Google Is Putting Ads Inside AI Search. Here's What Marketers Need to Do Now.
Google announced on February 11 that it's testing shopping ads inside AI Mode, the conversational search experience now used by more than 75 million people daily. If you run Shopping campaigns or manage a product catalog, this is the single biggest change to Google's ad platform since Performance Max launched in 2021.
The ads don't sit in a sidebar or above organic results. They appear as sponsored product tiles within the AI Mode response, displayed alongside or below the AI-generated text. When someone asks AI Mode "what's the best running shoe for flat feet," the answer now includes sponsored product recommendations labeled "Sponsored," positioned right next to the AI's own picks.
This is Google's answer to the same question OpenAI tackled the same week with ChatGPT ads: how do you monetize an AI conversation?
What AI Mode Shopping Ads Actually Look Like
In traditional Google Shopping, your product appears in a carousel of tiles above the search results. In AI Mode, there is no carousel. The AI assembles a written response, compares products, and recommends specific items. Your sponsored product tile appears within that same interface, adjacent to the AI's recommendations.
Think of it like a product placement in a personal shopping assistant's recommendation. The AI might say: "For flat feet, stability shoes with firm midsoles work best. Here are three options," and alongside those options, a sponsored product tile surfaces a paid placement. It's labeled, but it's visually integrated into the same experience.
Google commissioned a study with 4,773 participants (October to December 2025) and found that users reported a "more helpful shopping experience" in AI Mode when they could compare multiple brands and stores in a single response. That's the justification for placing ads within the AI Mode experience, not in the margins.
Who's Already In
Google isn't waiting for mass adoption. Etsy and Wayfair were the first retailers to go live with shoppable checkout inside AI Mode, launching Universal Cart Purchase at NRF on January 11, 2026. Users can browse, compare, and complete purchases without leaving the AI conversation. Target, Walmart, and Shopify were announced as "coming soon" partners at the same event.
On the advertiser side, Google launched a pilot called "Direct Offers," where retailers surface exclusive discounts inside AI Mode when the system detects high purchase intent. Petco, e.l.f. Cosmetics, Samsonite, Rugs USA, and select Shopify merchants are the first participants. Retailers configure offers in campaign settings; Google's AI decides when to show them based on shopper behavior and context.
Powering all of this is the Universal Commerce Protocol (UCP), an open standard Google introduced alongside those partnerships on January 11. UCP lets AI agents discover products, negotiate checkout, and complete transactions across retail platforms.
What Changes in Practice
Here's where it gets concrete for anyone managing campaigns.
Keyword strategy shifts from terms to intent. In AI Mode, users write queries two to three times longer than in traditional search. "Best laptop" becomes "best laptop for video editing under $1,200 that works with DaVinci Resolve." Google's AI infers intent from the full conversation, not individual keywords. Exact Match and Phrase Match are too narrow. Broad Match gives Google the flexibility to interpret conversational queries and match them to your products.
Product feeds become your competitive edge. When AI Mode assembles a recommendation, it pulls from structured product data: titles, descriptions, attributes, pricing, availability. If your Merchant Center feed has thin descriptions or stale pricing, the AI skips you for a competitor with cleaner data. AI Mode amplifies the penalty for poor feed hygiene.
Bidding depends on Smart Bidding. Manual CPC bids don't translate when an AI is assembling conversational responses. Strategies like Maximize Conversion Value are essential because Google's models evaluate intent signals from the full conversation. Feed your campaigns first-party data through Customer Match lists so the system learns which segments convert.
Fewer slots, higher stakes. A traditional Shopping carousel shows 8 to 12 products. An AI Mode response recommends 2 or 3. Early signals suggest selection is based on ad relevance, feed quality, bid strength, and the AI's assessment of fit for the user's stated needs. Organic relevance matters: the AI won't recommend a camping tent to someone asking about running shoes just because the tent bid higher.
The Harder Question: How Does the AI Choose?
Google hasn't published the exact ranking signals for AI Mode shopping ads. But we can piece together the logic.
The AI evaluates the full conversation, not a single query. It weighs the user's stated preferences, budget constraints, brand affinities, and how well your product data matches those signals. The ad auction still exists, but it's filtered through an intent layer that the AI controls.
A product with a lower bid but a precisely optimized feed (detailed attributes, accurate pricing, strong reviews) can beat a higher bidder with a generic listing. Google VP Vidhya Srinivasan positioned this in her annual ads letter as "agentic commerce," where AI agents research, compare, and complete purchases on behalf of users. The agent cares about relevance, not just who paid the most.
It's still an ad platform. Money talks. But in AI Mode, money talks through the filter of product quality signals. If your data is bad, no bid will save you.
Your 30-Day Action List
Week 1: Audit your product feeds. - Review every product title and description in Merchant Center. Add specific attributes (material, size, compatibility, use case) that AI Mode can parse. - Fix stale pricing and out-of-stock items. AI Mode penalizes bad data harder than traditional Shopping because it's making specific recommendations, not showing a grid. - Confirm your product categories and custom labels are accurate. The AI uses these to match intent.
Week 2: Restructure your bidding. - If you're still on Manual CPC for Shopping campaigns, switch to Smart Bidding (Maximize Conversion Value or Target ROAS). - Upload or refresh Customer Match lists. First-party data trains the AI on your highest-value segments. - Expand Broad Match coverage in Search campaigns that feed into AI Mode impressions.
Week 3: Monitor AI Mode performance. - Check your Google Ads reports for AI Mode impression data (Google is rolling out segmentation for this). - Compare click-through and conversion rates from AI Mode vs. traditional Shopping. Early adopters report fewer clicks but higher intent per click. - If you're eligible, apply for the Direct Offers pilot to test personalized discounts in AI Mode.
Week 4: Plan for vertical expansion. - Google has confirmed AI Mode ads are expanding beyond retail into travel and other verticals. If you're in travel or services, start preparing your feeds now. - If you sell through Shopify, check whether your store is UCP-enabled. - Brief your team on the shift from "user searches, clicks, buys" to "AI researches, recommends, checks out." Build it into your 2026 planning.
The Bottom Line
Google is rebuilding its ad business around conversational AI. Your product feed is now more important than your keyword list. Your Smart Bidding strategy matters more than your manual bids. And the quality of your product data determines whether an AI agent recommends you or your competitor.
This isn't a future scenario. It's live in the US with 75 million daily users. Marketers who optimize for AI Mode now will own the early advantage. Those who wait for Google to publish best practices will be optimizing against competitors who already figured it out.
