Discussion starter: an AI search question, posted with an answer from Niraj Raut to open the thread. If you have dealt with this on a site, reply with what you saw, especially where it differs.
- AI Search
Where ads appear inside AI Overviews or AI Mode, have you measured a further drop in organic clicks compared with similar queries without ads?
Disclosure: Niraj Raut, who posted this answer, runs the SEO consultancy linked at the end of it.
1 reply
-
adminAdmin
Public evidence on this specific effect is thin. No controlled public test exists that I can point to that separates the click cost of an ad inside an AI Overview from the click cost of the AI Overview itself. Google also gives you very little to do it with: Google Ads does not report these ads as a separate segment, and Search Console’s AI reporting shows impressions without clicks or queries. So any figure you see comes from SERP captures joined to click data, and it is only as good as the comparison behind it.
An extra loss is plausible on commercial queries, but the comparison in the question (“similar queries without ads”) is the weak point. Google says it considers both the query and the content of the AI Overview when serving these ads, so queries that attract them are likely to be more commercial than queries that don’t, and commercial queries already carry more ads and shopping units. Compare the two groups and you mostly measure intent. The comparison that isolates the ad is the same query, at a similar position, with and without the ad inside the Overview.
What Google documents about the placement, as of September 2026
Google’s help page About ads and AI Overviews sets out the parts that matter for measurement:
- Ads can show above, below or within AI Overviews. Above and below is available in every market where AI Overviews show (Google says more than 200).
- Ads within AI Overviews are listed as available in English, on mobile and desktop, in 12 countries: Australia, Canada, India, Indonesia, Kenya, Malaysia, New Zealand, Nigeria, Pakistan, the Philippines, Singapore and the US. The UK is not on the list.
- Text and Shopping ads from Search, Shopping and Performance Max campaigns are eligible, and advertisers cannot opt out of the placement.
- Google Ads does not currently offer segmented reporting for ads shown within AI Overviews. They are reported as top ads.
- Both the user’s query and the content of the AI Overview are considered when these ads are served.
For AI Mode, the picture is less clear. Google announced further AI Mode ad formats at Google Marketing Live 2026, and a June 2026 agency summary of those announcements described formats such as conversational discovery ads as US tests at that point. Check what actually serves in your market before you build a comparison around it.
Why “similar queries without ads” is the hard part
- Selection. Because the Overview’s content feeds ad selection, the ad is likely to appear where the answer itself is commercial: product types, comparisons, “best X for Y”. Those queries tend to have lower organic CTR for reasons that have nothing to do with the ad.
- Ads are not a fixed property of a query. Each impression runs its own auction, so the same query can show an ad inside the Overview on one search and not on the next. “Queries with ads” really means “queries with a high ad rate”, and one daily rank check samples one impression out of many.
- The Overview itself comes and goes. If AI Overview presence shifts during your comparison window, that shift can swamp any ad effect. Overview presence needs to be its own variable, not folded into “ads”.
Comparison What it actually isolates Main confounder Queries that show ads inside the Overview versus queries that don’t Intent and the ad, mixed together Commercial queries differ in many other SERP features Same query: Overview with an ad inside versus Overview without one The ad inside the Overview Changes in total ad count, position drift, capture sampling Same query: Overview versus no Overview The Overview itself Overview triggering varies by time and wording Same page types in a market with the placement versus one without, before and after launch A market-level effect Different layouts, competitors, seasonality and regulation How an ad inside the Overview could cost organic clicks
Picture the AI Overview as a shop assistant who answers the question before the customer reaches the aisles. An ad inside it is a promoter standing next to the assistant. Whether the promoter costs the aisles extra visits hinges on whether they are an extra person, or the same promoter who used to stand at the door. In search terms, the mechanisms are plausible rather than proven, and they pull in different directions:
- Displacement. An ad unit inside the Overview adds height above the first organic result. On mobile, where the Overview can take up much of the first screen, extra height can push your listing out of view.
- Attention. The ad sits in the block the user is already reading. If the Overview recommends a product type and the ad sells one, the ad is the nearest next step.
- Substitution. If an ad inside the Overview takes the place of a top ad that would have shown anyway, total ad load does not change and the extra organic loss could be small. Your own captures can show whether the ad adds to the page or relocates an existing one.
- Citation competition. If your page is cited in the Overview, the ad competes with your citation link as well as your organic listing.
A measurement design that isolates the ad
- Pick the market and queries. Use a market where ads within AI Overviews serve, such as Australia or the US. Choose 150 to 300 commercial queries where your page holds a fairly stable organic position and earns enough impressions for weekly CTR to mean something.
- Capture the full layout repeatedly. Use a SERP API or rank tracker that stores the full layout, not just your position, and capture each query several times a day on mobile and desktop. Record whether an Overview showed, whether an ad sat inside it, the number of top ads, shopping units, whether you were cited, and the pixel depth of your first listing.
- Turn captures into rates. For each query, device and day, calculate the share of captures with an ad inside the Overview and the share with an Overview at all.
- Join to Search Console. Pull daily clicks and impressions by query, page, device and country through the API. Rare queries are anonymised, so the panel will lean towards head and mid-tail terms.
- Compare within each query. Model CTR against the ad-inside rate with query fixed effects, controlling for position, Overview rate and top-ad count. The question you are asking: when this query shows an ad inside the Overview more often, does its CTR at the same position fall further than when it shows the Overview alone?
- Cross-check with a market difference. If you have comparable page sets in Australia and the UK, compare how CTR at fixed positions moved in each around the date the placement reached your market. The UK is a weak control because its results pages differ in other ways, but an effect that shows up in both designs deserves more weight than one that shows up in only one.
Test length should follow the variation you need, not a calendar habit. Ad rates move daily, so a few weeks of captures across 200 queries produces thousands of query-days. Low-volume queries need weekly aggregation, and windows that include major sales events (Black Friday, end-of-financial-year sales in Australia) change ad load and click behaviour at the same time, so either exclude them or model them separately.
Hypothetical example:
An Australian outdoor retailer tracks 200 commercial queries. Grouped naively, queries that ever show an ad inside the Overview have roughly half the CTR of queries that never do. Within each query, the gap shrinks to a fraction of a percentage point on mobile, disappears on desktop, and appears only on days when the ad inside the Overview comes on top of three top ads rather than replacing one. That pattern would point to total ad load on mobile as the mechanism, and the naive comparison would have overstated the placement’s effect several times over.
What Search Console and analytics can and cannot show
The Search generative AI performance report shows impressions for AI Overviews and AI Mode by page, country, device and date, with no clicks, CTR or queries. It was extended to all eligible sites worldwide on 31 August 2026. It shows which pages appear in AI features, not what happened to clicks when an ad was present.
AI Mode is harder still. Its clicks are counted in Search Console’s Web totals without a separate breakdown, and practitioner reports indicate they arrive in GA4 as google / organic. So you cannot yet isolate your own AI Mode clicks from first-party data, let alone split them by ad presence. Claims that ads in AI Mode cut organic clicks are, for now, inferences from captured responses rather than measurements.
If your comparison period reaches back to September 2025 or earlier, watch one more artefact. Google stopped supporting the
&num=100results parameter that month, which cut reported desktop impressions for many sites and improved average position, while clicks were affected far less, according to Brodie Clark’s analysis. CTR rose on paper, so a comparison spanning that date shows a change unrelated to ads.What result would change the decision
What you find What it suggests What I would do CTR at a fixed position falls further on high ad-rate days, mainly on mobile Displacement through extra ad load Use a lower CTR curve for commercial queries in placement markets when forecasting, and review paid coverage for queries where organic cannot hold clicks No difference once Overview presence is controlled The Overview, not the ad, is doing the damage Stop attributing losses to the placement and report Overview presence as the variable A drop only when the ad adds to existing top ads Total ad load matters more than where the ad sits Track top-ad count as a SERP feature in reporting A drop concentrated on queries where you are cited The ad competes with your citation link Report cited and uncited queries separately, because they behave like different channels If this were my site, I would stop asking whether ads in AI Overviews cost clicks in general and start recording SERP layout as a dimension of every commercial query I report on: Overview present, ad inside, top-ad count, and whether I am cited. That turns a question public data cannot yet answer into one your own data can answer for your queries in your market, and it gives you a defensible explanation when a stakeholder asks why CTR fell at a stable position.
Need help with this on your own site?
Niraj Raut works with businesses in Nepal, Australia, the UK, Europe and the US on technical SEO, ecommerce SEO, local SEO and AI search.
Add a reply Cancel reply
Replies are open to approved members. Log in or apply to join.