Why Most Retailers Aren't Ready for the AI Shopping Agents Showing Up This Holiday Season

Two numbers from this year's holiday retail research should sit next to each other on every ecommerce leadership team's whiteboard. Narvar's 2026 Holiday Shopping Report, based on a survey of 100 retail decision makers and 1,348 US consumers, finds that 65 percent of shoppers plan to use AI for at least one part of their holiday shopping. The same report finds only 8 percent of retailers describe themselves as very confident using AI to improve that experience. That is not a small gap. It is 57 points, and it is the gap between what shoppers are already doing and what retailers believe they are ready to support.

What shoppers are actually doing with AI this season

The 65 percent headline breaks down into specific, useful behavior rather than vague AI curiosity. Narvar finds 43 percent of consumers using AI for gift discovery, 33 percent for product comparison and review summaries, and 29 percent for budgeting and holiday spend planning. A separate Optimove 2026 report puts regular or occasional AI consultation for shopping ideas and gift recommendations at 72 percent, with ChatGPT, Claude, and Gemini named as the tools people actually reach for. These are not fringe behaviors anymore. A majority of your holiday customers are starting at least part of their research in a chat interface before they ever land on your site, and 78 percent say they would use AI powered tools more if doing so made shopping feel more personalized, which means the appetite is still growing, not plateauing.

The traffic is already showing up, whether or not you are ready

Narvar's numbers describe intent. Adobe's data describes what already happened. AI driven traffic to retail sites rose 693 percent during the 2025 holiday season, with Cyber Monday alone seeing a 670 percent spike. That traffic converts differently than a typical visitor: Adobe found AI referred shoppers spend 48 percent longer on site, view 13 percent more pages, and convert 42 percent better than other traffic. Salesforce is forecasting that 20 percent of all 2026 holiday ecommerce traffic will come from AI chat agents directly, not just AI influenced human browsing. If your infrastructure planning for this holiday season was built around last year's traffic mix, these are not incremental adjustments, they are a structurally different traffic composition than the one most commerce platforms were architected around five years ago.

Why the confidence gap is really an infrastructure gap

The 8 percent confidence figure is not retailers underestimating AI's potential. It tracks closely with a harder, more technical problem: most retail product data is not actually built to be read by an agent. Structuring a product catalog so an AI shopping agent can reliably parse pricing, availability, and variants is a real integration project, not a settings toggle, and most commerce platforms were not designed with that consumer in mind. Adobe separately found that roughly 34 percent of retail product pages are effectively inaccessible to AI systems today, whether because of JavaScript rendering agents cannot execute reliably, inconsistent structured data, or catalog information scattered across systems that were never meant to serve a machine reader directly. A retailer cannot be confident serving a shopping agent well when a third of its own catalog is invisible to one.

What the 6.2 percent versus 3.9 percent growth gap does and does not tell you

Salesforce reports that retailers who ran a branded AI shopper agent during the 2025 holiday season saw 6.2 percent sales growth, compared to 3.9 percent for retailers that did not. That is a real, meaningful difference, and it is also a correlation drawn from retailers that chose to invest in agent infrastructure early, which likely describes organizations that were already ahead on commerce technology more broadly. Treat it as directional evidence that early agent investment correlates with stronger holiday performance, not as proof that bolting a chat widget onto an existing storefront produces a guaranteed 2.3 point lift. The retailers actually capturing that advantage are the ones that solved AI discoverability at the catalog and storefront level first, not the ones that treated an agent as a marketing layer on top of unchanged infrastructure.

Delivery expectations have not relaxed, even as AI adoption rises

One more data point worth planning around: 76 percent of consumers say they would accept slower delivery in exchange for free shipping, but nearly 30 percent still expect same day or next day delivery regardless. AI adoption is rising at the discovery and research stage of the shopping journey, but it has not lowered the operational bar at fulfillment. A faster, more personalized front end still has to connect to a fulfillment and inventory system that can back up the promise an AI powered experience makes.

When this applies to your stack

If your holiday readiness checklist this year covers load testing, payment uptime, and promotional calendar configuration but has nothing specific about whether an AI shopping agent can actually parse your product pages, that is the gap worth closing before traffic peaks, not after. Closing the confidence gap means catalog structure, page accessibility to agent crawlers, and checkout flows that can handle non human initiated transactions, which is exactly the headless commerce and platform infrastructure work we help retailers get in front of before the season that makes or breaks the year.

FAQ

What percentage of holiday shoppers plan to use AI this year? Narvar's 2026 Holiday Shopping Report finds 65 percent of consumers plan to use AI for at least one part of their holiday shopping, most commonly gift discovery, product comparison, and budgeting.

Why are retailers so much less confident than consumers are ready? A large part of the gap traces to infrastructure, not sentiment. Adobe found roughly 34 percent of retail product pages are effectively inaccessible to AI systems today, which limits how well any AI shopping experience can actually perform on top of existing catalog data.

Does running a branded AI shopping agent actually increase holiday sales? Salesforce reports retailers running one in 2025 saw 6.2 percent holiday sales growth versus 3.9 percent for those without, though this is correlational data from retailers that chose to invest early, not a controlled experiment.

How much of holiday ecommerce traffic will come from AI agents directly? Salesforce forecasts 20 percent of all 2026 holiday ecommerce traffic will come from AI chat agents, separate from human shoppers who are influenced by AI research earlier in their journey.