The short answer
AI Overviews cut clicks to the top organic result by roughly half when they appear. Ahrefs' December 2025 study of 300,000 keywords found the top-ranking page gets 58% fewer clicks when an AI Overview is present, up from a 34.5% drop the same team measured in April 2025. So the ai overviews traffic impact is real, it is worse than a year ago, and it is still widening. But here is the part most panic pieces skip: if your page is cited inside the AI Overview, you claw back a chunk of that loss. Seer Interactive's clickstream data shows cited pages earn about 35% more organic clicks than pages that rank but go uncited. The move is not to fight AI Overviews. It is to segment your content by intent, defend the pages an AIO can't satisfy, and start measuring informational pages by citations instead of clicks.
This post is a spoke of our larger guide to generative engine optimization. If you want the SEO-versus-AI framing first, read how GEO differs from traditional SEO, then come back.
What the numbers actually say
The findings worth trusting, and who measured them:
| Source | Finding | Sample |
|---|---|---|
| Ahrefs (Dec 2025) | 58% lower CTR for the top page when an AIO shows | 150k AIO vs 150k non-AIO keywords |
| Ahrefs (Apr 2025) | 34.5% lower CTR, same setup | earlier snapshot |
| Seer Interactive (Sep 2025) | Organic CTR falls from 1.76% to 0.61% on AIO queries (a 61% drop) | 25.1M impressions, 42 orgs |
| Seer Interactive (Sep 2025) | Cited pages: 0.70% CTR vs 0.52% uncited when an AIO shows | same panel |
The trend line matters more than any single number. Between April and December 2025 the CTR penalty on the top result nearly doubled. That is not because AI Overviews got worse; it is because Google expanded them to far more queries and made the answers more complete, so more searchers get what they need without scrolling.
Google's own framing is that these are "bounce clicks", low-quality visits where someone lands, skims, and leaves. Take that with salt: it is a convenient story for the party keeping searchers on its own surface. But there is a grain of truth in it. The clicks you lose to AIO are disproportionately informational, top-of-funnel, and low-intent. Those were rarely your best-converting visits.
Which pages lose the most
AI Overviews are answer machines. They fire hardest on queries they can fully answer in a paragraph, and barely at all on queries that need a person to compare, decide, or buy. That split is the whole strategy.
High loss, expect real declines:
- Definitional and "what is" queries ("what is a canonical tag")
- Simple how-to and step queries ("how to add alt text")
- Factual lookups, unit conversions, quick reference
Low loss, mostly protected:
- Comparison and "X vs Y" queries where the searcher wants a verdict
- Transactional and commercial queries ("best keyword research tool India")
- Tool pages, calculators, and interactive utilities
- Deep, opinionated, experience-driven content an AIO can't compress without gutting
A concrete example. A page ranking #1 for "what is bounce rate" is exposed; the AIO answers it in two sentences. A page ranking for "Ahrefs alternative" is not, because the searcher wants to weigh options and pick one, and an AIO summary rarely closes that decision. If your traffic sits in the first bucket, your ai overviews traffic decline will be steep. If it sits in the second, you will barely feel it. Audit your top pages by intent before assuming you are in trouble.
The non-obvious fix: get cited, not just ranked
Here is the insight that changes the plan. Ranking in classic organic results and being cited inside the AI Overview are two different games with two different payoffs. When your page is one of the sources linked inside the AIO, you recover a meaningful share of the lost clicks. Seer Interactive's panel puts cited pages at 0.70% CTR against 0.52% for uncited pages on the same AIO queries, roughly a 35% lift. The citation acts as both an endorsement and a second doorway back to your site.
So the goal shifts from "rank #1" to "rank well and get cited." Getting cited is its own discipline: clear extractable answers near the top of the page, strong topical authority, structured data, and content written the way answer engines quote it. We cover the mechanics in how to appear in Google AI Overviews and the broader craft in how to get cited by AI. The same practices carry over to ranking in ChatGPT and ranking in Perplexity, because all three reward the same signals.
Two technical levers do most of the work:
- Structured data. Marking up FAQs, how-tos, products, and articles gives engines clean, machine-readable answers to lift. See structured data for AI search and Google's intro to structured data.
- Answer-first content. Lead with the answer, then support it. This is the core of answer engine optimization and optimizing content for AI search.
Stop measuring informational pages by clicks alone
If half your top-funnel clicks are gone and are not coming back, the honest response is to change the metric, not to keep grieving over a chart that will never recover. For informational pages, the real outcome now is: did you get mentioned, cited, or quoted inside the AI answer? A page that earns zero clicks but is cited in the AIO for a high-volume query is still doing brand work and still feeding the decision that leads to a converting visit later.
That means running two dashboards side by side:
- Search Console for classic impressions and clicks, to spot which queries flipped to near-zero clicks while impressions held steady. Google explains how to read those numbers in the Performance report documentation. That divergence is the AIO fingerprint.
- AI-visibility tracking for whether you are cited across AI Overviews and LLM answers, to see whether the "lost" query is actually a won citation.
Reading them together is the point. A query with flat impressions and collapsed clicks in Search Console, cross-referenced against an AI-visibility tool showing you cited in the AIO, is not a loss; it is a channel shift you can defend. We go deep on the measurement side in LLM visibility tracking. DeployFlare pairs Search Console integration with AI-visibility monitoring so you can see, per query, which ones went zero-click and whether you are cited there.
A practical playbook
What to actually do this quarter:
- Segment your top pages by intent. Tag each as informational, comparison, or transactional. This tells you what is exposed and what is safe.
- Defend the money pages. Double down on comparison, tool, and buying-intent content, the pages AIO rarely satisfies and that still convert. Build more of them.
- Re-optimize exposed informational pages to get cited. Add answer-first summaries, FAQ and how-to structured data, and clear extractable claims. Losing the click but winning the citation is a fine trade.
- Rebuild the report. For informational pages, track citations and impressions, not just clicks. Set expectations with stakeholders before the traffic chart does it for you.
- Watch the divergence. Flat impressions plus falling clicks on a query is your signal it went AIO-heavy. Prioritize those queries for citation work.
The framing that keeps you sane: AI Overviews did not delete your audience, they moved where the answer gets delivered. If you understand GEO versus SEO and treat citation as the new ranking, the ai overviews traffic impact becomes a reallocation you can steer rather than a leak you can't plug. Start with a free rank check to see which of your pages sit on AIO-heavy queries, or compare capabilities on our Ahrefs alternative page.