63% of Shoppers Will Use AI Next Purchase; Only 29% of Dealers Ready

By Elena Vasquez

63 per cent of vehicle shoppers plan to use AI for their next purchase, according to a new Cox Automotive AI in Auto Retail Tracker, while only 29 per cent of dealers have begun adjusting to AI-powered search. That gap is the number the aftermarket should pin to the wall. New-vehicle retail is showing the stress fracture first, and parts is unlikely to stay intact.

The data gets worse on close inspection. The share of dealers who know they need to adjust but have not started rose from 26 to 32 per cent. Awareness, in other words, is growing faster than action. So is complacency dressed as activity: 82 per cent of dealers use AI in some form, mostly for task automation and content generation, yet only 22 per cent report the sales growth that 69 per cent expect.

This is new-vehicle data, not parts data. But the same disconnect is already visible on the aftermarket side, with tools adopted, impact untracked and results assumed rather than confirmed. Sound familiar. It should.

Does fitment complexity protect parts sellers?

The instinct says yes. A brake rotor is not a running shoe, and AI cannot casually recommend one without year, make, model and trim. Wrong. Complexity does not block AI mediation. It raises the bar for who gets recommended while lowering the bar for misleading or false information.

Survey data from global consulting firm Oliver Wyman, spanning nearly 300,000 respondents, found that 62 per cent of consumers across generations have already used AI to find and purchase products. The zero-click, AI-first customer may never land on a website at all. They may never see the merchandising, the pricing ladder or the brand story — only whatever the AI decided to repeat.

Take the obvious query. An AI agent summarizing "best oil filter for a 2021 RAV4" pulls from whoever has clean, crawlable, structured data. A thin digital catalogue is not just harder to find. In the future, it may not be found at all.

Canada's position makes this sharper. Its online parts presence and penetration have run comparatively weaker than most advanced economies. Historically that gap meant lost visibility, softer price transparency and a smaller share of a smaller pie — a nuisance. Under AI-mediated shopping, it becomes closer to structural exclusion.

What is GEO, and why does it replace SEO?

The common thread across the value chain is GEO — generative engine optimization — replacing SEO as the operative discipline. The difference is mechanical, not semantic. SEO optimized for ranking and click-through. GEO optimizes for becoming a source that an AI model cites or paraphrases, often without ever sending a click.

The citation is the new impression, whether the shopper ever visits the site.

How should each tier of the market respond?

• Distributors and retailers: start with the catalogue itself. ACES/PIES compliance is no longer a back-office exercise. Clean fitment attributes, OEM cross-references and structured feeds are the entry ticket to being legible to an AI agent at all.
• Workshops: local visibility shifts away from keyword-stuffed service pages toward structured business data — service menus, schema markup and FAQ content that answers diagnostic questions directly rather than burying them in a blog post nobody clicks. "Who performs ADAS calibration near me?" gets answered from structured profiles now, not backlink counts.
• Suppliers: the marketing shift runs from optimizing for search rank to optimizing for citation. Machine-readable spec sheets and technical content that directly answers questions such as "Why does X fail?" or "What is the best X for Y?" get pulled into AI summaries. Marketing budgets built entirely around SEO keyword strategy are optimizing for a channel that is shrinking as a share of discovery.

How urgent is this, really?

Honest answer: nobody knows yet whether AI-mediated shopping becomes the dominant discovery channel for aftermarket parts or simply another layer alongside search, marketplaces and counter relationships. Probably somewhere in between. The Cox Automotive gap documented in new-vehicle retail is, however, a pattern the aftermarket will likely repeat in a year or two.

The economics favor moving early. Getting the catalogue and content foundation right now costs relatively little, especially with new AI tools available. Rebuilding it after the gap has widened — and after the AI agents have already learned to route around you — will not.