AI Overviews CTR: what happens to clicks and how to measure yours

What is known about AI Overviews and clicks

Four sources, measured different ways:

Read side by side, the independent studies put the click gap on affected searches at roughly half or more: 8% versus 15% in Pew's data is about 47% lower, Ahrefs reports 58% lower, and Seer's September figures are about 62% lower. All three are correlations, not proof of cause, and Seer covers informational searches only. Google's developer documentation adds that clicks from results pages with AI Overviews are "higher quality", meaning people are more likely to spend more time on the site (Google Search Central (opens in a new tab)).

If you want a broader baseline alongside these numbers, the free marketing audit scores your website, SEO and AI visibility.

Why the impact differs by query type

The studies and Google agree on one thing: the type of question decides the effect.

• Quick-answer questions lose most. Google itself says that for these, "people may be satisfied with the initial response and not click further." "What is a root canal" or "how long does Botox last" fit that pattern.
• Complex questions keep more clicks. Google says people keep clicking through on many other kinds of questions to dig deeper or act.
• Short and local searches see fewer AI Overviews at all. In Pew's data, one- or two-word searches produced an AI summary 8% of the time, versus 60% for searches starting with a question word.
• Being cited helps a little. Seer's cited-versus-uncited figures suggest some recovery when your page is the one linked.

For most practices, that means the blog and condition pages feel it first, and the service, insurance and booking pages feel it least.

Measuring it in your own Search Console data

You don't need a study; you have the data. Here's a comparison anyone can run in about an hour. Our guide to using Google Search Console covers setup if you're new to it.

1. Pick two equal periods. Use the Performance report's Compare option, for example the last three months against the same three months a year earlier, which keeps seasonality similar. Google's help suggests weekly or monthly granularity to avoid comparing a Monday with a Saturday (Search Console Help (opens in a new tab)).
2. Isolate question-style searches. Add a Query filter, choose Custom (regex) and use a pattern such as ^(how|what|why|when|does|do|can|is|are|should)\b. Search Console uses RE2 syntax and isn't case-sensitive by default.
3. Build the comparison group. Repeat with "Doesn't match regex" on the same pattern. Exclude your practice name too, since branded searches behave differently.
4. Compare CTR at similar positions. CTR is clicks divided by impressions, and average position is the position of your topmost result (Search Console Help (opens in a new tab)). Export the queries, group them into position bands (1–3, 4–10) and compare CTR band by band. A CTR drop at a steady position is the signal; a drop caused by slipping from position 2 to 7 is a different problem.
5. Find the pages Google links in AI features. The generative AI performance report (opens in a new tab) lists pages with impressions in AI Overviews and AI Mode. Check those pages' CTR trends in the main report.
6. Remember what's included. Google counts AI features traffic within the Web search type of the main report, so AI Overview impressions and clicks are already in your totals.

Pages most likely to lose clicks

• General condition explainers: "what is plantar fasciitis", "symptoms of sleep apnea". National health sites already dominate these, and a summary often answers them.
• Definitions and simple facts: "what does a periodontist do", "how long does a filling take".
• National averages: "average cost of braces" with no local detail.
• Lists that could be summarized in two lines: "five signs you need a new mattress".

These pages can still earn their keep as sources of trust and as citations, but judge them on more than raw clicks.

Pages that can still win the click

• Service pages where the next step is a decision: "Mohs surgery in Tucson", "dental implants in Lakewood".
• Pages with local specifics only you can supply: insurance accepted, your prices or ranges, new-patient wait times, parking.
• What-to-expect pages written from your own process.
• Clinician bios and first-person explanations. Google says people are more likely to click content that helps them learn more, such as in-depth reviews and first-person analysis.
• Anything with an action: call, get directions, book, check coverage.

Adjusting pages: answer the question, offer the next step

The answer is going to be summarized either way. Make sure the summary comes from you, and give the reader a reason to come through.

1. Answer first. Put a clear 40-to-60-word answer at the top. It makes your page easier to cite; see answer engine optimization.
2. Add what a summary can't. Your clinician's experience, your process, photos of your office, local logistics.
3. Offer the next step right after the answer: "Check whether we take your plan", "See our self-pay prices", "Book a consultation" in your own scheduling tool.
4. Link informational pages to service pages, so a reader who arrives with a question can find the visit that answers it.
5. Re-measure in 8 to 12 weeks with the same Search Console comparison.

None of this promises that clicks return to earlier levels. It makes the clicks you do get more likely to become patients. To see which searches show AI Overviews at all, set up AI Overview tracking; for the wider picture, see Google AI Overviews.

Where Medfy fits

Medfy's Google reporting puts Search Console and Google Analytics in the same dashboard, and its first-party analytics counts visits from AI assistants, which our guide to traffic from AI assistants explains how to measure yourself. A weekly brief by email summarizes what moved. To see where you stand now, start with the free marketing audit.

Frequently asked questions

Do AI Overviews reduce website traffic?

On many searches, yes. Pew Research Center found users clicked a traditional result on 8% of Google visits with an AI summary, versus 15% without, and Ahrefs and Seer Interactive report similar or larger gaps on informational searches. Google says total organic clicks have stayed relatively stable. The effect depends on the question type, so check your own Search Console data.

How can I tell whether AI Overviews affect my clicks?

In Search Console, compare two equal periods. Filter question-style queries with a Custom (regex) filter, then compare their click-through rate with non-question queries at similar positions. A CTR drop at a steady position on question searches, with steady CTR elsewhere, points to AI Overviews. The generative AI report shows which pages Google links in AI features.

Which pages are most affected by AI Overviews?

Pages that answer quick, general questions: condition explainers, definitions, simple facts and national cost averages. A summary can answer those without a click, and national health sites often dominate them anyway. Service pages, pages with local details such as insurance and prices, what-to-expect pages and anything with a clear action tend to keep their clicks better.

Should I change my content because of AI Overviews?

Adjust it rather than rewrite it. Put a clear answer at the top of each page, add details only your practice can give, and offer an obvious next step, such as checking insurance or booking. Link informational pages to the matching service pages. Then measure again in Search Console after 8 to 12 weeks. No change can promise traffic returns.