Key takeaways
- Ads in AI Mode embed labelled ads inside AI-generated answers, so Google’s models now frame why you are the right choice, not your headline.
- Expect click volume on informational queries to fall and cost-per-click on commercial queries to rise. Judge the trade on cost-per-acquisition, not clicks.
- Merchant Center feed quality and conversion values become the throttle: if AI cannot read your product or price, it cannot pick you as the answer.
- At least 56% of active UK accounts have a conversion-tracking fault distorting their numbers, and that is a floor, not a ceiling.
- Fix tracking, feeds and negatives before enabling AI Brief or AI Max, or the model will confidently scale the wrong outcome.
Ads in AI Mode are Google’s new paid formats that place labelled ads directly inside AI-generated Search answers, where Gemini reasons about the query and frames why your product or service is the right response. Announced at Google Marketing Live 2026 and built on AI Max and Performance Max, they move the ad from above ten blue links to inside the answer itself. For UK PPC teams, that is not a cosmetic change. It shifts where the click happens, what the user has already decided by the time they reach you, and how much of the buying decision Google’s models make before your landing page loads. If you already worry about lead quality on matching search intent to conversions, this raises the stakes.
The commercial risk is simple. When ads become answers, the systems deciding which answer wins lean even harder on your feed, your conversion data and the context you give them. Thin or broken inputs will show up faster, and cost more, than they do today.
This piece is about what Ads in AI Mode actually change, why the money moves, and what to fix now rather than when the formats hit your account.
What Ads in AI Mode actually change
Google announced a new generation of formats for the AI era of Search. Ads in AI Mode, powered by Gemini and built on AI Max and Performance Max, embed labelled ads directly inside AI-generated responses. The formats reason about the query, adapt their creative, and surface verified product details at the point of decision.
Alongside it, Google introduced AI Brief: a way to steer AI Max for Search campaigns in plain language. You write messaging, matching and audience guidelines, for example “never mention prices” or “avoid searches for in-person degrees”, and the model works within those boundaries. Shopping formats got AI-Powered Shopping Ads and AI Max for Shopping, both pulling harder on Merchant Center feeds to answer conversational queries. That is the substance. The rest of the launch, from journey-aware bidding to demand-led pacing, orbits the same shift towards models making more of the call.
Why Ads in AI Mode matter for PPC performance
Here is the mechanism. In classic Search, a user types a query, scans results, and clicks your ad because your headline matched their intent. You control the headline. In AI Mode, Gemini reads the query, drafts an answer, and decides which advertised product or service belongs inside that answer. Your ad still shows, but the model, not your carefully written headline, is now framing why you are the right choice. Control moves from the advertiser to the system.
That has three direct consequences for performance.
Fewer clicks, higher intent, tighter measurement
When ads answer the question inside the response, the user who does click has already been half-sold. Click volume on informational queries drops because the AI satisfied the query without a click. The clicks you do get are later in the journey and worth more. Your cost-per-click will rise, your conversion rate should rise too, and if your tracking is weak you will only see the first half of that trade. That is where accounts get misread as “getting more expensive” when they are actually getting more efficient.
Feed and conversion quality become the throttle
AI-Powered Shopping Ads and AI Max for Shopping build answers from your Merchant Center data. A product missing GTINs, weak titles or stale prices cannot be reasoned about, so it will not be chosen as the best answer. The same logic applies to conversions: journey-aware bidding and Smart Bidding need clean signals to price these auctions. Feed hygiene stops being a Shopping housekeeping task and becomes the difference between appearing in an AI answer or vanishing from it. Our guide to tightening product feed management covers the groundwork this now depends on.
Lead-gen accounts feel the tracking gap first
The accounts we worry about most here are lead-gen accounts running broad match with Smart Bidding and a thin conversion history. AI Mode widens the query pool your ads can answer. If your conversion feedback is patchy, the model optimises towards the wrong answers, and you get more enquiries that look fine in the dashboard and convert badly in the CRM. Our Q1 2026 (Jan to Mar) audit data shows at least 56% of active UK accounts carry a conversion-tracking fault serious enough to distort the numbers they optimise on. That is a floor, not a ceiling: the Google Ads API cannot see Consent Mode, web Enhanced Conversions or tag-firing errors, so the true rate is likely higher. Feed AI Mode broken signals and it will confidently scale the wrong outcome. Our guide to offline conversion tracking shows how to close that CRM gap.
PPC Geeks’ View
The specific problem advertisers will face is control loss dressed up as convenience. AI Brief hands you plain-language guidelines, which feels like more control. It is actually less. You are describing intent to a model that decides execution, and the gap between “focus on my autumn collection” and what the model serves is where budget quietly leaks. We see this most often in accounts that treat automation as set-and-forget, then cannot explain a month-on-month cost shift because nobody owns the inputs.
Ads in AI Mode reward the accounts with clean feeds, accurate conversion values and disciplined negatives, and punish the ones running on autopilot. The single act that protects you is auditing what the model is actually being fed before you let it answer on your behalf.
Automation does not fail loudly. It fails quietly, optimising perfectly towards the wrong number while your dashboard stays green.Dan Trotter, Head of PPC, PPC Geeks
This is exactly the type of issue we surface in a free Google Ads audit, especially where automation, tracking or feed quality may be steering performance without anyone noticing.
What advertisers should do next
Do not wait for Ads in AI Mode to appear in your account. The work that protects you is the same work that improves performance today.
- Fix conversion tracking before you touch AI settings. Open Google Ads > Tools > Conversions and check every action fires, deduplicates and reports a real value. Cross-reference against your CRM for lead-gen; anything with an import lag over three days needs offline conversion uploads.
- Audit your Merchant Center feed this week. Segment by disapprovals, missing GTINs and stale prices. If AI-Powered Shopping Ads cannot read a product, it cannot pick it as the answer.
- Write AI Brief guidelines as if briefing a new hire. Draft explicit messaging, matching and audience rules before AI Max touches your budget. Include what the model must never say and which queries to avoid.
- Rebuild your negative keyword lists for a wider query pool. AI Mode expands what your ads can answer, so tighten the boundaries manually rather than trusting broad match to police itself.
- Feed conversion values, not just conversion counts. Journey-aware bidding needs value signals to price late-stage clicks correctly. Counting a £50 enquiry the same as a £5,000 one will misdirect spend.
- Separate high-intent from research queries in reporting. Expect click volume on informational terms to fall and cost-per-click on commercial terms to rise. Judge the trade on cost-per-acquisition, not clicks.
If you want a partner to run this properly, our Google Ads agency team does exactly this groundwork before letting automation loose on a budget. It is the same discipline we apply to Smart Bidding Exploration controls, where more model freedom demands cleaner inputs first.
What this means for your campaigns
Ads in AI Mode are still testing, so you have a window. Use it to get your house in order rather than to wait and see. The advertisers who win when these formats scale will be the ones whose feeds are clean, whose conversion values are accurate, and whose AI Brief guidelines are written with intent. The ones who lose will hand a powerful model poor inputs and wonder why costs drifted up while lead quality drifted down.
Google confirmed the direction of travel at Google Marketing Live 2026, and the official ads and commerce announcements make the reliance on first-party signals and feed quality explicit. For the mechanics of how AI reshapes bidding freedom, the Smart Bidding documentation is worth reading against your own value setup, and Google’s product data specification is the checklist your feed must pass before any AI format can use it.
The common thread is inputs. Ads in AI Mode make your data quality the deciding factor in whether you appear as the answer or disappear from it. We can help you stress-test your account against this. A free PPC audit is the fastest way to see where you stand. For the detail behind this, see Google’s own Ads in AI Mode announcement.
Frequently asked questions
What are Ads in AI Mode?
Ads in AI Mode are a new format Google announced at Google Marketing Live 2026. Powered by Gemini and built on AI Max and Performance Max, they place clearly labelled ads directly inside AI-generated Search answers, adapting creative and explaining why your product or service best answers the query.
How do Ads in AI Mode affect my cost-per-click?
When the AI answers informational queries without a click, click volume on those terms falls. The clicks you keep are later in the journey and more valuable, so cost-per-click on commercial queries tends to rise while conversion rate improves. Measure success on cost-per-acquisition, not raw clicks.
What is AI Brief in Google Ads?
AI Brief lets you steer AI Max for Search campaigns in plain language. You set messaging, matching and audience guidelines, such as never mentioning prices or avoiding certain searches, and the model works within those rules. It feels like more control but shifts execution to the model, so your written boundaries matter.
Do I need to fix my Merchant Center feed for AI Shopping Ads?
Yes. AI-Powered Shopping Ads and AI Max for Shopping build answers from your Merchant Center data. Missing GTINs, weak titles or stale prices mean the model cannot reason about your product, so it will not choose it as the answer. Feed hygiene now decides visibility.
Why does conversion tracking matter more with Ads in AI Mode?
AI Mode widens the pool of queries your ads can answer, and Smart Bidding uses your conversion signals to price those auctions. With broken tracking, the model optimises towards the wrong outcomes. Our Q1 2026 audit data found at least 56% of UK accounts carry a tracking fault serious enough to distort optimisation, and that is a floor.
