AI Overviews Ads Are Here. Here Is What Changes for Paid Search.

Google’s AI Overviews now carry ads. What started as an organic feature sitting above the traditional search results has become a commercial surface, and paid search teams need to understand what that means for their campaigns, their budgets, and their measurement models before the landscape shifts further under them.

The short version: ads appearing inside AI Overviews give Google a new way to monetise the top of the search page, and give advertisers a new placement that behaves differently from anything they have managed before. The mechanics are familiar enough. The implications are not.

What Has Actually Changed With AI Overviews Ads?

Google began rolling out AI Overviews to US users at scale in 2024, replacing or supplementing the traditional featured snippet with a synthesised, multi-source answer generated by its Gemini models. For most of that initial period, the AI Overview was a purely organic feature. Ads sat above or below it in the traditional positions.

The expansion that matters now is the integration of ads directly within the AI Overview unit itself. Google has confirmed that Shopping ads and, in some configurations, text-based ads can now appear embedded inside the AI-generated answer block. The user does not leave the AI Overview to see the ad. It is woven into the response.

This is not a minor formatting update. It changes the context in which your ad appears, the intent state of the user seeing it, and the signals available to measure whether it worked. I have been watching Google restructure the search results page for two decades, and this is one of the more structurally significant shifts I have seen, not because the ad unit itself is revolutionary, but because of where it sits in the user’s decision process.

If you want to understand how AI is reshaping search and advertising more broadly, the AI Marketing hub at The Marketing Juice covers the commercial implications across channels and tools.

Why Intent Is the Variable That Changes Everything

When I was running paid search at scale, managing spend across dozens of client accounts in sectors from travel to financial services, the discipline that separated good campaigns from expensive ones was intent mapping. Not keyword research. Intent mapping. Understanding where in the decision process a user was when they typed a particular query.

AI Overviews are triggered most often by informational and navigational queries. Someone asking “what is the best running shoe for flat feet” or “how does a balance transfer credit card work” is in research mode. They are gathering, not deciding. The AI Overview is built to satisfy that research impulse by synthesising multiple sources into a single answer.

Placing an ad inside that answer does not transform the user’s intent. They are still in research mode. The ad appears in a high-attention position, which has value, but the conversion behaviour you would expect from a bottom-of-funnel placement is not what you should be forecasting here. Teams that treat AI Overview ad impressions as equivalent to a standard commercial query placement will over-invest and be disappointed by the return.

This is not speculation. I saw a version of this play out early in my career at lastminute.com, where we were running paid search campaigns against queries that looked commercially promising but were predominantly informational. The click volumes were strong. The conversion rates were not. The lesson was that position and intent are separate variables, and conflating them costs money.

Which Campaign Types Are Most Exposed?

Google has indicated that Shopping ads are the primary format appearing inside AI Overviews in the current rollout. Performance Max campaigns, which give Google broad discretion over where ads are placed, are also likely to serve into this placement. Standard Search campaigns with tightly defined match types and placement controls have more insulation, at least for now.

This creates an asymmetry. Teams running highly automated, broad-targeting campaigns have the least visibility into where their ads are actually appearing and the least control over whether those placements are efficient. Teams running tighter, more manually managed campaigns have more protection, but also less reach into a placement that could, in the right context, perform well.

The practical implication is that if you are running Performance Max and you have not recently audited where your impressions are being allocated, now is the time. Semrush’s overview of AI optimisation tools is a reasonable starting point for understanding what visibility options exist across the ecosystem. The tools are imperfect, but they are better than flying blind.

For Shopping specifically, the question is whether your product feed, your imagery, and your pricing are optimised for a placement where the user is comparing options within an AI-generated context rather than a traditional results grid. The competitive frame is different. The user is seeing your product alongside an AI-generated narrative about the category, not just alongside other product listings.

What Does This Do to Your Attribution Model?

Attribution was already a problem before AI Overviews. I spent years in agency leadership watching clients make budget decisions based on last-click data that systematically undervalued brand and upper-funnel activity. We moved to data-driven attribution, which was better, but still a model built on assumptions about how credit should flow through a linear path to conversion.

AI Overviews add a new complication. If a user sees an ad inside an AI Overview, does not click, continues researching, and converts three days later through a branded search, how does your attribution model account for the AI Overview exposure? It probably does not. The impression happened inside a Google-generated surface, and the connection between that exposure and the eventual conversion is likely invisible to your measurement stack.

This is not a reason to panic. It is a reason to be honest about what your data is actually telling you. The measurement frameworks we use are approximations of reality, not reality itself. That has always been true. AI Overviews make it more true, and any team that treats their attribution report as a precise account of how their ads are performing will make worse decisions than a team that holds the data more lightly and triangulates across multiple signals.

Incrementality testing becomes more important in this environment. If you want to understand whether AI Overview placements are contributing to business outcomes, you need to design tests that can isolate the effect, not rely on attribution credit that the platform assigns by default. Ahrefs has covered the intersection of AI and search measurement in ways that are worth reviewing if you are trying to build a more rigorous framework.

How Should Paid Search Teams Respond Right Now?

The wrong response is to do nothing and assume your existing campaigns will adapt automatically. The equally wrong response is to restructure everything immediately based on limited data about a placement that is still evolving. The right response is measured and specific.

First, audit your current campaign structure to understand which campaigns are most likely to serve into AI Overviews. Performance Max and broad-match Shopping campaigns are the obvious candidates. Review the placement reports available to you and establish a baseline before the placement becomes more widespread.

Second, revisit your creative assets for Shopping campaigns. If your product imagery, pricing, and copy were optimised for a standard grid placement, they may not perform as well in an AI-generated context where the surrounding content is richer and more contextual. HubSpot’s guide to AI copywriting tools covers some of the options available for testing ad copy variations at scale, which becomes more relevant when you are optimising for a new and unfamiliar placement.

Third, set up measurement checkpoints. Agree internally on what metrics matter for AI Overview placements before you start spending meaningfully against them. Impression share, click-through rate, and conversion rate are the obvious starting points, but also track view-through behaviour and branded search volume as secondary indicators of awareness impact.

Fourth, resist the pressure to scale. Every new ad placement comes with a wave of vendor enthusiasm and a push to increase budgets before the data is there to justify it. I have seen this pattern play out with every major Google product launch for two decades. The teams that do well are the ones that test methodically, not the ones that commit early and spend their way through the learning period.

What This Means for Organic Search Teams

Paid search teams are not the only ones affected. The expansion of ads inside AI Overviews has a direct implication for organic search strategy, because it changes the economics of appearing in an AI Overview at all.

If the AI Overview now contains both organic citations and paid placements, the organic citation becomes relatively less prominent. A user reading an AI Overview sees the synthesised answer, potentially a paid placement, and then the source citations. The organic citation is still valuable for brand visibility and trust signals, but the direct traffic benefit may be lower than it would have been in a purely organic AI Overview.

This does not mean organic optimisation for AI Overviews is pointless. It means the goal of that optimisation shifts slightly. Being cited in an AI Overview is increasingly about authority and brand recognition rather than click volume. Moz’s analysis of content and AI tools is a useful reference for teams thinking about how to structure content to earn AI Overview citations more consistently.

The broader point is that paid and organic search teams need to be in the same conversation about AI Overviews. The placement affects both channels, and a coordinated strategy, where paid covers the commercial intent queries and organic targets the informational queries where AI Overviews are most prevalent, is more effective than two teams operating independently against the same page.

When I grew iProspect from 20 to 100 people and moved it from loss-making to a top-five UK agency, one of the structural changes we made was breaking down the wall between paid and organic teams. Not because it was organisationally tidy, but because the search results page does not care about your internal org chart. The user sees one page. Your strategy should reflect that.

The Longer-Term Shift in How Search Works

AI Overviews ads are a symptom of a larger structural change in search. Google is building a search experience where the answer is increasingly provided on the results page rather than accessed through a click. That is good for users in many cases. It is a significant challenge for advertisers and publishers who built their models on click volume.

The response to this cannot be purely tactical. It requires a strategic rethink of what search is for in your marketing mix. If organic search is delivering fewer clicks because AI Overviews are satisfying queries without requiring a click-through, the value of organic search shifts from traffic generation to brand presence and authority. If paid search ads inside AI Overviews are reaching users earlier in their decision process, the value of those placements shifts from direct response to influence.

Neither of these shifts makes search less valuable. They make it differently valuable, and the teams that adapt their measurement and their objectives accordingly will be better positioned than those who keep applying 2019 frameworks to a 2026 search landscape.

Semrush’s thinking on AI content strategy and Ahrefs’ webinar series on AI tools both cover the evolving relationship between content, search, and AI-generated surfaces in ways that are worth bookmarking as reference points while this space continues to develop.

For a broader view of how AI is reshaping marketing measurement, creative, and channel strategy, the AI Marketing hub at The Marketing Juice brings together the most commercially relevant thinking across all of these areas.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions