one of these groups did the obvious thing. They added capacity.

• AutoNation grew same-store franchise technician headcount by more than 2% year over year, and its CFO tied that directly to the target: growing the technician workforce is what makes mid-single-digit after-sales gross profit growth possible. Customer-pay repair orders rose 5% and warranty repair orders rose 8%. Same-store after-sales gross profit still landed at +0.1%.
• Group 1 added technicians in the quarter and reported same-store technician headcount up 2% in both the U.S. and the U.K. It also reallocated collision capacity toward traditional service work, which pulled same-store collision revenue down 15%. Same-store parts and service gross profit: −0.2%.
• Sonic added both bays and technicians. Same-store franchised fixed operations gross profit grew 2%, and management said plainly on the call that low-single-digit growth is not acceptable and that there was a "wobble" in the quarter.

Three groups, three deliberate capacity investments, and essentially no incremental output. When you add inputs and the output does not move, you have not found a demand ceiling. You have found a utilization ceiling.

Group 1's CEO framed the next move almost exactly that way on the call — the focus now is on getting more out of the additional technician capacity the company has already built. That is the right instinct. The hard question is how you would even know whether you were succeeding.

The Hidden Problem

We read all six releases, all six investor decks where available, and the available call transcripts. Between them, these companies disclosed technician headcount growth, effective gross margin, customer-pay growth rates, repair-order growth rates, DMS conversion progress, and days' supply down to the unit.

Not one of them disclosed a shop utilization figure. Not one disclosed an average repair-order cycle time. Not one disclosed hours per repair order, bay turns per day, or the elapsed time between check-in and dispatch. The word "reconditioning" appeared essentially once across the entire set — and only as an explanation for why internal work was down.

That is the gap. The industry manages fixed operations in dollars and headcount because those are the numbers the DMS produces. It does not manage fixed operations in minutes, because almost nobody measures the minutes.

Cox found that high-performing service departments run their bays at 90% utilization or better. That is a wonderful benchmark. It is also unusable to a dealer who cannot state their current number — and most cannot, because bay utilization is not a field in the DMS. It is a physical fact about where vehicles, keys and technicians actually are, minute by minute, and it lives entirely outside the systems dealers already own.

Think about where a repair order actually loses its day. The customer drops the car at 7:40 a.m. and it sits in the lot because nobody knows it arrived. The technician is dispatched at 9:15 and spends eleven minutes walking rows looking for it. The key is in an advisor's pocket, or in a jacket, or in the wrong drawer. The vehicle comes off the lift at 2:00 p.m. and sits another ninety minutes because no porter knows it is finished. The customer calls at 3:00 for a status update and the advisor puts them on hold to go look.

None of that appears on the repair order. The RO shows 1.4 hours of billed labor. The day shows nine hours of elapsed time. The difference is the entire growth opportunity, and it is invisible on every dashboard in the building.

Meanwhile the front end is quietly documenting the same failure from the other side. CDK data reported this week shows 61% of service appointments are still booked by phone, 29% of service customers report difficulty scheduling by phone, and average service hold time runs 9.3 minutes. Car Wars' mid-year review of roughly 44.4 million inbound calls found dealers missed more than 3.4 million service calls, with follow-up on only 46% of them, and service connection rates of 59% against 78% at top performers. A nine-minute hold is not a phone-system problem. It is what happens when the person answering has to go find the answer.

Sources: CBT News — CDK finds phone friction persists at dealerships as AI adoption grows (Aug 3, 2026) · Car Dealership Guy News — Car Wars 2026 Mid-Year Review (Aug 3, 2026)

The MDD Perspective

The Q2 results are a clean natural experiment, and the finding is uncomfortable: capacity you cannot see does not convert into output. Adding a technician to a shop with no visibility into vehicle location, key status or job stage adds cost with a lag and adds throughput only by accident.

This is the distinction MDD builds around. There is a difference between a dealership that reports on what happened and a dealership that controls what is happening. The DMS is a reporting system — it tells you, accurately and after the fact, what was billed. Real-time operational control is a different capability. It answers, right now: where is this vehicle, where is its key, who has it, what stage is it in, how long has it been there, and is that longer than it should be.

Real-time location and key control and stage-level workflow visibility do three things to the numbers these six groups just reported:

• They convert dwell time into a measured number. Once you can see that the average vehicle spends 47 minutes between check-in and dispatch, that becomes a target instead of an assumption. Most of the gap between billed hours and elapsed hours is recoverable, but only after it is visible.
• They make added capacity actually produce. A 2% increase in technicians should produce more than 0.1% growth in gross profit. It does when the technician spends the recovered minutes turning wrenches instead of hunting for a vehicle or a key.
• They make transparency operationally possible. Cox found customers who received photos or videos during a service visit spent about $230 more per repair order, and 49% said seeing visual evidence made them more likely to approve recommended work. Surveyed dealers using photos and videos cited a 53% lift in consumer trust. That entire mechanism depends on a technician being with the right vehicle, on time, with a documented stage — which is an operational precondition, not a marketing one.

The same logic runs through the used-car side. Lotlinx reported this week that only 9% of dealers can flag a struggling unit inside its first 15 days on the lot, and nearly two-thirds take 15 to 45 days. But a large share of those first fifteen days is not merchandising — it is recon dwell, and recon dwell is a location and workflow problem before it is a pricing problem. Group 1 explicitly attributed part of its aftersales softness to lower internal reconditioning volume. Recon is the seam where fixed ops and used-car performance meet, and it is the least instrumented process in the building.

Source: CBT News — Lotlinx survey finds only 9% of dealers catch early inventory risks (Jul 30, 2026)

What Dealers Should Do Next

1. Instrument elapsed time, not just billed time. Before the next capacity investment, establish a baseline for four timestamps on every repair order: arrival, dispatch, work start, and ready-for-delivery. The gap between billed hours and elapsed hours is your real throughput opportunity, and you cannot negotiate a labor-rate increase to make up for it. If you cannot state your current bay utilization to within five points, that is the first project — not the next hire.
2. Solve key control and vehicle location before you add headcount. Every group that added technicians in Q2 got low-single-digit growth for it. The cheapest capacity in the building is the capacity you already pay for and lose to search time. Audit it directly: for one week, log how long technicians and porters spend locating vehicles and keys. Most operators are surprised by the number, and it is recoverable without a single new hire.
3. Make live status visible to the advisor and the customer at the same time. If an advisor has to leave the phone to find an answer, the 9.3-minute hold and the 3.4 million missed service calls are structural, not behavioral. Status that updates automatically — and can be pushed to the customer as a text with a photo — shortens the call, raises approval rates, and removes the single most common reason a service customer never calls back.

The Bottom Line

The Q2 prints made the strategic case for fixed operations undeniable and the operational case urgent in the same week. Service is now the engine — close to half of gross profit at the largest groups in the country — and in the second quarter, the engine did not accelerate even when six sophisticated operators fed it more technicians and more bays.

The constraint is not demand. Cox's data says the car park is aging into the dealer's favor for the next several years. The constraint is not headcount. Three groups proved that by adding it. The constraint is visibility — the plain inability to see, in real time, where a vehicle is, where its key is, what stage it is in, and how long it has been sitting there.

Dealerships that stay reactive will keep managing fixed operations by reading yesterday's gross. Dealerships that move to real-time operational control will manage the minutes, and the gross will follow. That shift — from reporting on the service drive to running it — is the whole opportunity Q2 just put on the table.