Prediction: The First Real 'Agentic' Holiday Shopping Season Is Coming

My bet for this year: Q4 2026 is the first holiday season where AI shopping assistants meaningfully influence what people buy. With ChatGPT reportedly fielding around 50 million shopping-related queries a day, agentic commerce protocols live, and "discover in AI, buy on site" now a normal path, the assistant increasingly shapes the shortlist before anyone reaches your product page. This is a prediction — but the 2026 groundwork is already in place.

The prediction

This holiday season, the buying decision will more often start with "ask the AI what to get" than with a search box. The retailers who win won't be the ones with the biggest ad budget — they'll be the ones whose product data, reviews, and availability an AI can read, trust, and recommend by name.

What's driving this?

Agentic commerce got real in 2026. OpenAI built shopping directly into ChatGPT, then reworked the experience around research and recommendation after its Instant Checkout experiment; Stripe and OpenAI shipped the Agentic Commerce Protocol; and Google introduced its Universal Cart approach. Early retailer data is striking — Walmart reported ChatGPT drove roughly double its normal new-customer rate. Add ~50M shopping queries a day and you have the raw ingredients for a genuinely agentic holiday season.

What could this look like in Q4?

Shoppers asking "what's the best gift under $100 for a runner?" and getting a short, named list — then buying from whichever retailer the assistant surfaced, usually on that retailer's own site. Availability and price accuracy become conversion factors, because an assistant that recommends an out-of-stock item looks broken. The brands that show up are the ones with catalogs an engine can parse and reviews that signal trust.

How should retailers prepare now?

1
Clean up your product data
Ensure every product page has structured, accurate price, availability, specs, and identifiers an AI can lift. Ambiguity gets you left off the shortlist.
2
Build review depth and freshness
Assistants lean on social proof. Recent, substantive reviews make you the safer recommendation than a competitor with none.
3
Answer buying questions on-page
Write answer-first content for the real questions — sizing, compatibility, returns — so the model can quote you confidently.
4
Confirm crawlers can reach you
Check that the AI crawlers driving commerce can actually access your catalog. Being unreadable is the same as being unavailable.

For the deeper playbook, see how AI recommends products and agentic commerce. For where the checkout story went, read why ChatGPT's Instant Checkout was discontinued.

Frequently asked questions

Will people really buy through AI this holiday season?

My prediction is that AI will more often influence the purchase than complete it — people research in ChatGPT or Gemini, then buy on the retailer's own site. Either way, being the option the assistant recommends is what matters, and that's decided before the click.

It's July — is it too late to prepare for Q4?

No, but the window is closing. Product data, reviews, and availability need to be clean and crawlable before the shopping season ramps. Starting now gives engines time to index and trust your catalog ahead of peak demand.