If AI Overviews Ate Your Clicks, Stop Panicking — Change the Job of the Page

AI Overviews eat some of your clicks when Google answers the informational query on the results page and your blue link becomes optional. That is not a ranking collapse, and panic publishing will not restore 2022 CTR. First confirm the fingerprint: impressions steady or up, average position stable, CTR down on queries that now trigger an Overview. Then change the job of those pages — win the citation, capture brand demand, and stop treating every explainer like a click farm.

This spoke sits under the Answer Engine Optimization playbook. If the URL is not even eligible to be quoted, fix that first with how to get cited in AI Overviews. Lane work lives on /visibility.

The short answer

• Separate Overview zero-click from ranking loss before anyone rewrites a homepage.
• Industry CTR headlines vary by study, country, device, and whether the URL was cited. Treat a single “everyone lost X%” number as directional, not your KPI.
• Informational how-to and definition queries lose clicks first. Brand and many bottom-funnel queries behave differently.
• Being cited can recover some attention and trust even when raw clicks stay below the old SERP. It is not a full CTR restore.
• Protect revenue URLs. Reinvent commodity explainers. Report brand search plus citations beside sessions.

What is the impressions-up, CTR-down fingerprint?

The Overview incident looks like a healthy SEO report until you split the fraction. Impressions hold or rise. Average position barely moves. Clicks sag. CTR collapses because the denominator grew or held while the numerator shrank.

That pattern is not a vibe. Seer Interactive’s April 2026 AIO CTR update documented a version of it on cited brands in late 2025: in their cohort, brand-cited organic impressions more than doubled from September to October while clicks stayed roughly flat, so CTR halved. Seer could not attribute the impression surge to Google alone versus brands earning citations on more queries. Treat that as a measurement lesson, not a promise that your property will double impressions.

Signal	Overview fingerprint	Ranking-loss fingerprint	What to do first
Impressions	Flat or up on the cluster	Often down	Do not celebrate reach yet
Average position	Stable	Worse	If position fell, stop blaming Overviews
Clicks	Down or flat while impressions rise	Down with position	Split brand vs non-brand
CTR	Down on Overview-triggering queries	Down in line with rank	Freeze the query set
SERP check	Overview present; you may still rank	Overview optional; you slipped	Screenshot 20 queries
Brand queries	Often steadier	May fall if awareness dips	Keep brand as a control

Operator rule: never call it an Overview problem from a sitewide CTR chart. Call it from a frozen query set plus a SERP log.

How do Search Console impressions and clicks actually count here?

Google’s own counting rules explain why the fingerprint can look “healthy” on impressions. The Search Console impressions, position, and clicks help page is explicit for AI Overviews:

• A click is a click on an external link inside the Overview.
• An impression is counted under standard rules, and the link must be scrolled or expanded into view.
• The Overview occupies a single position. Every link inside it inherits that same position.

That last line is why average position can look stable while the click curve dies. You are not “ranking #1 in the Overview” in the blue-link sense. You are sharing one SERP slot with every other cited URL.

GSC object	What Google counts	How it can fool a weekly report
Impression	Link to your site was shown (Overview links need scroll or expand)	Reach looks fine while nobody leaves Google
Click	User left Search for your URL	The number you actually cash
CTR	Clicks ÷ impressions	Falls when impressions hold and clicks do not
Position	One slot for the whole Overview	“We still rank” hides a job change
Web performance	Overview and AI Mode sit inside the Web search type	You cannot read Overview-only CTR from the classic report alone

In June 2026 Google launched a Search generative AI performance report that isolates impressions in AI Overviews and AI Mode. The help-center writeup says the report was rolling out to a subset of properties, that it is impression-led, and that Search Labs experiments are excluded. If you do not see the report yet, do not invent a workaround metric. Use the Web performance report plus a manual citation log.

Checklist before you brief leadership:

• Frozen query set (same 30–80 queries for 90 days)
• Brand vs non-brand split in Search Console
• Device split (Ahrefs’ public CTR work is desktop-weighted; your mobile mix may differ)
• Manual SERP log: Overview yes/no, cited domains, your blue-link position
• Generative AI impression report if the property has it
• Clicks and impressions read as separate series, never CTR alone

If you only watch CTR, a citation win can look like a failure. Seer’s cited-brand example is the cautionary tale.

How do you confirm Overviews vs a ranking loss?

Use a triage table before anyone opens twenty content tickets.

Fingerprint	Overview-driven zero-click	Ranking loss / algorithm
Impressions	Flat or up	Often down
Average position	Stable	Worse
CTR	Down on specific queries	Down with position
SERP check	Overview present; you may still rank	Overview optional; you slipped
Landing pages	Informational cluster	Mixed / money terms too
Brand query traffic	Often steadier	May fall if awareness dips

Procedure:

1. Pull 28–90 days in Search Console for the suspected pages.
2. Annotate when Overviews visibly expanded on your priority queries. Manual SERP log, not a vendor screenshot of someone else’s niche.
3. Segment brand vs non-brand.
4. Spot-check 20 queries: Overview yes/no, your position, cited domains, whether your URL is in the Overview.
5. Label each query: Overview zero-click, ranking loss, mixed, or unrelated (seasonality, PR spike, site bug).
6. Only then change strategy.

If position collapsed, fix SEO fundamentals. If position held and CTR fell while Overviews appeared, change the page’s job. Do not chase the old CTR target as a moral right.

Failure mode: a redesign sprint on the homepage because “organic is down,” while the actual bleed is a glossary cluster that still ranks and still gets impressions.

What have Pew, Ahrefs, Seer, and Google actually reported?

Quote the primary page and the date. Do not average three recaps into a fake industry standard. As of August 2026, the public studies disagree on magnitude and even on direction in early 2026.

Source	Window / method	Headline number (theirs)	Hedge you must keep
Pew Research, July 22, 2025	900 U.S. adults; March 2025 browsing; 68,879 Google searches	Traditional-result click on 8% of visits with an AI summary vs 15% without; 1% clicked a link inside the summary; session ended on 26% of AI-summary pages vs 16% without	Panel, U.S., one month. Pew re-collected SERPs in April, so the Overview on the original visit may have changed. This is visit behavior, not GSC position-1 CTR.
Ahrefs, April 2025	300,000 keywords; desktop GSC CTR; Mar 2024 vs Mar 2025	Presence of an Overview correlated with ~34.5% lower position-1 CTR vs their forecast	Correlation plus a forecast, not a controlled experiment. Desktop only.
Ahrefs update, Feb 4, 2026	Same method; Dec 2023 vs Dec 2025	Presence correlated with ~58% lower position-1 CTR vs forecast	Two-year gap. Informational CTR also fell without Overviews. Ahrefs themselves call it a correlation.
Seer, Sept 2025 update	Informational queries across 42 orgs	Organic CTR on AIO-present queries described as down ~61% in that window	Client-mix study, not a census of the web.
Seer, April 2026 update	53 brands; 5.47M queries; 2.43B organic impressions; Jan 2025–Feb 2026	AIO-present organic CTR rose from a 1.3% Dec 2025 floor to 2.4% in Feb 2026 in their set; cited vs not-cited still mattered	Early-2026 bounce is two months. Seer says they do not expect a return to pre-AIO CTR.
Google Search Central, AI features	Official product docs	Google says Overviews can send people to a wider set of sites, and that clicks from Overview SERPs have been higher quality (more time on site)	That is Google’s claim. It does not restore your old click volume. Measure quality in Analytics, not in a press quote.

Pew’s other useful specifics, still from that July 2025 short read: about 18% of the Google searches in the study produced an AI summary; 58% of respondents had at least one such search in March 2025; question-shaped queries began with who/what/when/why 60% of the time produced a summary; one- or two-word searches did so 8% of the time, ten-or-more-word searches 53%. Those are Pew’s panel rates, not your coverage.

Ahrefs’ position table in the February 2026 update is the one operators misuse. They reported a forecast-vs-actual CTR impact that was largest at position 1 (~58%) and smaller further down (they list about −19% by position 10). That is still their model, still desktop, still not your vertical.

Operator rule:

• If you put a percentage in an exec deck, put the study name, date, and method on the same slide.
• If you cannot open the primary chart, say “reports vary” and show your Search Console delta.
• Never average Pew’s 8-vs-15 visit rate with Ahrefs’ 58% position-1 forecast into one “industry CTR.”

Your site’s before/after CTR on a frozen query set beats any LinkedIn screenshot.

Why do the study numbers disagree?

They are not measuring the same object.

Study object	What it captures	What it misses
Pew visit-next-URL	Did a person leave Google after this search?	Your rank, your citation, your GSC CTR
Ahrefs position-1 desktop CTR	Average GSC CTR for keywords that later show an Overview	Mobile, local packs, your brand mix
Seer brand-cohort CTR	How a set of client brands moved, including cited vs not cited	Brands outside that roster
Google product copy	How Google wants the feature understood	Your P&L
Your GSC + SERP log	What happened to this property	Other people’s headlines

Two mechanical reasons the percentages will never match:

1. Denominator games. CTR falls when impressions rise faster than clicks. Seer’s cited-brand October 2025 slice is the clean example: clicks ~flat, impressions doubled, CTR looked disastrous.
2. Selection. Overviews fire more on question and comparison queries. Pew saw summaries on 60% of question-word searches and 18% of searches overall. Ahrefs compared Overview keywords to informational keywords without Overviews. Those are different baskets.

Google’s AI features documentation also says Overviews are only shown when systems decide they add something beyond classic Search, and that they often do not trigger. A week with no Overview on a money query is a SERP-feature decision, not proof your page “recovered.”

Which query types lose the most clicks?

Pattern most teams see, and the public studies roughly rhyme with it — confirm on your own data.

Pew: longer queries, question syntax, and full sentences produced summaries more often. Seer’s 2026 update, in their informational slice, put comparison queries at a 95.4% Overview rate (267 of 280) and question-format queries at 85.9% (1,214 of 1,413). They also reported Overviews on about 36% of informational queries vs about 8% commercial and 5% transactional in that set. Those rates are Seer’s classification, not a Google transparency report.

Query class	Overview behavior (directional)	Click reality	Page job after the drop
Definition / what-is	High coverage	Sharp CTR pressure	Citation engine: dated definition + table
How-to / steps	Often answered on-SERP	Clicks shift to tools, video, or service	Procedure plus a gated calculator or audit
Comparison / best	Very high coverage in Seer’s slice	Partial recovery if you are the cited source	Criteria table, not a listicle
Local / services	Variable; Seer flagged “near me” informational as hot in their set	Maps + Overview interplay	NAP + offer sentence that matches the profile
Brand + navigational	Lower theft	Protect these — they fund the business	Keep fast, accurate, unblocked
Bottom-funnel commercial	Less “answer complete”	Still watch, panic less	Protect; do not starve for blog volume

Failure mode: cutting the informational cluster because CTR fell, then watching brand search decay six months later because you stopped being the cited explainer.

Checklist per query class:

• Overview present on 5 of the last 10 weekly checks?
• Are you cited, mentioned without a link, or absent?
• Did impressions rise while clicks fell on that exact query?
• Is there a protectable URL this page should hand off to?
• Would deleting this URL remove a fact Google currently attributes to you?

If the last box is yes, you do not have a “low CTR blog post.” You have a brand fact sitting in a weak template.

Should you fight zero-click or change the job of the page?

Fighting means trying to restore 2019 click curves on definitional queries. Changing the job means accepting some answers stay on Google while you compete to be the named source and the brand people search next.

Fight (usually loses)	Change the job (usually wins)
Doorway pages to dodge Overviews	Answer-first pages built to be cited
Clickbait titles that mismatch the Overview	Titles and leads that match the quoted truth
Hoarding facts behind lead forms on every URL	Give the answer; gate the calculator, template, or service
“SEO is dead” budget freeze	Split KPIs: citation share + brand + non-brand CTR
Publishing 40 thin posts to “outrank the AI”	Merge, deepen, and date the five pages that can be quoted

Zero-click is not permission to stop measuring. It is a reason to stop using sessions as the only scoreboard.

Google’s eligibility floor has not changed: the AI features page says a supporting link must be indexed and snippet-eligible. There is no special Overview markup. If you are going to change the job of the page, start from that floor, then write a passage that can stand alone. The citation checklist lives in the get-cited spoke.

What does “change the job” look like on a live URL?

Pick one informational URL that matches the fingerprint. Do not start a new content calendar.

Old job: rank, collect the click, convert later if you are lucky.

New job: be the sentence and the table the Overview lifts, put the brand in the unit the buyer actually reads, and send the people who want depth, a tool, or a service to a protectable URL.

Rewrite skeleton that only fits this problem:

1. Lead answer in the first 60 words — the definition or the decision, dated if it includes a number.
2. One citeable object — a table, a numbered procedure, or a bounded checklist. Not a mood paragraph.
3. Entity facts that match About / schema — so the Overview does not invent your pricing, city, or offer.
4. A next step that is not “read more blog.” Demo, audit, calculator, pricing, or a case page you can stand behind.
5. A monthly SERP + citation log — same query, same device class, screenshot, cited domains.

Page element	Old click-farm version	Citation + brand version
H1 / title	“Ultimate guide” + year stuffed	The question people type, answered in the title
First paragraph	Brand mythology	The answer a model can lift without you
Middle	2,000 words of synonym coverage	Tables, steps, caveats next to claims
Proof	“Trusted by industry leaders”	Named receipts you can defend
CTA	Newsletter on every URL	One handoff to a protectable URL
Measurement	Sessions	Citation on the frozen query + brand search + leads

I will cite receipts I can defend on a sales call: SEO certified since 2021, 500+ automations built, 20,000+ hours on agentic systems, 35,000+ hours saved for clients, hundreds of production sites. I will not invent an “Overview recovery rate” for your deck. If you cannot say the number out loud, do not put it in the passage you hope Google lifts.

Which pages do you protect vs reinvent?

Protect (revenue and trust)

• Pricing / packages
• Demo / contact / audit landing pages
• High-intent service pages
• Case studies with named outcomes you can stand behind
• Brand FAQ that prevents hallucination

Reinvent (informational CTR casualties)

• What-is and glossary pages → citation engines (definitions, tables, dated claims)
• Generic how-tos everyone ranks for → proprietary steps, screenshots, or calculators
• Thin listicles → merge or kill
• Comparison posts on queries that now always show an Overview → criteria tables, not ten near-duplicate “best” pages

Candidate for demotion

• Pages with rising impressions, collapsing CTR, no citation, no brand assist, no leads — for two full quarters

Decision	Keep when	Merge or kill when
Protect	Pipeline or trust depends on it	Never “optimize” it into a glossary
Reinvent	You still rank or still get impressions	You cannot name a citeable object after one rewrite
Demote	Two quarters of no citation, no brand, no leads	You are deleting a fact Google still attributes to you — fix the fact first

Checklist for each reinvented URL:

• Lead answer in the first 60 words
• One table or procedure worth citing
• Clear next step to a protectable URL
• Entity facts aligned with About / schema
• No nosnippet / max-snippet:0 accident on the answer block
• Owner assigned to a monthly SERP + citation log
• Index request only after the fetched HTML contains the answer

Do not open twenty tickets on day two. Overviews punish undifferentiated libraries. They do not reward volume for its own sake.

Do citations recover clicks — or just brand?

Sometimes clicks. More often a mix of residual clicks plus brand. Almost never a full return to the pre-Overview baseline.

Seer’s April 2026 writeup, in their informational slice, said being cited delivered about +120% more organic clicks per impression than not being cited on the same Overview-present SERP, and still underperformed no-Overview queries by about −38%. They also gave a per-million-impression sketch: roughly 33,500 clicks with no Overview, about 20,700 when the brand was cited, about 9,400 when an Overview showed and the brand was not cited. That is Seer’s cohort math. Your vertical will not photocopy it.

Pew’s panel is colder on in-unit clicks: users clicked a link inside the summary in 1% of visits to those pages. If your strategy is “get cited so the Overview becomes a traffic source,” Pew is the study that should humble that slide. Citation is still worth doing. It is a terrible plan if the only success definition is “the Overview becomes our new #1 referrer.”

Outcome	What you can honestly claim	What you cannot
Cited in the Overview	Brand in the unit people read; a shot at residual clicks	“We got our old CTR back”
Ranked, not cited	You are still in the candidate pool	The page is doing strategic work
Impressions up, clicks flat, cited	Possible citation expansion — check query count	Automatic failure
Impressions up, clicks down, not cited	Classic fingerprint; reinvent or demote	“SEO is dead”
Brand search up after citation work	The page may be doing brand work	You can skip the citation log

Treat citation as a visibility KPI adjacent to clicks. A page that lost a large share of its old CTR (your GSC number — measure it) but became the cited source on three money queries may still be doing strategic work. Pair click reporting with a frozen panel, not with a vendor “AI traffic” estimate.

How do you replace informational CTR in the scoreboard?

Rebuild the funnel math executives already understand. Do not invent a vanity “AEO score.”

1. Informational impressions — still show reach. Keep the series.
2. Overview citation rate — share of Overview-triggering queries in the frozen set where you are credited with a link.
3. Assisted brand search — branded queries and direct after content or campaign pushes.
4. Conversion on protectable URLs — demos, audits, purchases.
5. Quality of the clicks you still get — Google claims Overview-SERP clicks spend more time on site. Prove or disprove that in Analytics for your landing pages. Do not paste the docs sentence into a board deck as if it were your result.

KPI	Source	Review	Kill criterion
Non-brand CTR on frozen queries	Search Console	Weekly	Useful only after you split Overview vs rank
Clicks (not CTR) on the same set	Search Console	Weekly	If clicks and impressions both fall, demand problem
Citation rate	Manual log + gen-AI report if available	Weekly	0% for a quarter on money queries → reinvent or stop
Brand queries	Search Console	Weekly	Free-fall after you deleted explainers → you cut demand
Protectable conversions	Analytics / CRM	Weekly	The only number that pays rent
Time on site from organic	Analytics	Monthly	Test Google’s “higher quality” claim; do not assume it

If informational sessions fall while (2)+(3)+(4) hold or rise, you are adapting. If all four fall, you have a real demand problem — not just an Overview problem.

Semrush-style share-of-voice views help on classic SERPs. They do not replace a manual Overview citation log. I still use competitive SERP tools the way I use a packing list: to see shape, not to grade your week.

How do you report this to executives without a freeze?

Bring one page. Not a TED talk. Not “AI killed SEO.” That phrase invites a budget freeze. Invite a redesign of page jobs.

Slide structure

1. Fingerprint: impressions / position / CTR for the affected cluster. Show clicks and impressions as two lines.
2. SERP reality: five annotated screenshots with Overview on/off and cited domains.
3. Study context: “Third-party CTR studies disagree on magnitude. Ahrefs published a 34.5% then 58% position-1 forecast gap. Pew measured 8% vs 15% traditional-result clicks in a March 2025 U.S. panel. Seer saw compression in 2025 and a partial bounce in early 2026. Here is our delta.”
4. Decision: protect vs reinvent list with owners.
5. New scoreboard: citation rate + brand search + pipeline, reviewed monthly.
6. Ask: approve a 30-day reinvent list or fund a visibility audit.

Exec question	Bad answer	Better answer
“How much traffic did AI steal?”	A viral percentage	Our frozen-set CTR and click delta, dated
“Should we pause the blog?”	Yes / no as a slogan	Protect vs reinvent list, two quarters of evidence
“Are we cited?”	“We do AEO”	20-query log with domains
“Is Google lying about quality?”	Debate	Time-on-site for the clicks we still get
“What do you need?”	More posts	Owners, 30 days, two reinvented URLs, two protected URLs

Avoid: asking for sympathy. Ask for a decision.

What is a 30-day response that is not a content panic?

Days	Action	Done looks like
1–5	Triage fingerprint; freeze query set	Each priority query labeled Overview vs ranking loss vs mixed
6–12	Eligibility + extractability on the top 5 URLs	Lead answers live; no nosnippet surprises; URL Inspection clean
13–20	Reinvent two informational pages; strengthen two protect pages	Shipped HTML, internal links to protectable URLs, index requests
21–30	Citation + brand log; exec one-pager	Keep / merge / invest decision, not a new editorial calendar

Do not open twenty content tickets on day two.

Weekly ops after the first month:

Cadence	Owner	Artifact
Weekly	SEO lead	10-query Overview citation log
Weekly	Analytics	Brand vs non-brand sessions, clicks, CTR
Biweekly	Content	Two reinvent tickets shipped or explicitly deferred
Monthly	Marketing lead	Exec one-pager with protect/reinvent status
Quarterly	Leadership	Budget split: classic SEO vs answer-engine work

Miss the weekly log and you will argue from vibes again by week six. The ritual is the product.

When is a visibility audit the right next step?

Run or hire an audit when:

• You cannot tell Overview zero-click from a ranking event after two weeks of Search Console work
• Executives want an external prioritization hammer before cutting headcount or content
• Citation gaps show competitors owning every Overview on your category terms
• Entity conflicts or technical eligibility issues keep blocking wins
• The generative AI report, the Web report, and the SERP screenshots tell three different stories and nobody can reconcile them

DIY the triage ladder above first if you have an SEO lead with Search Console access. Hire when the argument is political or the data is muddy. The lane page is /visibility. The system map is the AEO playbook.

I have been SEO certified since 2021. The certification does not make Overviews kinder. It makes the fingerprint boring to diagnose if you actually open Search Console.

What does good look like 90 days later?

You will not get every informational CTR back. You should be able to show:

• Clear labels: which clusters are Overview-affected vs rank-affected
• A handful of money queries where you are the cited source
• Brand search not in free-fall
• Protectable URLs converting at least as well as before on a lead or revenue basis
• A kill/merge list for pages that help neither citations nor pipeline
• An exec one-pager that quotes a named study only as context, and your GSC as the decision

90-day artifact	Pass	Fail
Frozen query log	12 weeks, same queries, Overview + citation marked	Ad-hoc screenshots
Protect list	Owners, no “optimize into a blog”	Revenue URLs rewritten as glossaries
Reinvent list	Two URLs shipped with tables and handoffs	Twenty outlines, zero HTML
Brand series	Stable or up	You deleted the explainers and called it “focus”
Pipeline	Protectable conversions held	Sessions were the only slide

If none of those moved, you optimized titles while the page jobs stayed wrong.

FAQ

What CTR changes have studies reported?

Reports vary by study, market, device, and whether the page was cited in the Overview. Ahrefs published a forecast-based position-1 gap of about 34.5% (April 2025) and about 58% (December 2025 data, posted February 2026). Pew measured traditional-result clicks on 8% of visits with an AI summary versus 15% without in a March 2025 U.S. panel. Seer reported 2025 compression and a partial early-2026 bounce in their brand cohort. Quote the named study and date; prefer your own Search Console delta.

Which query types lose the most clicks?

Definitional and generic how-to queries usually see the sharpest CTR pressure when an Overview fully answers the question. Pew found summaries more often on longer and question-shaped searches. Seer’s 2026 informational slice put comparison and question queries at very high Overview rates. Brand, navigational, and many bottom-funnel commercial queries tend to hold up better — confirm the split on your property.

Does being cited recover some clicks?

Sometimes. Seer’s 2026 cohort math said a citation beat “Overview present, not cited” on clicks per impression and still lagged queries with no Overview. Pew found people rarely clicked a source inside the summary (1% of those visits). Treat citation as brand-plus-residual-clicks, not a restore of informational CTR.

Should I cut informational content?

Not by default. Cut or merge pages that create neither citations, brand lift, nor leads. Reinvent commodity explainers into citeable assets. Protect revenue URLs. Blanketing “delete the blog” usually trades today’s CTR pain for tomorrow’s demand hole.

How do I report this to executives?

Show the fingerprint chart with clicks and impressions as separate lines, five annotated SERPs, your site’s CTR delta (not a viral statistic), a protect-versus-reinvent list, and a scoreboard that includes citation rate and brand search. Ask for a 30-day decision, not sympathy.

When is a visibility audit the right next step?

When you cannot separate Overview zero-click from ranking loss, when competitors own every category Overview, or when leadership needs a prioritized 30/60/90 before more publishing. Start with the triage ladder. Escalate when the data or the politics exceed in-house bandwidth.

CTA

If Overviews chewed your CTR, stop mourning the old click curve — prove the fingerprint, then redesign what those pages are for.

Lane: /visibility · Book a visibility audit.