Person holding a printed chart report above a laptop on a desk
Fewer clicks means each one has to be judged on what it becomes. Photo by Jakub Żerdzicki on Unsplash

Google is taking ads in AI Mode, its conversational search experience, out of the US. In an October 2026 announcement, Google said it is scaling AI Mode ads to ten more countries, including Australia, Canada and Singapore, and plans to add the United Kingdom, Brazil and Mexico in the fourth quarter of 2026. In the US it is also launching multiturn ads, which can appear on follow-up questions inside an AI Mode conversation.

The detail that matters most for B2B advertisers is in Google Ads Help: you cannot opt out. Existing Search, Shopping and Performance Max campaigns are automatically eligible, and you cannot target AI Mode ads, or switch them off, separately from Search. If you run Google Ads into the US, AI Mode ads are already in your account. If you run them into the UK, they are coming. This is a checklist for getting ready.

AI Mode ads and B2B search: the key numbers
KEY NUMBERS3xlonger queriesin AI Mode45%of B2B buyers hadused GenAI+12%Google Search adrevenue, Q3 2026
Source: Google, October 2026; Gartner, May 2026 (645 B2B buyers); Alphabet Q3 2026 earnings release

What actually changes with AI Mode ads

The queries are different. Google says AI Mode queries are, on average, three times longer than traditional searches (Google, 2026). A long query carries context a short one doesn’t: the team size, the system it must integrate with, the problem behind the search. That helps an advertiser whose keywords and landing pages match specific problems, and exposes one that relies on loose targeting to fill gaps.

The matching rules are different too. According to Google Ads Help, exact and phrase match keywords are eligible only when a query expresses direct commercial intent. AI Max, broad match and Performance Max are eligible for both direct and “contextual” matching, which covers broader, exploratory questions. In practice, your AI-powered targeting settings decide most of your exposure in research-stage conversations.

Reporting gives you less to go on. Performance in AI Overviews and AI Mode is included in aggregate in standard campaign reports, and the triggering queries appear in aggregate in the search terms report. There is no separate AI Mode line. One more caveat for regulated sectors: Google’s AI Overviews help page says ads do not appear within AI Overviews for sensitive verticals, naming finance and healthcare among them. It does not explain how a B2B supplier to those sectors is classified, so check your own data rather than assuming.

Fewer clicks, and each one matters more

The best independent evidence on AI answers and clicking comes from organic search, not ads, and it is older than we would like. Pew Research Center (2025) analysed March 2025 browsing data from 900 US adults. On visits where Google showed an AI summary, users clicked a traditional result 8% of the time, against 15% where there was none. They clicked a link inside the summary on just 1% of visits. Longer queries were more likely to produce a summary: 53% of searches of ten words or more did.

Share of Google search visits where users clicked a link (US, March 2025)
15%No AI summary:clicked a result8%AI summary:clicked a result1%AI summary:clicked its link
Source: Pew Research Center, July 2025 (browsing data from 900 US adults, March 2025). Covers AI Overviews and organic results, not ads.

B2B buyers are already doing research this way. In a Gartner survey of 645 B2B buyers (published 2026, fieldwork August to September 2025), 45% said they had used generative AI, mainly to gather information on vendors and products. We covered what that means for sales in our piece on AI agents in B2B buying.

Google has a reason to keep pushing ads into these answers. Search and other advertising revenue grew 12% to $63.35 billion in the third quarter of 2026 (Alphabet, 2026), and on the earnings call Philipp Schindler, Google’s chief business officer, said strong growth in AI Mode queries “had an impact because of the lower monetization we were seeing in AI Mode.” Our reading is that you should expect more ad placements in AI Mode over time, not fewer.

Our position: a click that comes out of a long AI Mode conversation should be judged on what it becomes, not on click-through rate or cost per click. The buyer may have done most of their research already. Or they may be a student, a job seeker or a competitor. Only your pipeline data can tell you which.

A checklist for AI Mode ads in a B2B account

  1. Read your long queries now. Pull 90 days of search terms and filter for queries of eight words or more. That is the closest view you will get of the traffic AI Mode will bring. Add negatives for the patterns that never convert, such as jobs, salary, free templates, courses and definitions.
  2. Make a deliberate call on broad match and AI Max. They carry the contextual eligibility, so they will drive most of your AI Mode exposure. If you keep them, back them with tight negative lists and brand exclusions, which Google says still apply. If you stay on exact and phrase match only, accept that you will be largely absent from exploratory conversations. Neither is right by default.
  3. Bid on outcomes, not form fills. Enhanced conversions for leads lets you send hashed first-party data, such as email addresses, and match it to leads that later convert in your CRM, so bidding learns from qualified leads rather than every submission. Uploading hashed customer data is still processing personal data, so check your privacy notice and lawful basis first (general information, not legal advice).
  4. Fix the CRM data first. Offline conversion signals are only as good as the records behind them. Consistent email capture, accurate lead status and a defined date for when a lead is accepted by sales all matter more than the bidding setting. Clean contact and CRM data is the unglamorous part that makes this work.
  5. Write landing pages for the long question. If the query names a use case, an integration or a company size, the page should answer it directly. A generic product page that made sense for a three-word query wastes a click from a twelve-word one.
  6. Agree what a qualified paid search lead is. Before volumes shift, write down the criteria sales will accept, the same way you would for any lead supplier. Our guide to BANT, MQL and SQL sets out the distinctions, and independent lead qualification can check paid search leads against them before they reach your account executives.

“Wait until there is data”

The strongest objection is that nobody has published B2B performance data for AI Mode ads. Google’s only performance claim in its announcement is that click-through rates are already similar between sponsored and organic product list items in AI Mode. That tells a software or services advertiser almost nothing, and with no separate reporting, you cannot easily measure the effect yourself.

That is a fair point about measurement, and a weak argument for doing nothing. Every item on the checklist above improves a B2B search account whether or not AI Mode ads perform well: better negatives, bidding on qualified outcomes, cleaner CRM data and agreed lead criteria. The cost of doing them early is low. The cost of discovering in the spring that broad match has been buying research-stage clicks you cannot see is higher.

Your next step

This week, export 90 days of search terms, split them into queries under and over eight words, and compare how each group converts to an opportunity that sales accepted, not just to a form fill. That gives you a baseline before AI Mode ads reach more of your markets, and it will show quickly whether your bidding is learning from the right signal.

If you want a second view on how paid search leads are qualified once they reach your pipeline, talk to Asott.