Automakers Now Expect AI to Raise Dealer Profits. The Real Question Is Where.

An unusual thing just happened: the automakers took AI's side

For three years, "AI at the dealership" sat in the same drawer as every other vendor pitch - interesting, unproven, easy to defer to next quarter. That framing got harder to defend this month.

In its 2026 OEM Survey, the buy-sell advisory firm Kerrigan Advisors asked roughly 155 manufacturer executives - the people who study dealer economics for a living - what they expect AI to do to dealership profitability over the next five years. Fifty-nine percent said it will increase profits. Just 4% said it will lower them, with the rest expecting no change (Kerrigan Advisors 2026 OEM Survey, reported by Car Dealership Guy and CBT News, July 2026). It was the survey's largest response in its four-year history.

That is a striking vote of confidence from an audience with no reason to hype dealership software. And it matches what dealers have been signaling themselves: in Kerrigan's 2025 Dealer Survey, 90% of dealers said they already use AI or plan to. The industry's question has quietly moved from "does this work" to "where does it pay."

Myth: AI in the showroom is still a science project - lots of pilots, little proven profit.

Data: 59% of OEM executives expect AI to increase dealership profitability over the next five years; only 4% expect a decrease (Kerrigan Advisors 2026 OEM Survey, ~155 respondents). The manufacturers grading their own dealer networks are betting on the upside.

Why the factory cares about your bottom line

It is worth pausing on who is making this forecast. Manufacturers do not answer surveys to flatter their dealers. They track dealer profitability because a well-capitalized, profitable network is the one that can afford to build the facilities, floor the inventory, and fund the customer experience the brand depends on. When 59% of them expect AI to lift dealer margins, they are describing a lever they expect their own retail channel to pull.

Kerrigan is specific about the mechanism. The firm expects wide-scale AI deployment to reduce selling expenses and lift revenue per employee - which is, in plain terms, the definition of margin expansion for a retail business. You either capture more gross from the same traffic, or you produce the same output with less overhead. AI's profit case in auto retail is some combination of those two things, and nothing more exotic.

The market is already paying for operating leverage

If the survey is the opinion, Carvana is the price tag. Kerrigan singled out the online retailer - whose AI-enabled model has grown retail sales 91% year over year and carried its market value to roughly $80 billion, more than the combined worth of all six public dealer groups - as the clearest evidence of what operating leverage is now worth (Kerrigan Advisors, via CBT News, July 2026).

The smaller detail is the more instructive one. According to Kerrigan, Carvana's first new-vehicle store - a Stellantis franchise in Casa Grande, Arizona that had been selling about 35 cars a month - sold close to 1,000 new vehicles in June, reportedly without a general manager. Treat that specific number as a reported anecdote, not an audited result; one store's single month is not a trend, and extraordinary figures deserve a raised eyebrow. But the direction is the whole point. What the market is rewarding is not "selling cars online." It is throughput per rooftop - pushing far more volume through the same physical footprint and headcount.

That distinction is good news for a franchised dealer, because throughput is reproducible in a way an $80 billion balance sheet is not. You cannot copy Carvana's cost of capital. You can copy the discipline of not leaking the demand you already pay to create.

Where AI actually moves the number

Strip the profit thesis down and it rests on two levers - the same two Kerrigan named.

Lever one: capture more of the demand you already have. Most stores lose a meaningful share of ready buyers before a salesperson ever touches them. Foureyes' 2026 benchmarks, built on more than a billion dealer-site visits, found 42.7% of qualified sales leads were mishandled - a missed call, a dropped follow-up, or a lead never worked at all. The phone is the leakiest pipe of all, and it is where the highest-intent shoppers go. Answering every call, working every lead, covering nights and weekends, and booking the service appointment on the first contact is not glamorous work - but it is gross profit the store already paid to generate and then let walk out. (We have put numbers on that leak in what missed calls really cost and why speed-to-lead decides so many deals.)

Lever two: lower the cost of each sale. This is the "revenue per employee" half of Kerrigan's thesis. A voice agent that handles status checks, routing, after-hours calls, and routine service scheduling lets a fixed BDC and service team cover far more contacts without adding headcount - the repetitive, high-volume, off-hours work that quietly burns labor. That is the same math the fixed-ops side of the store runs on: capture the demand that reaches you, and do it without proportionally growing cost.

Neither lever is a moonshot. Both are what the OEM executives are quietly pricing in when they forecast higher dealer profits.

The backdrop that makes this urgent

None of this is happening in a growth market. The same survey found 45% of OEM executives expect their networks to have fewer dealers in five years - up sharply from 33% a year earlier. Consolidation is the base case: fewer, larger, better-capitalized groups, not a retreat from the franchise model. Buy-sell activity is expected to stay hot - 88% of executives expect deal volume to hold or rise - and blue-sky values to hold or climb (82%). The strong stores are getting more expensive to buy, and the pressure to perform per rooftop is rising with them.

Kerrigan's own conclusion points the same way. Dealers, she argues, "need to focus on growth, both on a rooftop level in terms of growing the revenue and market share of each of their stores," and on scale within their markets. In a market that is not adding customers - Cox Automotive still projects full-year 2026 U.S. sales near 15.8 million units, down about 2.9% from 2025 - you grow revenue per rooftop by capturing more of the demand already coming at you. That is precisely the job AI is being bought to do.

Source: Kerrigan Advisors 2026 OEM Survey, about 155 OEM executive respondents, fielded December 2025 to June 2026.

The measured version

Optimism is not ROI, and a survey of manufacturers is a forecast, not a receipt. A few cautions belong next to the headline number.

• "AI will help profits" is not "any AI tool will help yours." The gains show up in narrow, measurable jobs - answered-call rate, leads logged, appointments set, after-hours capture - not in a general-purpose bot sprayed across the store. Pick the job first, then the tool.
• Measure the baseline before you buy. If you cannot state your current answer rate, after-hours miss rate, and average speed-to-answer, you cannot prove a fix. The stores that see ROI instrument the phone the way they already instrument the website.
• Keep humans on the high-stakes moments. The best deployments know their limits and hand off cleanly on price, financing, and anything emotional. Automation that captures and routes, with people closing, is the pattern the evidence keeps supporting - a theme we go deeper on in the real ROI of agentic AI at the dealership.
• Compliance does not get an AI exemption. Outbound calls and texts still live under the TCPA and state rules, and "the software did it" is not a defense. Confirm consent and opt-out handling before anything dials out - see our rundown on AI calling and compliance.

What to do with this

The useful takeaway from the 2026 OEM Survey is not "buy AI." It is that the people with the clearest view of dealer economics now expect the profit to be there - and they expect it to come from lower selling cost and higher output per person, in a market consolidating around the stores that execute.

So start where the leak is biggest and the math is clearest. For most stores, that is the phone: the highest-intent channel, the one that most often rolls to voicemail after hours, and the one where a single missed call can hand a ready buyer to the dealer across town. Fix that, measure it honestly, and you are pulling the exact lever the manufacturers are betting on.

The manufacturers, the market, and the buy-sell data are all pointing at the same thing: in a flat, consolidating market, the winners will be the stores that capture more of the demand they already have. That is the problem Carbuki was built for - AI voice agents for U.S. dealerships that answer every call, book the appointment, and make sure no lead goes unlogged. If you are trying to figure out where AI actually moves your number, it is a conversation worth having.

Sources

• Kerrigan Advisors, "2026 OEM Survey" (report overview): https://www.kerriganadvisors.com/our-reports/oem-survey
• Business Wire, "AI to Lift Dealership Profits as Automakers Plan to Absorb the Bulk of Tariff Costs, 2026 Kerrigan OEM Survey Finds," July 7, 2026: https://www.businesswire.com/news/home/20260707469261/en/AI-to-Lift-Dealership-Profits-as-Automakers-Plan-to-Absorb-the-Bulk-of-Tariff-Costs-2026-Kerrigan-OEM-Survey-Finds
• Car Dealership Guy News, "Survey finds that OEM execs predict fewer dealers, steady buy-sells," July 15, 2026: https://news.dealershipguy.com/p/survey-finds-that-oem-execs-predict-fewer-dealers-steady-buy-sells
• CBT News, "2026 OEM Survey reveals the future of dealership consolidation and profitability," July 14, 2026: https://www.cbtnews.com/2026-oem-survey-reveals-the-future-dealership/
• The Motley Fool, "Prediction: Carvana's New-Car Business Will Work. Early Numbers Are Stunning," July 11, 2026: https://www.fool.com/investing/2026/07/11/prediction-carvanas-new-car-business-will-work/
• Foureyes, "2026 Automotive Dealer Benchmarks Report," 2026: https://www.foureyes.io/blog/2026-automotive-dealer-benchmarks-report
• Cox Automotive, "Cox Automotive Forecast: New-Vehicle Sales Pace Holds Strong Through First Half of 2026," June 24, 2026: https://www.coxautoinc.com/insights/cox-automotive-forecast-june-2026-u-s-auto-sales-forecast/