Agentic commerce lets a shopping assistant help a buyer choose or purchase a product. For brands, preparation starts with accurate product descriptions, prices, availability and fulfillment information. Evaluate each shopping surface separately: appearing in an answer, supplying product data and supporting checkout are different capabilities.
Why this matters now, not next year
AI shopping agents are already helping people choose and buy products. ChatGPT's Instant Checkout, codeveloped with Stripe on the open Agentic Commerce Protocol, opened with Etsy and is rolling out to over a million Shopify merchants including Glossier, SKIMS, Spanx, and Vuori (OpenAI). Google launched the Universal Commerce Protocol on January 11, 2026 with Shopify, Etsy, Wayfair, Target, and Walmart, plus 20+ endorsers including American Express, Mastercard, Stripe, and Visa (Google). McKinsey now estimates agentic commerce could orchestrate $1 trillion in U.S. retail revenue and $3–5 trillion globally by 2030 (Digital Commerce 360). The early-mover window is open, and it is closing in retail categories first.
What is actually new for brand visibility
AI shopping changes three parts of the buying journey: how products are found, how they are compared, and where the purchase happens. Agents compare price, availability, delivery and seller reputation using product data. A brand that earns editorial citations still needs accurate feeds and clear fulfillment terms to compete at checkout.
How the major shopping agents pick brands
The three agentic surfaces with the most volume each use different selection logic. The differences matter because optimizing for one does not transfer cleanly to the others, and reading them as one channel will lead to wasted budget.
| Agent surface | Discovery basis | What it weighs at selection | Source of brand record |
|---|---|---|---|
| ChatGPT shopping | Web index plus merchant feeds via ACP | Relevance, availability, price, primary-seller status, Instant Checkout enabled | Third-party metadata and direct merchant integrations |
| Google UCP / AI Mode | Google Shopping graph, Merchant Center feeds | Price, availability, shipping, returns, retailer eligibility, Google Pay/PayPal rails | Merchant Center, retailer-of-record APIs |
| Alexa for Shopping | Amazon catalog, prior shopping history, Rufus IP | Catalog completeness, Q&A, A+ content, reviews, fit-to-intent synthesis | Amazon listings, A+ content, backend attributes |
| Perplexity / Claude | Live web retrieval, no checkout layer yet | Cited source mix at answer time; community sources and review platforms over-indexed | Open web, Reddit, review platforms, brand sites |
ChatGPT explicitly states product results are not ads and not influenced by partnerships, and that merchants for the same product are ranked on availability, price, quality, primary-seller status, and whether Instant Checkout is enabled (OpenAI). Google's UCP runs through Merchant Center and ties checkout to Google Pay and PayPal (Google). Amazon's pivot, announced May 13, 2026, folds Rufus's recommendation logic into Alexa for Shopping, which can both answer queries and take actions on behalf of users (CNBC). The pattern across all three: structured data wins, and the editorial layer is downstream of the feed layer.
If you want to see how AI engines describe your own brand, run a free brand check. It takes a minute.
What changes for brand visibility work
Global agentic commerce by 2030 (McKinsey)
Holiday 2025 retail traffic from generative AI vs prior year (Adobe)
Conversion rate of AI referrals vs other traffic, holiday 2025 (Adobe)
Brands and payments firms backing Google's UCP at launch
- Keep your product feeds complete. Check attributes, availability, return policies and shipping terms in Google Merchant Center, Shopify’s ACP integration and Amazon’s catalog. Agents use this data when choosing products.
- Measure each shopping experience separately. Track ChatGPT product results, Google AI Mode product cards and Alexa shopping recommendations with their own question sets.
- Record what happens after selection. Where your integrations expose them, retain purchase permissions, consent and fulfillment confirmations from ACP, AP2, A2A and MCP. These records help you connect agent activity to transactions.
Where the agentic story misleads operators
McKinsey's $3–5 trillion is a 2030 number, not a 2026 number. Morgan Stanley estimates the more conservative slice at $190–385 billion in U.S. e-commerce by 2030, and Bain at $300–500 billion (Digital Commerce 360). Treat agentic commerce as a high-conviction multi-year build, not a quarter-to-quarter revenue line.
The OpenAI–Stripe scope has already narrowed once. Industry reporting suggests the original vision of buying directly from any merchant inside ChatGPT search results was scaled back, with Instant Checkout focused on a tighter merchant set (Stripe). Protocol races usually consolidate, and the brands that bet on a single protocol take on integration risk if the standards reshuffle. Amazon's discontinuation of standalone Rufus less than 18 months after launch is the latest reminder that surfaces themselves are unstable. The right posture is protocol-agnostic plumbing: clean structured data, an updated Merchant Center feed, an ACP-ready product catalog, and clear API access, so the same source of truth feeds whichever agent surface scales next.
What to ship in the next 60 days
A pragmatic 60-day program for a mid-market brand: audit your Merchant Center feed for completeness on price, availability, shipping, returns, and product identifiers, since UCP-eligible categories require this data to participate at all. Confirm whether your ecommerce platform (Shopify, BigCommerce, or commercetools-class) already supports the Agentic Commerce Protocol and Google's UCP; if it does, enable the integrations and verify the test transactions in a sandbox. Build a 25–50 prompt set covering buying-intent queries in your category, run them in ChatGPT shopping, Google AI Mode, and Alexa for Shopping, and log which competitors are appearing where. For the cross-platform context behind these differences, see how AI platforms differ on brand recommendations and the brand-impact data in Google AI Overviews: what the data shows about brand visibility. The goal of the first 60 days is not lift; it is baseline visibility and protocol readiness, so you can tell what is happening before you commit budget to changing it.
How to think about ROI before the channel scales
Agentic commerce attribution will be cleaner than AI-referral attribution in regular search, because the buy happens inside the agent surface and the protocols carry transaction identifiers. That is the reason the channel is worth instrumenting before it is the largest one. Adobe's 2025 holiday data showed AI-referred retail traffic converting 31% more than other sources, with revenue per visit up 254% year-over-year (Adobe). Those numbers describe AI-influenced traffic landing on your site; the agentic commerce variant collapses that funnel further by closing the loop in-agent. Brands that wait for the channel to become "obvious" will be optimizing in 2027 against competitors who started collecting data, integrations, and customer relationships in 2026. HBR's view is consistent with what the data shows: the brands that win agentic commerce treat their product information as a machine-readable knowledge base, not a marketing asset (Harvard Business Review).
Search your brand
Search your brand to see how AI describes it and which competitors appear alongside it before you make changes.