Cox Says 82% of Dealers Now Use AI. Only 29% Have Started on the Part That Decides Whether Shoppers Find You.

On August 11, Cox Automotive published the first edition of a new quarterly study called the AI in Auto Retail Tracker. It surveys both sides of the counter: 504 franchise and independent dealers plus 1,505 in-market consumers in Q1, then 483 dealers and 1,502 consumers in Q2.

The headline number is the comfortable one. 82% of dealers report using AI today. Adoption is basically settled. If you are running a store and you have been feeling behind on this, the survey says you are not.

Then there is the number underneath it, and it is the one worth a meeting.

63% of in-market shoppers say they will definitely or probably use AI on their next vehicle purchase. Only 29% of dealers have started to adjust to AI-powered search.

Those two figures come from two different survey populations, so do not subtract them and treat the difference as a score. Read them for what they are: the shopper side has already moved, and roughly seven in ten dealers have not started moving with it. Cox titled the release around exactly that gap, and they are right to.

The 82% is real, and it is pointed at the wrong wall

Look at what dealers are actually buying. Cox lists the top three uses of AI in a store as automating routine and complex tasks (40%), coordinating customer follow-up (40%), and generating content and creative (38%).

Every one of those lives inside your building. Task automation, follow-up cadence, ad copy and descriptions. Useful work, all of it. It also happens entirely after a shopper has already found you.

The disruption Cox is measuring is happening before that. It is happening in the moment a shopper opens ChatGPT or Google's AI Mode and asks which dealership near them has the truck they want, at a price that makes sense, with a service department worth trusting. That moment is upstream of your CRM, your chat widget, and your follow-up cadence. None of your AI tools are in the room for it.

So the 82% and the 29% are not in conflict. Dealers bought AI for the parts of the business they can see and control. The part they cannot see is the part that decides whether the shopper ever arrives.

We have watched this exact movie before

Rewind to the mid-2000s. Dealers adopted the internet fast, in the sense that everybody got a website and everybody got a CRM. The tools went in. The department got a name.

What most stores did not do for another five years was accept that the shopping had moved off their property entirely, onto third-party listing sites and search results they did not own. Those dealers had a website. They just did not have a presence anywhere the buying decision was actually being made. The stores that figured that out early, that treated their listings and their search footprint as the real storefront, took share from the ones who kept polishing a site nobody was starting at.

The tooling era always arrives before the visibility era. Adopting the technology is the easy half, because you can buy it. Restructuring around where the customer went is the hard half, because you have to change what you publish and how you are described by systems you do not own.

We are at exactly that point again, and Cox has now put a number on it.

The results gap, stated plainly

The other finding worth your attention is uncomfortable. 69% of dealers expected AI to drive sales and revenue growth. Only 22% of AI users report seeing that growth so far.

Before anyone reads that as proof AI is a bust, read the line Cox published next to it: about 1 in 3 dealers either are not measuring AI's impact at all or have no clarity on how they are measuring it.

You cannot report growth you never instrumented. A store that turned on an AI follow-up tool nine months ago, never tagged the traffic, never separated AI-assisted appointments from the rest, and never compared to a baseline, has no way to answer the question. It will honestly report no growth, because nobody built the meter.

Notice what dealers did report: the most commonly cited benefits were improved customer experience (28%) and employee productivity (26%). Those are the things you feel without measuring. Revenue lift is the thing you have to prove, and most stores did not set up the proof.

Read that 22% as a measurement failure before you read it as a product failure. Most dealerships bought AI without a scoreboard and now cannot tell either way.

About that "external partner" statistic

Cox also found that dealers working with an external AI partner were more likely to say they are using AI optimally (66% vs. 46% without), more likely to express high confidence in AI outputs (30% vs. 8%), and more likely to report sales and revenue growth from AI (36% vs. 21%).

We are an external partner. So take this next part as our own disclosure rather than our sales pitch: that is a correlation, and Cox drew it from the combined Q1 and Q2 base rather than a controlled comparison. Stores that hire an outside partner for AI tend to be stores that already committed budget, already assigned an owner internally, and already decided to measure something. Some of that 20-point gap is the partner. Some of it is that the kind of dealership that hires help was going to execute better anyway.

Anybody who quotes that stat at you, including us, should say that out loud. What it genuinely supports is narrower and still useful: the dealers getting results are the ones who treated AI as a project with an owner and a scoreboard, not as software they turned on.

What this means for your dealership

The single most actionable line in this study is that 29%, and its trend. Cox found the share of dealers who know they need to adjust to AI-powered search but have not started rose from 26% to 32% quarter over quarter.

Awareness is climbing faster than action. More dealers understand the problem, and the pile of dealers who understand it and are doing nothing is the pile that is growing. Knowing does nothing for you.

Concretely, here is what "adjusting to AI-powered search" means at a franchised store, and none of it is exotic:

Your inventory pages are the thing being read. When an assistant answers a shopper's question about a specific trim at a specific price, it is pulling from structured, machine-readable inventory data. If your VDPs bury the answer under a hero image and boilerplate, or your feed is stale, you are not in the answer.

Your Google Business Profile is customer-facing copy now. Hours, address, phone, categories, and departments for every rooftop. AI assistants lean heavily on local records, and a wrong service-drive phone number does not get corrected by a chatbot.

Your reviews are training material for how you get described. The corpus of what people say about your store is what an AI summarizes when someone asks if you are worth visiting.

Your crawler access decides whether you are eligible at all. If you are blanket-blocking AI crawlers at the WAF, you have opted out of the 63%. We have written about why that backfires.

And Cox's consumer side adds one more thing worth planning for: 24% of shoppers say AI already helps them feel more prepared when working with dealerships. Lori Wittman, President of Retail Solutions at Cox Automotive, framed it as AI "bringing dealers a more informed, more decisive buyer." Your salespeople are going to meet more customers who arrive with an AI-assembled version of your pricing and your reputation. That version is either accurate or it is not, and you influence it upstream, before the walk-in.

What to do about it

1. Assign an owner for AI-powered search visibility this week. Not the DMS admin, not the OEM website vendor by default. One person accountable for how your store reads to an assistant.
2. Build the scoreboard before you buy anything else. If you cannot say what AI-influenced traffic and appointments looked like last month, fix that before renewing a single AI tool.
3. Audit every rooftop's Google Business Profile and third-party listings for accuracy. Old DBAs from a prior owner are the usual culprit at franchised stores.
4. Fix the top of your highest-intent pages so the answer is in the first 200 characters, in plain language, ahead of the boilerplate.
5. Confirm you are not blocking AI crawlers at the WAF or in robots.txt, and check that separately from your SEO vendor's report.
6. Ask every AI vendor in your building what they measure and how often they sample. Confidence without a range is a sales technique.

The gap is the opportunity

Two thirds of your shoppers are heading into their next purchase planning to use AI. Seven out of ten dealers have not started adjusting to that. This is what a market looks like right before the early movers take share, and it will not stay this open, because the 32% who know and have not started will eventually start.

The work is unglamorous. Clean data, accurate listings, readable pages, real answers to real buying questions, and a way to tell whether any of it moved. It is the same foundation that won the listings era, and it is worth more now because more of the decision happens before anyone talks to you.

If you want to see what an AI assistant currently says about your store, and where the gaps are, book a walkthrough and we will go through it with you.