AI Overviews are changing the way search traffic is distributed. For many publishers, ecommerce brands, SaaS companies, and local businesses, the problem is no longer just ranking on page one. The problem is that even strong rankings can produce fewer clicks when Google answers part of the query directly in the results. Recent studies suggest this is especially true for informational and question-based searches, but the impact is spreading into commercial-intent SERPs as well.
If you want to recover traffic from AI Overviews, the most practical approach is not to chase a separate optimization theory. Google’s official position is clear: there is no special “AI Overview SEO” requirement, and the same SEO fundamentals still apply. At the same time, Google has introduced new reporting, new generative-search guidance, and richer source visibility inside AI-driven results. That means recovery depends on diagnosing click loss accurately, protecting vulnerable query sets, and increasing your chances of being surfaced as a trusted source inside AI experiences.
Understand what AI Overviews are really taking away
The first step is to define the loss correctly. In many cases, AI Overviews do not eliminate visibility; they reduce the number of clicks available to traditional organic listings. Google’s own messaging in 2026 is that AI Overviews are contributing to Search growth, with Sundar Pichai describing a seamless experience across AI Overviews and AI Mode. For site owners, however, search growth at the platform level does not automatically translate into website traffic growth.
Third-party data helps explain the gap. Ahrefs reported in July 2026 that the top organic result can lose about 58% of its clicks when an AI Overview appears above it, a much sharper decline than earlier estimates around 34.5%. Their conclusion was important: the main damage happens on queries where the AI Overview intercepts the click before users reach the classic results.
This distinction matters because it changes your recovery strategy. If your rankings are stable but clicks are falling, the issue may not be ranking loss at all. It may be SERP design and query treatment. In other words, recovering traffic from AI Overviews starts by accepting that you are fighting click compression, not only competition from other websites.
Use Search Console to separate visibility from traffic
Google’s June 2026 launch of dedicated Search Generative AI performance reports is one of the most useful developments for recovery work. These reports help site owners measure impressions in AI Overviews, AI Mode, and generative AI features in Discover. That gives you a more direct view into how often your content is being surfaced across generative search experiences.
You should watch clicks, impressions, and CTR together rather than relying on sessions alone in analytics tools. Google’s own documentation makes this practical: Search Console performance reports reveal what users saw and whether they clicked. If impressions are steady or rising while clicks and CTR fall, that is a strong sign that AI Overviews may be cannibalizing traffic rather than your site losing relevance.
This diagnostic step is essential because referral traffic often tells only part of the story. For example, research cited by Search Engine Land showed that AI referrals represented just 1.1% of visits to news publishers after AI chats, while also warning that last-click attribution misses the full user journey. Search Console gives a better SERP-level picture of what is happening before the visit is made or lost.
Follow Google’s guidance instead of inventing a separate playbook
Google has been unusually direct on this point. Its documentation says there are no additional requirements to appear in AI Overviews or AI Mode, and that standard SEO best practices remain relevant for AI features in Search. That means recovery should begin with the same fundamentals that have always supported search performance: crawlability, indexability, relevance, content quality, internal linking, and technical health.
In May 2026, Google published a new guide specifically for optimizing for generative AI features in Search. The guide did not introduce a radically different framework. Instead, it emphasized valuable, unique, non-commodity content and reinforced the importance of local, shopping, image, and video content. The takeaway is simple: the path into AI surfaces is still built on strong search fundamentals, but with more pressure to offer something distinctive.
This matters because many teams waste time searching for a hidden GEO or AEO trick. Google’s 2026 guidance suggests the opposite. If traffic is declining, the official answer is to improve usefulness, originality, and technical SEO rather than to chase speculative hacks. A recovery plan grounded in Google’s stated rules is far more durable than a playbook based on rumors.
Create content that gives Google a reason to cite you
If AI Overviews compress clicks, the best available recovery path is to increase the probability that your site becomes one of the sources surfaced inside the AI result. Google has said AI Overviews and AI Mode are meant to help people find relevant links and original content, not just summaries. In May 2026, Google also rolled out updates to help users discover original content and trusted sources more easily through article suggestions, direct links, and site previews.
That creates a clear content standard. Commodity articles that simply repeat what is already known are easier for AI systems to summarize without sending a click. By contrast, original research, firsthand experience, expert commentary, proprietary data, strong comparisons, and clearly structured insights give both users and Google a stronger reason to surface your page as a source.
Content formatting also matters. Pages that answer a question and then go beyond the answer are better positioned than pages that stop at the basic definition. If a query can be satisfied by two lines, your page may lose the click. If your page provides unique examples, updated data, actionable frameworks, visuals, local context, or decision support, it becomes more likely to earn inclusion and more compelling to click after the overview appears.
Protect the query types most exposed to click loss
Recent studies show that not all keywords are equally vulnerable. Ahrefs found that AI Overviews heavily target question searches, which makes answer-driven and informational content especially exposed. If a large share of your organic traffic comes from “what is,” “how to,” “why does,” or other question-led patterns, you should assume a higher risk of click suppression.
At the same time, commercial SERPs are no longer safe territory. Semrush reported in July 2026 that AI Overviews expanded 71% across commercial-intent results in a six-month period, based on more than 600,000 keywords. The same study found the strongest commercial-growth surge in finance, up 231% in six months, showing how quickly AI coverage can move into revenue-generating queries.
Your recovery strategy should therefore segment queries by intent and business value. Identify which informational pages are losing clicks because the answer is now visible in the SERP, and which commercial pages are being squeezed by a more crowded layout. Then prioritize updates for the keywords where click loss has the biggest downstream effect on leads, sales, subscriptions, or ad revenue.
Adapt to a more crowded SERP, not just a smarter one
AI Overviews are only one part of the traffic problem. Semrush also found that Google Ads and AI Overviews appeared together about twice as often as a year earlier. This means many SERPs now stack paid placements, AI-generated summaries, and classic organic results in a way that reduces the available click opportunity for every listing below.
That SERP crowding changes the kind of pages that win. If users are less likely to click a generic blue link, your snippet appeal, page relevance, and source authority have to be stronger. Titles and descriptions still matter because they help users decide whether your page offers more than the overview. So do visible signals of trust, freshness, expertise, and specificity.
It also means that recovering traffic may require defending multiple channels at once. Adthena’s February 2026 analysis found AI Overviews were not just a visibility issue for paid search but a revenue issue, with some industries seeing persistently lower CTR when AI Overviews were present. In sectors such as telecom and technology, the traffic pipeline itself can be drained. Brands should evaluate paid and organic performance together because the same SERP changes can hurt both.
Expand beyond text pages to increase search surface area
Google’s May 2026 guidance for generative search features emphasized not only unique written content but also local, shopping, image, and video content. That is a strong signal that recovery should not rely exclusively on traditional article formats. The broader your eligible content types, the more opportunities you have to appear in different search experiences and alongside different query intents.
For local businesses, this may mean improving business profiles, location pages, and local proof signals so that Google can connect your brand to nearby intent. For ecommerce, it may mean richer product content, comparison assets, merchant data quality, and strong image coverage. For publishers and B2B brands, it may mean creating supporting video, charts, original visuals, and expert explainers that offer value beyond plain text.
This is also a hedge against answer commoditization. When a short text explanation is easy for AI to summarize, supplementary assets can still attract attention and clicks. Images, tools, videos, calculators, interactive comparisons, and local details often remain more difficult to replace fully inside a summary box. They can increase both discoverability and click motivation.
Measure recovery by source inclusion and qualified clicks
Recovering traffic from AI Overviews does not always mean restoring the old volume of clicks to every query. In many cases, the better objective is to win more of the clicks that still happen and to earn visibility inside the AI result itself. Google has said AI Overviews create unique opportunities for more types of sites to appear, and its 2026 updates show closer integration with preferred sources and richer result treatment in AI Overviews and AI Mode.
That means success metrics should evolve. Beyond raw traffic, track whether your content is appearing in generative reports, whether CTR improves on affected query groups, and whether the visits you do receive are more qualified. A lower click volume can still support growth if those users arrive deeper in the funnel and convert at a higher rate.
You should also monitor source patterns over time. If certain article structures, content types, or author formats correlate with stronger presence in AI surfaces, replicate those patterns across your most valuable clusters. Recovery is not a one-time technical fix. It is an ongoing process of understanding which pages are being summarized away, which pages are being cited, and which pages still persuade users to click.
The companies that recover best from AI Overviews will be the ones that stop treating the change as a purely ranking problem. The evidence from Google, Ahrefs, Semrush, Adthena, and publisher studies points to a more nuanced reality: visibility can remain strong while traffic falls, and commercial queries are increasingly affected alongside informational ones. The winners will be the sites that measure the difference clearly, adapt content to the new click environment, and earn a place inside the AI-generated experience.
Google’s own guidance offers a useful anchor in a noisy market. There is no separate secret formula for AI Overviews. Standard SEO best practices still matter, but they now need to be paired with more distinctive content, broader search formats, and closer performance analysis in Search Console’s generative AI reports. If you want to recover traffic from AI Overviews, the most reliable strategy is to build pages that deserve citation, diagnose cannibalization with the right metrics, and optimize for clicks that remain valuable enough to win.
