Google Is Putting Ads in AI Mode Follow-Up Questions. What Should B2B Advertisers Do?

What did Google announce about ads in AI Mode follow-up questions?

Google said on October 28, 2026 that it is launching multiturn ads in the U.S., which lets ads appear in follow-up queries inside AI Mode, not only on the first question. The same post scaled ads in AI Mode to ten more countries. For B2B advertisers, the second and third question in a conversation are now paid inventory.

The announcement came from Shashi Thakur, who leads Search Ads and Ads on Google Experiences, on Google's Ads and Commerce blog. The new markets are Australia, Canada, India, Kenya, Malaysia, New Zealand, Nigeria, Pakistan, the Philippines, and Singapore. Google also said it plans to add more locales in Q4 2026, including Brazil, Mexico, and the United Kingdom. Ads will carry a "Sponsored" label.

I work on go-to-market systems for B2B SaaS teams, and it is easy to read this as a consumer shopping story. I think that is a mistake. The follow-up question is where a B2B buyer stops browsing and starts comparing, and that is exactly the moment you want to be present.

Why do follow-up queries matter more than the first question?

The first question in AI Mode is usually broad, like "how do I automate lead routing." The follow-up is where intent sharpens: "which tools work with HubSpot for a ten-person team." That second question carries budget, stack, and team size. An ad that can appear there reaches a buyer who has already told Google what they need.

Google's own numbers point the same way. In the October post, Google said that queries in AI Mode are, on average, three times longer than traditional searches, citing its AI Mode Report for the U.S. from May 2026. Longer queries carry more context, and follow-ups carry even more because they build on the answer the person just read.

In classic search, a buyer who refines a query starts a new search and sees a fresh page of ads. In AI Mode the refinement happens inside one conversation. Until now, paid placements could miss that refinement entirely. Multiturn ads close that gap, which is why I expect them to matter more for considered B2B purchases than for quick consumer buys.

Which campaigns are already eligible for ads in AI Mode?

According to Google Ads Help, existing Search, Shopping, and Performance Max campaigns are automatically eligible to serve in AI Overviews and AI Mode. You do not build a new campaign or set a separate budget. That means many B2B accounts are already in this inventory without anyone having made a decision about it.

This is the part I would check first. If your account runs a Performance Max campaign that someone set up for a trial signup goal and then forgot, it may already be eligible to show inside AI answers. The question is no longer whether you will be there. It is whether the campaign that shows up is the one you would choose.

Google Ads Help also lists the markets where ads appear within AI Overviews and AI Mode: the United States, Australia, Canada, India, Indonesia for AI Overviews only, Kenya, Malaysia, New Zealand, Nigeria, Pakistan for text ads only, the Philippines, and Singapore, all in English. Google notes the list may not be exhaustive, so check the help page before you plan budgets by country.

How does Google decide when to show an ad inside an AI answer?

Google Ads Help describes two kinds of matching. Direct matching ties an ad to the query itself. Contextual matching ties it to both the query and the AI response. Exact and phrase match keywords only qualify when the query shows direct commercial intent, while broad match, AI Max, Shopping, and Performance Max can use both methods.

That split matters for how most B2B accounts are built. Many SaaS advertisers lean on exact and phrase match because they want tight control over cost per lead. Under the rules Google describes, those keywords will not be matched to broad, exploratory questions. They wait for a query that already sounds like a purchase.

Contextual matching is the new behavior. It means an ad can be chosen partly because of what the AI said in its answer, not only what the buyer typed. If the answer mentions CRM integration, an ad about CRM integration becomes relevant even when the person never typed that phrase. That is powerful, and it is also harder to predict.

Can you see how your ads perform in AI Mode on their own?

Not cleanly, based on what Google publishes. Google Ads Help says performance metrics for ads in AI Overviews and AI Mode are included in aggregate with your standard campaign reports. It also says you cannot target ads exclusively to AI Overviews or AI Mode. You can still use negative keywords and brand exclusions to shape where ads serve.

For a revenue team, this is the uncomfortable part. You may see cost per lead move and have no clean way to say whether AI Mode caused it. I would not wait for a perfect report. I would build a proxy instead: watch the search terms report for longer, question-shaped queries, and tag the leads they produce in your CRM.

In HubSpot, that can be as simple as a property that records whether the first-touch search term was a long natural-language question. It is not exact attribution, and I would not pretend it is. But it gives you a trend line you control, which is better than a blended number you cannot explain to a founder.

Should B2B teams lean into broad match and AI Max now?

Not by default. Broad match and AI Max are eligible for contextual matching, so they reach more AI Mode moments. But reach is not the goal for most B2B teams. The goal is qualified pipeline. I would test broad match in one tightly scoped campaign, with strong negatives, before changing a working account.

My rule is simple. If your average deal is large and your ideal customer profile is narrow, a single bad lead costs real sales time. In that situation, an extra click from a loosely related AI answer is not free, even if the cost per click looks fine. Protect the sales team's calendar first.

If your product is lower priced and self-serve, the math is different. More contextual reach may be worth it, because a trial signup costs the team very little. The decision depends on your motion, not on what Google is promoting this quarter. That is the part of go-to-market planning that no ad platform can make for you.

What should your landing pages say to a buyer who arrives mid-conversation?

A buyer who clicks an ad on a follow-up question has already read an AI answer. They know the basics. Your landing page should skip the category explainer and answer the specific follow-up: integrations, team size, pricing model, setup time. If the page starts from zero, it wastes the context the buyer just built.

This is where most B2B landing pages fall short. They are written for a cold visitor who needs to learn what the category is. A follow-up click is warmer than that. I would build at least one page per core use case that leads with the comparison or the constraint the buyer is likely asking about.

On Webflow, a clean way to do this is a CMS collection of use-case pages that share one template, so each page can speak to a narrow situation without a full redesign. The structure matters more than the platform. Clear headings, a direct first answer, and a visible next step will serve this buyer on any site.

How does this change the balance between paid and organic AI visibility?

It makes organic visibility more valuable, not less. When an ad sits next to an AI answer, the answer itself still shapes what the buyer believes. If your content is the source the answer draws on, the ad becomes a second touch, not the only one. Paid placement without organic presence is renting attention.

Google offered one data point here. In the October post, it said click-through rates are already similar between sponsored and organic product list items within AI Mode, based on Google internal data from the U.S. between August 20 and September 3, 2026. That figure is about product listings, not B2B text ads, so I would not stretch it further than that.

What it does tell me is that buyers in AI Mode are not ignoring sponsored results. That raises the stakes for the organic side too. I have written before about writing content for the follow-up question in AI search, and that work now pairs naturally with paid. The same follow-up questions should shape both your content plan and your keyword list.

How should a small B2B team budget for this without overspending?

Start with what you already pay for. Since existing campaigns are eligible automatically, the first step is not new spend but an audit of where current spend may already land. Review search terms weekly for a month, add negatives for off-target questions, and only then decide whether a dedicated test budget makes sense.

For an early-stage team, I would also ask whether paid search is the right channel at all before AI Mode enters the picture. I covered a similar question in my piece on whether B2B SaaS should run LinkedIn ads before product market fit, and the logic carries over. If you cannot yet say who buys and why, more ad inventory will not fix that.

For a team that already has a working search program, the best use of a small test budget is learning. Pick one use case, one landing page, and one campaign. Track lead quality in the CRM, not just cost per lead. After a month, you will know more than any industry commentary can tell you.

What should you do next?

Open your Google Ads account this week and list every Search, Shopping, and Performance Max campaign that is live, because each one is likely eligible for AI Mode already. Check its goal, its keywords, and its landing page. Then pick the single follow-up question your best buyers ask and build one page for it.

After that, set up a simple way to track the leads that come from long, question-shaped search terms, and pair it with a prompt list for your organic visibility. My notes on how many prompts you need to track AI visibility are a good place to start. Paid and organic should be measured against the same buyer questions.

If you want help auditing your campaigns, tightening your CRM tracking, or building landing pages for the questions your buyers actually ask, reach out. I build websites and the automations behind them for lead generation, and this is exactly the kind of system I like to set up. Let's chat.