Google AI Commerce: everything you need to know
Best for: brands and retailers who want their products visible, and eventually purchasable, inside Google’s AI surfaces such as AI Mode and the Gemini app, by enriching the product data those models reason with.
|
|
|
|
| Fast | Longer |
|
|
|
|
|
|
| Discovery | Purchase |
Google rebuilt shopping around AI in 2026, in two moves six months apart. In January, at NRF’s Big Show, it launched the Universal Commerce Protocol (UCP), an open standard that lets AI Mode and the Gemini app complete a purchase without sending the shopper to a retailer’s website. In May, at Google Marketing Live, it switched on Universal Cart in beta with retailers including Nike, Sephora, Target, Walmart and Wayfair. Google AI Commerce is the umbrella over three layers: discovery in AI Mode and Gemini, native checkout through UCP, and measurement through AI Performance Insights.
This is not an advertising channel in the classic sense: there is no auction and no bid. Visibility is earned through product data. Gemini and AI Mode don’t retrieve a keyword match, they reason about a product, and reasoning needs context a traditional feed doesn’t carry: what pairs with it, what buyers ask about it, which variant differs how. Google’s six new Conversational Attributes exist for exactly that, and while Google documents no direct ranking benefit today, its AI Performance Insights report will turn attribute completeness into a share-of-voice number competitors can beat you on.
Through Lengow, the discoverability side is live today: all six conversational attributes are native to Lengow’s Google feed structures, for both API and URL feeds, a configuration change rather than a project, with nothing to install and no Google approval to wait for. Google Native Checkout is different: it sits in an early access programme, gated behind Google’s UCP Integration Interest Form and onboarding, and its consumer rollout is US-first, with no published timeline for continental Europe.
Key figures
Sources: Google I/O 2026 (AI Mode users and query growth); Google, Google Marketing Live 2026 (Shopping Graph); Adobe holiday shopping data (January 2026). Discovery traffic is growing fast; Google has not yet published transaction figures for AI-driven checkout.
Three layers, three speeds
Who will see your products?
The audience is the billion-plus people using AI Mode, plus Gemini app users, who shop by describing what they need: a lightweight carry-on under 2 kg, a stroller safe for a newborn, a gift for someone who already owns everything. Price, location, colour, brand and availability are the five things Google says these shoppers specify most, and planning-style queries grew 80% faster than AI Mode overall in the six months to May 2026.
Behaviour is shifting ahead of transactions: roughly one in eight US consumers now starts a shopping journey on a generative AI chatbot according to Bain, and Adobe measured a 693% jump in retail traffic from generative AI over the 2025 holidays. The caveat belongs next to the promise: this is discovery traffic, and it still converts markedly worse than classic search. The race today is to be present in the answers, not to count revenue from them.
Because the answer is generated rather than ranked, absence is silent: a product whose data can’t answer the shopper’s question simply never appears, whatever its price or quality.
Spec-rich catalogues
Electronics, appliances, DIY and equipment, where manuals, compatibility data and FAQs answer exactly the questions shoppers ask AI Mode.
Variant-heavy catalogues
Fashion, beauty and home, where item group titles and variant options stop twelve colourways being read as twelve unrelated products.
US-facing merchants
Native Checkout’s consumer rollout is US-first, then Canada and Australia, then the UK. Merchants selling into the US have a checkout race to enter; others have time.
A weaker fit: ROI-now advertisers
There is no documented ranking benefit yet, no ad product to buy, and no published transaction figures. Merchants needing measurable returns this quarter should treat this as preparation, not performance.
What you need to know before starting
Split the work the way Google has: a discovery half that is open to everyone today, and a checkout half that is gated, US-first and worth preparing for without betting a roadmap on.
Merchant profile
Any merchant with a Google Merchant Center feed can map the discoverability attributes today. Native Checkout requires applying through Google’s UCP Integration Interest Form and passing its onboarding.
Cost & economics
No ad auction and no media budget: visibility is earned through data. Under UCP you remain Merchant of Record, Google Pay handles payment, and fulfilment and the customer relationship stay with you.
Availability & rollout
The attributes are available wherever Merchant Center operates. Universal Cart is in US beta, with Canada and Australia next and the UK after; Google has published no timeline for continental Europe.
Technical setup
Discoverability attributes map inside your existing Google API or URL feed. Native Checkout, once approved, uses a dedicated Google UCP channel in Lengow, delivered to Merchant Center as a supplemental data source.
Feed requirements
Six conversational attributes for discovery: question_and_answer, related_product, document_link, item_group_title, variant_option, popularity_rank. UCP adds a checkout identifier, checkout eligibility and regulatory notice fields such as Prop 65.
Measurement
AI Performance Insights will benchmark share of voice on AI surfaces and flag missing attributes. Until it reaches you, test your best sellers in AI Mode using real shopper language and see whether you appear.
How to get ready for Google AI Commerce with Lengow
The discovery work starts in your existing feed: map the six conversational attributes in your Google API or URL configuration in NetAmplify, feeding them from the FAQ, documentation and sales data your organisation already holds. The checkout work starts with Google: prepare Merchant Center to spec, apply through the UCP form, and once approved, configure the dedicated Google UCP feed in Lengow.
The channel rewards merchants who treat product content as structured data rather than page copy. Check out Lengow’s Google AI Commerce setup guide for the step-by-step configuration.
| Channel challenge | What successful merchants do | What Lengow automates |
|---|---|---|
| From keywords to context | Populate the six conversational attributes so Gemini has material to reason with, not just match against. | All six attributes are native to Lengow’s Google feed structures, for API and URL feeds. |
| Product Q&A | Turn the questions support teams and product pages already answer into structured pairs. | question_and_answer mapping brings FAQ content into the feed via rules. |
| Product relationships | Declare complementary, compatible and substitute items so agents can build recommendations. | related_product IDs are mapped from catalogue data through attribute rules. |
| Documentation depth | Link manuals, spec sheets and warranties directly, the detail a “will this work for me” question depends on. | document_link is carried per product in the feed structure. |
| Variant clarity | Give variants a shared parent title and precise differentiating parameters. | item_group_title and variant_option are structured through variation handling. |
| Popularity signal | Rank items by sales or traffic within the catalogue, a signal beyond price and availability. | popularity_rank is fed from source data through rules. |
| Native Checkout eligibility | Apply through the UCP form, then declare exactly which products can be bought inside Google. | The dedicated Google UCP channel carries checkout identifiers and eligibility flags. |
| Regulatory notices | Attach consumer-facing notices such as Prop 65 where products require them. | notice_type and notice_message are part of the UCP feed structure. |
| Deploying without disruption | Enrich Google’s view of the catalogue without touching the primary feed that runs live campaigns. | The compiled UCP feed URL plugs into Merchant Center as a supplemental data source. |
With Lengow AI Studio, AI is embedded directly into the feed workflow: AI Category Mapping classifies products into Google’s taxonomy faster, AI Content Localisation adapts product content per market, and AI Image Compliance prepares visuals for Merchant Center requirements at scale, the same catalogue quality that AI surfaces now read.
Should Google AI Commerce be on your roadmap?
Google has tried native commerce before, from Checkout to Buy on Google, and retired every attempt. This one has stronger incentives, a billion-user surface and an industry coalition behind it, and it still hasn’t published a single transaction figure. Lengow’s analysis of the rollout draws a clear line for merchants.
The discovery race has started and runs on the product data you already send to Merchant Center. The checkout race, for a European merchant, hasn’t begun. Prepare the data now; don’t bet the roadmap yet.
Lengow channel analysis · Google AI Shopping: will this attempt finally work?
FAQ
Frequently asked questions
Can’t find what you’re looking for?
Google AI Commerce is the umbrella over Google's AI-powered shopping stack: product discovery in AI Mode and the Gemini app, native checkout through the Universal Commerce Protocol (UCP), and measurement through AI Performance Insights. It is not a classic ad channel: there is no auction and no bid, and visibility is earned through product data.
Any merchant with a Google Merchant Center feed can map the six discoverability attributes today, with nothing to install and no approval to wait for. Google Native Checkout is different: it runs as an early access programme, and you must apply through Google's UCP Integration Interest Form and complete its onboarding before configuring a feed.
The discoverability attributes work wherever Merchant Center operates. The consumer-facing checkout, Universal Cart, is in beta in the US with retailers including Nike, Sephora, Target, Walmart and Wayfair; Canada and Australia come next, then the UK. Google has published no timeline for continental Europe, so for European merchants the discovery race has started and the checkout race hasn't.
question_and_answer (structured FAQ pairs), related_product (complementary and substitute item IDs), document_link (manuals, spec sheets, warranties), item_group_title (the shared title across variants), variant_option (the parameter separating one variant from another) and popularity_rank (an item's rank in your own catalogue). They are optional, and Google documents no direct ranking benefit today; what they do is give Gemini context to reason about a product instead of keywords to match.
All six conversational attributes are native to Lengow's Google feed structures, for both API and URL feeds, so mapping them in NetAmplify is a configuration change rather than a project. Once Google approves your UCP application, Lengow provides the dedicated Google UCP channel with checkout identifiers, eligibility flags and regulatory notice fields, compiled into a URL you add as a supplemental data source in Merchant Center.
Take control of your advertising performance
Join thousands of brands and retailers using Lengow to structure, optimise and distribute their product feeds across every advertising channel, with data-driven insights and full control.