What AI Overviews Are Costing You Right Now
Position One No Longer Means What It Used To
Your rankings may look fine. Your impressions are climbing. But your clicks are quietly disappearing—and AI Overviews are responsible. Ahrefs studied 300,000 keywords in December 2025 and found that AI Overviews now reduce the average position-one CTR by 58%. That is not a rounding error. A keyword delivering 580 clicks from 1,000 impressions now delivers roughly 243. This happens before any other SERP feature gets involved.
The Ahrefs finding is not an outlier. Pew Research Center tracked 900 U.S. adults across 68,879 real Google searches in March 2025. Users who saw an AI summary clicked a traditional result just 8% of the time. Users who saw no AI summary clicked 15% of the time. Only 1% of those who encountered an AI Overview clicked any link inside the summary itself. For keyword categories where AI Overviews appear consistently, you are competing for roughly half the clicks you once earned—even at position one.
How Widely AI Overviews Spread During 2025
Understanding the keyword impact starts with understanding the rollout. Semrush analyzed more than 10 million keywords from January through November 2025. AI Overviews began the year appearing on 6.49% of queries. They peaked at 24.61% in July. By November they stabilized at 15.69%. Google pulled back from query types that were not serving users well. That calibration directly determines the value of every keyword on your target list.
Industry distribution matters as much as overall prevalence. Semrush found that Science leads all verticals at 25.96% AI Overview presence. Computers and Electronics follows at 17.92%. Food and Drink experienced the fastest growth of any category since March 2025. Meanwhile, Ahrefs data found Shopping carries just 3.2% AI Overview presence and Real Estate sits at 5.8%. Those numbers should directly influence where you invest your content budget.
How to Spot the Damage Before It Compounds
Most teams discover the problem months after it starts. AI Overviews do not trigger a Search Console alert. Rankings hold while CTR quietly drops. The standard dashboard does not separate AI Overview queries from traditional results. Pew found that 26% of users who encountered an AI summary ended their browsing session entirely. That compares to 16% of users who saw only traditional results. AI Overviews are not just intercepting clicks. They are ending user journeys before anyone reaches your site.
Use this checklist to determine whether AI Overviews are already eroding your keyword strategy.
AI Overview Exposure Self-Assessment
- More than 30% of your target keywords are question-format queries beginning with “what,” “how,” “why,” or “which.” Pew Research found that 60% of question-based searches triggered an AI summary in March 2025.
- Your keywords average 8 or more words in length. Analysis of 2.3 million keywords found these are 7× more likely to produce an AI Overview than shorter queries.
- Your primary content category falls in Science, Health, Education, or Computers and Electronics. Ahrefs measured AI Overview presence of 43.6%, 43.0%, and 17.92% in these sectors respectively.
- Your top pages include “best [product]” comparison or buyer’s guide content. BrightEdge documented 83% AI Overview presence on these queries in November 2025—up from just 5% a year earlier.
- Your organic CTR has declined year-over-year even though your average ranking position held steady or improved.
- You are targeting broad informational keywords with no transactional modifier. These match the query profile most affected by AI Overviews in every major study.
- You have not separated AI Overview queries from traditional results in Search Console. You have no baseline to measure the true impact.
- Your brand appears in fewer than 5 third-party sources online. Thin external citation footprints consistently predict AI Overview exclusion regardless of organic ranking.
0–2 items checked: Your keyword portfolio carries low AI Overview exposure. Monitor quarterly. 3–5 items checked: AI Overviews are likely already suppressing your CTR. Audit your keyword intent distribution immediately. 6–8 items checked: Your keyword strategy is misaligned with the 2026 SERP. Restructuring by intent tier is urgent.
Which Keyword Categories Are Now High-Risk
Informational Keywords Were Hit First and Hardest
Most practitioners assumed AI Overviews would stay narrow—handling simple factual queries while leaving competitive keywords untouched. The data does not support that assumption. Semrush found that in January 2025, 91.3% of all queries triggering an AI Overview were informational. Broad how-to content, explainers, and definition pages were hit earliest. By October 2025, informational queries had dropped to just 57.1% of all AIO-triggering searches. Commercial and transactional keywords now absorb a rapidly growing share of AI summaries. Building your content calendar around top-of-funnel educational keywords puts you squarely in the path of both waves.
The organic CTR data from Seer Interactive’s longitudinal study is direct. Seer tracked 3,119 search terms across 42 organizations and 25.1 million organic impressions from June 2024 through September 2025. For queries where an AI Overview appears but your brand is not cited, organic CTR dropped 61% year-over-year. An informational keyword that once reliably drove traffic now offers roughly one-third of the clicks it produced before. This happens even when your page still ranks in position one.
Broad Non-Branded Queries Lose the Most Ground
Not every informational keyword suffers equally. Severity depends heavily on whether the keyword is branded. Amsive analyzed 700,000 keywords across five industries—finance, education, SaaS, healthcare, and pets. Non-branded keywords saw an average CTR decline of 19.98%. Keywords ranking outside the top three positions experienced a steeper drop—27.04% on average. These were once considered reliable traffic drivers. They now represent a structural risk in any keyword portfolio that relies on them for volume.
Consider a broad non-branded keyword like “how to calculate marketing ROI.” It is informational. It has a clear answer that fits in a paragraph. Google’s AI has strong incentive to resolve it without requiring a click. Targeting that keyword with a traditional 1,500-word guide no longer works by default. It now requires a second step: optimizing for citation inside the AI Overview itself. That is a different problem than ranking. Most keyword strategies are not built to solve both simultaneously.
Comparison and Buyer’s Guide Content Faces a New Reality
The most alarming shift for content-heavy businesses is what happened to “best [product]” queries in 2025. BrightEdge’s year-over-year analysis compared November 2024 to November 2025. Informational shopping queries—”best air fryer,” “best project management software”—went from 5% to 83% AI Overview presence in a single year. That 78-percentage-point swing is the largest single-year shift in BrightEdge’s entire dataset. If your business depends on ranking for buyer’s guides or comparison articles, the SERP you optimized for in 2024 no longer exists.
Pure transactional queries remained at just 13–14% AI Overview presence. “Buy air fryer” and specific product name searches are still largely protected. Google has drawn a clear operational line between the research phase and the purchase phase. Comparison content sits on the research side. AI Overviews now dominate there. Purchase-intent content is relatively protected. That distinction should directly reshape how you categorize and prioritize keyword groups.
Long-Tail Informational Keywords Are Not a Safe Refuge
Many SEO teams shifted toward long-tail terms after early AI Overview rollouts. The reasoning was that Google would target only high-volume head terms with AI summaries. That assumption was wrong. Ahrefs found that 46% of all AI Overview appearances involve long-tail queries of seven words or more. Question-format queries account for 57.9% of all AIO appearances. Queries with eight or more words are roughly 7× more likely to trigger an AI Overview than one-to-two-word queries. Long-tail keywords do not escape AI Overviews—they attract them.
The learning platform Chegg experienced this dynamic at scale. Search Engine Journal reported that Chegg disclosed a 49% decline in non-subscriber traffic between January 2024 and January 2025. The timing aligned directly with AI Overviews answering homework and study questions that had previously driven traffic to the platform. Chegg’s content was optimized and authoritative. Their pages were well-ranked. None of that protected the business once AI Overviews resolved user queries before anyone reached the site.
Keyword Categories That Still Deliver Clicks
Branded Queries Show Consistent Resilience
AI Overviews are expanding into navigational searches—queries where users look for specific brands or destinations. Even so, branded keywords behave fundamentally differently from non-branded terms. Amsive found that only 4.79% of branded keywords triggered an AI Overview in their dataset. More importantly, branded keywords that appeared alongside an AI Overview saw an average CTR increase of 18.68%. When a user searches specifically for your brand and Google generates an AI summary that names you, the summary functions as a recommendation. The user was already looking for you. The AI confirmation reinforces their intent to click.
This creates a direct strategic implication. Brand-building is now also keyword strategy. The stronger your brand recognition, the lower your exposure to AI Overview cannibalization. Investment in branded search is no longer separable from investment in organic search health. They are the same decision approached from different angles.
Transactional Keywords Remain Structurally Safer
Google has consistently shown restraint with purchase-intent queries. Big Leap’s analysis found that only 10% of commercial or transactional keywords generate AI Overviews. AI is less likely to replace clicks when users are ready to buy. When a user searches “project management software for remote teams,” they need to visit a site. They must start a trial, compare pricing, or speak to sales. AI cannot complete that transaction for them. The user still needs to click. That structural necessity provides meaningful protection for transactional keyword targeting.
Additional context comes from research on Google’s AI Mode—a more aggressive AI experience than standard AI Overviews. Even there, 69% of transactional searches still result in a click to a website. The pattern is consistent across AI search experiences: users completing transactions click through. Concentrate your keyword investment at the point in the funnel where buying decisions are made. Top-of-funnel informational keywords are where AI Overviews are resolving queries before anyone reaches your content.
Local Keywords Carry a Structural Advantage
If you operate a local or regional business, the AI Overview threat is substantially smaller than industry averages suggest. Ahrefs data shows only 7.9% of local searches trigger an AI Overview. Local queries have inherent characteristics that make them poor candidates for AI summaries. They depend on proximity. They require up-to-date business information like hours and availability. They typically resolve through a call or a physical visit rather than content consumption. A search for “HVAC repair near me” is not a knowledge question. It is a proximity and preference decision that AI cannot make for the user.
This structural advantage extends across related industries. Real Estate sits at 5.8% AI Overview presence. Shopping carries just 3.2%. For businesses dominated by local-intent terms with city, neighborhood, or “near me” modifiers, traffic risk is meaningfully lower than industry statistics suggest. Local keyword investment remains one of the highest-ROI categories in the current SERP environment precisely because AI Overviews are less active there.
Highly Specific Technical Keywords Hold Their Value
AI Overviews synthesize answers from existing public sources. They cannot answer questions that depend on proprietary data, unique expertise, or context-specific situations requiring professional judgment. A keyword like “how to interpret GA4 multi-touch attribution for a B2B SaaS funnel” has several characteristics that reduce AI Overview risk. It is highly specific. It implies a need for customized guidance. It signals that the user wants practitioner expertise rather than a synthesized summary. Metrics Rule, an SEO and AI search consultancy, consistently finds in client audits that hyper-specific technical queries continue to drive meaningful CTR. Adjacent broader queries, by contrast, are losing significant traffic to AI Overviews.
Turning AI Overviews Into a Visibility Asset
Being Cited Changes the Entire Equation
Most practitioners focus only on the traffic they are losing to AI Overviews. They miss the opportunity to become part of the answer. Seer Interactive’s study found that queries where your brand is cited in an AI Overview produce 35% more organic clicks. That compares to queries where an AI Overview appears but your brand is not cited. Paid clicks increase by 91%. Users who click after seeing an AI Overview are higher-intent. They read the AI summary, saw your brand mentioned as a source, and chose to investigate further. That user profile converts at meaningfully different rates than typical organic traffic.
This creates a different strategic posture. Rather than treating AI Overviews purely as a threat, treat citation frequency as a keyword strategy metric. The question is not just “which keywords should I rank for?” Add a second question: “Which of those can I earn citation status on?” Both lists overlap—but they are not identical. Optimizing for both requires different content and technical decisions than traditional keyword targeting alone.
What Gets Cited and What Does Not
Google’s citation selection no longer mirrors organic rankings with the precision it once did. Citation pattern tracking found the overlap between AI Overview citations and top-10 organic rankings dropped from 76% in mid-2025 to approximately 38% by early 2026. Ranking position one no longer automatically earns AI Overview citation. SE Ranking research found that approximately 65% of pages cited by Google AI Mode include structured data markup. Research compiled across multiple studies found that 85% of AI Overview citations came from content published within the last two years. A full 44% came from 2025 alone.
Content structure matters in a specific, measurable way. Pages organized into sections of 120 to 180 words between headings receive 70% more ChatGPT citations than pages with shorter, fragmented sections. Self-contained answers that fully resolve a query in a single passage are the format AI systems most readily extract and cite. If your page buries its core answer after a long introduction, AI systems will frequently skip it. They will cite a page that leads with the direct answer instead. That is a structural content problem—distinct from a keyword selection problem.
Content Freshness Is Now a Competitive Advantage
AI citation systems demonstrate a strong preference for recent content. SE Ranking’s November 2025 study found that content updated in the past three months averages 6 citations. Outdated pages average just 3.6 citations. That is a 67% citation-rate advantage for actively maintained pages. A well-structured page from 2023 will consistently lose AI citation opportunities to a comparable page updated in 2025. This happens even if the older page has stronger backlinks and domain authority.
This creates a maintenance imperative that most teams underinvest in. Publishing new content while letting existing pages go stale is the standard content calendar approach. In the AI Overview era, that approach systematically disadvantages your citation profile. A quarterly content refresh targeting your highest-impression pages now delivers citation value. Updating statistics, refreshing examples, and adding structured data accomplish what net-new publishing cannot replicate alone.
External Mentions Drive Citation Probability
One of the most consequential citation findings involves where AI citations actually originate. Airops data found that brands are 6.5 times more likely to be cited in AI responses through third-party sources than through their own domain. This directly connects external brand authority building to the mechanics of AI Overview citation. A strategy that optimizes only your own pages while neglecting external brand mentions misses the primary pathway to AI citation visibility. Entity presence on LinkedIn, G2, and industry-specific directories creates the external citation network AI systems rely on. Those networks determine which brands appear in synthesized answers.
Measuring Performance in the AI Overview Era
Why Standard Metrics Miss the Shift
Google Search Console does not currently separate AI Overview traffic from traditional organic traffic. Clicks from AI Overview citations appear under the same “Web” search type as regular organic clicks. Your standard CTR reporting shows a blended figure that obscures the AI Overview impact on specific keyword clusters. Practitioners who segment AI Overview data consistently find a disconnect: impressions have risen significantly while CTR has declined. AI Overviews often generate two impression counts for the same page. The standard traffic dashboard does not reveal this tension between growing visibility and shrinking engagement.
The measurement gap creates a practical problem. Teams not auditing which keywords trigger AI Overviews are making investment decisions based on incomplete data. A keyword showing 15,000 monthly impressions at 0.9% CTR in Search Console may have dropped from a 2.3% baseline. That drop happened because AI Overviews appeared mid-year. Without segmenting by AI Overview presence, that keyword looks like a stable performer. It is actually a declining one.
Metrics That Actually Reflect Current SERP Reality
Effective measurement in the AI Overview era requires tracking signals that traditional SEO reporting does not capture. Citation frequency is now a meaningful leading indicator of branded traffic and brand authority. It measures how often your brand appears as a named source inside AI Overviews for your target keywords. Share of voice in AI responses gives you a picture of brand prominence that organic ranking alone no longer provides. Measure it by manually testing your top 30 to 50 queries monthly across Google AI Overviews, ChatGPT, and Perplexity. Users arriving after AI exposure are pre-qualified. Their landing page behavior should be tracked separately from standard organic sessions.
Think of it this way. Treating keyword rankings as your primary success metric in 2026 is like navigating a city using a map printed in 2022. The roads still exist, but new intersections have appeared, some old routes are now one-way, and landmarks you relied on have changed. The map is not wrong—it is incomplete. Adding AI citation rate, share of voice in AI responses, and CTR segmented by AI Overview presence gives you the updated map. The current SERP requires all three.
Isolating AI Overview Impact From Other Traffic Changes
One of the most common errors in 2025 traffic analysis was attributing AI Overview CTR declines to seasonal patterns or algorithm updates. The correct approach compares CTR trends for keyword clusters where AI Overviews are present against trends for equivalent clusters where AI Overviews do not appear. Seer Interactive’s methodology provides a useful template. They tracked 3,119 search terms from June 2024 through September 2025. They segmented keywords by AI Overview presence. Then they compared before-and-after CTR for the same terms. That segmented comparison isolates the AI Overview effect from everything else. Use the same approach before drawing conclusions about which keyword categories are performing or declining.
A Practical Keyword Targeting Framework for 2026
Tier 1 — Protect and Prioritize Transactional Keywords
Transactional keywords should receive your highest content investment and technical optimization priority. The data consistently shows they carry lower AI Overview presence. They maintain click-through rates far closer to pre-AI Overview levels. Separate your keyword portfolio by intent tier. Any keyword answerable by a paragraph of text should be treated as an AI Overview risk category requiring citation optimization—not just ranking optimization. Any keyword that requires a user to take action—contact, purchase, compare specific pricing, or download—is structurally protected. AI cannot complete that action for them.
The core strategic shift is away from keyword volume as the primary selection criterion. Keyword intent is now the governing filter. A 500-monthly-search transactional keyword in your exact service category now delivers more expected return. A 5,000-monthly-search informational keyword where AI Overviews appear on 70% of searches no longer does. Volume metrics that ignore AI Overview presence are overstating the value of high-volume informational keywords. They are simultaneously understating the value of lower-volume transactional terms.
Tier 2 — Citation-Optimize Informational Keywords You Cannot Abandon
Some informational keywords represent unavoidable strategic targets. They drive brand awareness. They support topical authority that benefits your transactional pages. For those keywords, the correct response is not to stop targeting them. It is to add citation optimization as a mandatory layer on top of traditional ranking optimization. Structure your content in self-contained answer units of 120 to 180 words. Lead with the direct answer rather than a lengthy introduction. Implement FAQPage or HowTo structured data in JSON-LD format. Ensure the page has been updated within the past 90 days.
A contrarian finding from Semrush is worth foregrounding here. The common assumption is that AI Overviews always destroy clicks on the keywords they appear for. Semrush tracked the same keywords before and after an AI Overview first appeared on them. The zero-click rate actually decreased—from 33.75% to 31.53%. The interpretation is counterintuitive but data-supported. Many keywords where AI Overviews appear were already high-zero-click keywords before AI Overviews existed. The AI Overview arrived on queries that were already not producing clicks. Ask this about each flagged informational keyword: was it driving clicks before AI Overviews arrived? If yes, citation optimization can recover value that ranking alone no longer delivers. If no, the keyword was low-value before AI Overviews and remains low-value now.
Tier 3 — Reallocate Toward External Citation Building
The clearest strategic reallocation from the AI Overview data: reduce investment in broad, high-volume informational keywords where AI Overviews have established stable presence. Redirect that capacity toward external brand mention building. The reason is structural. Brands are 6.5 times more likely to be cited in AI responses through third-party sources than through their own websites. Every resource invested in creating broad informational content that AI Overviews are already answering yields declining returns. That same resource invested in authoritative third-party coverage builds the citation network that determines AI Overview citation probability across dozens of queries simultaneously.
For a B2B company in a technical service category, this might mean prioritizing a G2 profile with verified reviews. It might also mean contributing to industry publications or earning vertical trade media coverage. It means publishing fewer blog posts about questions AI can already answer. Semrush’s AI search trend analysis identifies cross-platform presence as a top predictor of AI citation probability. This holds across Google AI Overviews, ChatGPT, and Perplexity. None of those platforms cite brands they cannot find in multiple credible places. Building that multi-platform presence is keyword strategy for the AI era—even if it does not look like keyword strategy in the traditional sense.
Apply a Decision Framework to Every Keyword in Your List
Run this diagnostic on every keyword cluster in your current strategy. First, check whether the keyword currently triggers an AI Overview in a manual Google search. If it does not, traditional ranking optimization applies without major modification. If it does, check whether your brand is cited inside the AI Overview. If you are cited, the keyword remains strategically valuable and citation quality is the priority. If you are not cited, the keyword requires both a ranking strategy and a separate citation strategy.
Second, assess the keyword’s intent. Informational keywords with AI Overviews where you are not cited should be deprioritized against transactional keywords where AI Overview presence is low. Third, evaluate specificity. Hyper-specific technical queries requiring practitioner-level answers are more resistant to AI Overview commoditization. Concentrate resources in high-specificity queries within your domain of expertise. SeoClarity found that 97% of AI Overviews cite at least one source from the top 20 organic results. Ranking still matters. But it is now a necessary condition for AI citation—not a sufficient one. The winning keyword strategy identifies where ranking investment can also produce citation outcomes, then concentrates resources precisely at that intersection.