See How AI Sees Your Dealership

Organic traffic is down at most dealerships right now, and the biggest single cause is that AI is answering shopper questions before the click ever happens. Google AI Overviews, ChatGPT and Gemini now handle a growing share of the research that used to land on your VDPs. The work still gets done. You just cannot see it in Analytics anymore. Falling sessions on their own do not tell you whether you are winning or losing. What matters is whether the AI is recommending your store or the one down the road. We break down the zero-click funnel in detail on the blog.

If your phone and your floor are holding steady while form fills drop, the problem is at the top of the funnel, not in your BDC. Shoppers who finish their research inside an AI assistant often skip the lead form entirely and walk in or call direct. Check your show rate and closing ratio before you change anything in the CRM. If those are flat or up while lead count falls, you are looking at the zero-click funnel, not a lead problem.

On research questions, almost certainly yes. AI Overviews answer the question directly on the results page, so pages that used to earn clicks now earn impressions instead. The effect is heaviest on informational searches and lightest on branded and transactional ones. The response is not to fight harder for the click. It is to make sure your store is the one being quoted inside the answer. More on how AI Overviews answer before the click.

Yes, and the share is growing quickly. Shoppers use AI assistants to shortlist dealers, compare two stores of the same brand, check reputation, and decide where to get a vehicle serviced. The consideration set gets narrowed to two or three names before anyone opens a dealership website. If your store is not in that shortlist, you never enter the process at all. Here's which AI assistants shoppers actually use.

Ask it directly. Open ChatGPT, Gemini and Claude and type the questions your customers would type, like "best BMW dealer near Bridgewater NJ" or "where should I service my car in Somerville". Read what comes back about your store and about your competitors. To do it systematically instead of one query at a time, the free check on this page runs a standard set of real shopper questions across all three engines and scores how your store comes back.

Usually because the AI can read them more clearly than it can read you. Engines lean on structured data, review volume and sentiment, and business information that stays consistent everywhere it appears. A competitor with clean schema, recent reviews, and matching hours and address across the web is simply easier to recommend with confidence. This is rarely about who sells more cars. It is about who is easier for a machine to describe without guessing. We wrote about how AI picks between same-brand dealers.

Because SEO and AI visibility are two different problems. SEO gets you ranked in a list of links. AI visibility decides whether an engine names you inside a sentence. A site can have good page speed, clean meta tags and solid rankings while still failing on what AI depends on: schema markup, entity consistency, and content an engine can extract and quote. Both matter. Most vendors are only measuring the first one.

It might be, and most dealers have never checked. Some platforms block AI crawlers in robots.txt by default, and many render inventory in JavaScript that non-rendering crawlers never see. When an engine cannot fetch or parse your pages, it falls back to third-party sites to describe you. Checking takes a few minutes: look at your robots.txt for GPTBot, ClaudeBot and Google-Extended, then load a VDP with JavaScript turned off and see what is left. Here's how to tell if your website is blocking AI crawlers.

Structured data first. Vehicle and Offer schema on your VDPs, AutoDealer or LocalBusiness schema with accurate hours, address and phone, and page content that exists in the HTML rather than only in JavaScript. Then keep it current, because engines quoting last year's model year or an expired lease offer is one of the most common and most damaging failures we see. When the reader is a machine, machine-readable beats beautifully designed. Start with the schema markup layer AI actually reads.

No, and this one trips up a lot of people. Your CRM and DMS are internal systems and AI engines never see them. What engines see is your website, your Google Business Profile, your reviews, and the third-party listings carrying your inventory. Your CRM decides what happens after a lead arrives. It has nothing to do with whether that lead gets created in the first place.

Constantly. These engines generate a fresh response every time, so the same question can return different dealers on different days. Answers also shift as your reviews, your website and your competitors change. Check monthly at minimum, and watch the trend line rather than any single result. One bad answer is noise. A three-month slide is a problem. There's a difference between being mentioned and being recommended, and it moves over time.

Right now it is usually nobody's, which is exactly why it goes unmanaged. Your website provider owns the platform but rarely owns schema strategy. Your marketing agency owns paid and social and reports on clicks. Neither one is measuring whether ChatGPT recommends your store. Until somebody owns it explicitly, the honest answer is that it lands with you. Start by measuring it, because you cannot assign a problem nobody has quantified.