Agentic Commerce Before Black Friday: A Pre-Peak Readiness Checklist

The First Peak Season With AI Checkout in Play

Black Friday falls on November 27, 2026, with Cyber Monday on November 30. It is the first holiday peak where several AI surfaces can not only recommend products but also complete purchases: Google's agentic checkout in AI Mode and Gemini (built on the Universal Commerce Protocol), Instant Checkout in ChatGPT, and TikTok's new Shopping Assistant and Buy Direct.

That does not mean AI agents will carry your holiday revenue. Analysts covering the season expect agent-completed purchases to remain a small share of total ecommerce this year, and paid search, email, and marketplaces will still do most of the work. But AI assistants are increasingly where gift research starts, and the stores they recommend in late November are chosen from data that has to be correct before your code freeze.

This checklist is about doing the high-leverage work in the next two weeks and deliberately skipping the rest.

What to Leave Alone

Start with restraint. Peak season is the worst time to introduce risk, and most agentic commerce work is not urgent enough to justify it.

Do not rebuild your promo calendar, shipping cutoffs, or returns policy around AI surfaces. They should stay consistent across every channel.
Do not start a new protocol integration (a custom UCP or ACP endpoint) in the final weeks before peak unless your platform already ships it as a toggle. Shopify, BigCommerce, and other platforms are rolling out native connectors; let them carry the engineering.
Do not install a pile of new "AI visibility" apps during the freeze window. Every new script is a new failure point on your highest-traffic days.

If you want the full picture of which protocols matter and why, our breakdown of ACP, AP2, and the other agentic commerce protocols covers it. For peak season, the short answer is: use what your platform gives you and focus on data.

What to Fix Now

1. Your top revenue SKUs, not your whole catalog

You will not fix thousands of products in two weeks. Pull your top products by holiday revenue from last year plus this year's planned hero items, and audit only those. For each one, check:

Title states the product type, brand, and key attribute in plain language ("Merino Wool Crew Sock, Men's, 3-Pack"), not a campaign slogan.
Description answers the questions shoppers ask assistants: materials, dimensions, compatibility, who it is for, and what is in the box.
Identifiers (GTIN, MPN, brand) are present and correct. Agents use them to match your product across sources and reviews.
Variants are grouped correctly, with each size and color carrying its own price, availability, and image. See our guide to variant data for AI search if this is a known weak spot.

2. Price and availability that hold up under promo pressure

Holiday promotions are exactly when feeds, structured data, and the live page drift apart. A sale price that updates on-site but not in your feed, or a sold-out variant still marked in stock, is the kind of mismatch that can get a listing suppressed or a checkout rejected.

Schedule sale prices in your feed with explicit start and end dates rather than editing them by hand at midnight.
Confirm that your Offer schema, Merchant Center feed, and any OpenAI or platform feed all read from the same source of truth.
Spot-check your hero SKUs after each promo change: page, schema, and feed should show the same price and stock status.

3. Shipping deadlines and return windows as data, not banners

"Will it arrive by Christmas?" is one of the most common questions shoppers put to an assistant in December. A site-wide banner with your cutoff dates is useful for humans but often invisible to an agent.

Publish shipping speeds, costs, and holiday cutoff dates in Merchant Center shipping settings and in structured data where your platform supports it.
Make sure your returns policy, including any extended holiday return window, is set in Merchant Center and stated plainly on a crawlable policy page.

4. Google agentic checkout eligibility

If you are one of the merchants with access to Google's agentic checkout, the requirements are specific. According to BigCommerce's guide to Google agentic checkout, as of September 2026:

Your Merchant Center account must be in good standing, with complete feeds covering titles, GTINs, price, availability, shipping, returns, and a support contact.
Only listings carrying the native_commerce(checkout_eligibility) attribute show a Buy button, and it is set per SKU. BigCommerce recommends adding it through a supplemental data source so existing Shopping ads and free listings are untouched.
You need a Google Pay & Wallet Console account, acceptance of the UCP checkout terms, and a payment processor that can accept Google Pay tokens.
Access is still limited to select U.S. merchants, with broader Merchant Center onboarding rolling out gradually.

If you do not have access yet, the same feed hygiene still pays off: those attributes power AI Mode product results whether or not a Buy button appears.

5. Crawler and agent access on your busiest days

Bot protection tends to get tightened before peak to fend off scalpers and scrapers. Check that the change does not also block the AI crawlers and shopping agents you want. Review your robots.txt, your CDN's bot settings, and any rate limits, and confirm that product pages render their key content without depending on client-side JavaScript.

Measure Before and After the Peak

AI-referred sessions often land in analytics as direct or unattributed traffic, which hides whether any of this work paid off. Before peak:

Add channel rules for referrers from ChatGPT, Perplexity, Gemini, Copilot, and other assistants so they report as their own group.
Record a baseline: run a fixed set of gift-style prompts in your category ("best gift for a home barista under $100") and note whether your brand appears.
Rerun the same prompts after Cyber Monday and again in mid-December.

Our guide to measuring AI search visibility covers the prompt set and tracking setup in more detail. If you want this run for you on an ongoing basis, that is what our visibility tracking service does.

A Two-Week Plan

This week: pick your hero SKUs, audit titles, descriptions, identifiers, and variants, and fix attribution rules.
Next week: lock promo pricing into scheduled feed updates, publish holiday shipping and returns data, and review bot settings.
Freeze: no new apps or integrations through Cyber Monday. Spot-check price and stock parity on hero SKUs daily.
After peak: compare AI visibility against your baseline and decide what deserves real investment in 2027, including protocol work you deferred.

Agentic checkout is real this season, but small. The stores that win it are the ones whose product data was already accurate when the assistant went looking. If you are not sure where your catalog stands, an AI visibility audit is the fastest way to find the gaps before the freeze.