Something paradoxical has started appearing in SEO analytics. Impressions in Google Search Console are rising. Clicks are falling. The two curves, which once moved in the same direction, are pulling apart — and AI Overview is at the center of that divergence.
A new dynamic in search performance
Over recent months, SEO analysts have increasingly observed an apparent contradiction: content receives more impressions than ever, yet drives less traffic. The most immediate explanation points to AI Overview — the AI-generated answers Google surfaces at the top of search results, directly in the SERP, before any organic link.
The widely circulated interpretation has been roughly this: AI Overview increases general information visibility, content gets cited more often, and direct traffic declines because the answer is already present on the results page. In other words, many have framed this as a paradigm shift in how SEO metrics are interpreted — not a problem, but a transition.
That reading is too comfortable. It risks normalising a structural erosion of visibility that will compound over time.
Impressions without real visibility
Marketing has long distinguished between technical impressions and impressions that are actually seen. If content appears theoretically in search results but the user never reaches it, that impression loses most of its value.
A simple analogy clarifies this. Imagine installing a large advertising billboard on a busy road. The investment makes sense because many people pass in front of it and can see it. Now imagine that every day a truck parks directly in front of the billboard, covering it almost completely.
Technically, the billboard still exists. Technically, it generates impressions. But the message is no longer seen.
AI Overview risks producing exactly this effect. When Google surfaces a complete AI-generated answer at the top of the page, the user finds the response immediately, the need to scroll through organic results diminishes, and many users never reach the links positioned below. The sites cited in the AI Overview continue to exist in the results — but their effective visibility may fall sharply.
The problem is not just the collapse of CTR. It is the reduction of real content exposure — and with it, the erosion of brand lift, recall, and conversion influence.
The risk to brand lift and conversion lift
For many organisations, SEO is not only a question of direct traffic. Presence in search results contributes to brand awareness, brand lift, and conversion lift through repeated exposure. When a user sees a brand name appear multiple times across search results, the probability of recognition, recall, and eventual selection increases — even if that user never clicks.
If AI Overview occupies the primary space in the SERP and users stop scrolling, this accumulative exposure process weakens. In the medium term, organisations may begin to observe lower organic traffic, reduced brand recognition, and a measurable impact on indirect conversions — without an obvious explanation visible in standard analytics.
The damage is slow, invisible, and structural. Which makes it more dangerous than a sudden traffic drop.
Google's response: advertising inside AI answers
Google is fully aware of these dynamics. One of the announced solutions involves the introduction of advertising within AI Overview, planned to roll out progressively through 2026. The commercial logic is clear: if organic traffic migrates into the AI layer, monetisation must follow it there.
But this move opens new questions that remain unanswered. Will users accept advertising embedded in AI-generated answers, or will it undermine the perceived neutrality of those responses? How much organic traffic will continue flowing to websites as the AI layer expands? And what happens to brands that are not cited in AI Overview at all — neither organically nor commercially?
Citation becomes the new strategic variable
In this new landscape, the concept of visibility is shifting fundamentally. Appearing in organic results is no longer sufficient. What becomes increasingly critical is being cited directly by AI — included by name, described accurately, and recommended in the AI-generated answer that captures most of the user's attention.
This transforms the nature of SEO work. The challenge is no longer only to earn a position in the SERP, but to build content that AI systems understand, trust, and select as a source. It means structuring information in machine-readable formats, establishing consistent topical authority, and ensuring that what AI says about your brand matches what you want to be known for.
This is precisely the gap that the Mismatch Index measures: the distance between how a brand is positioned in its own communications and how AI systems actually describe it. For most brands, that gap is significant — and invisible until it is measured.
Citation Rate as a structural response
The Citation Rate model emerged from this precise problem. In a context where impressions are decoupling from clicks, and where the AI layer is capturing an increasing share of user attention, the relevant metric is no longer position — it is citation probability.
How likely is it that an AI model will include your brand in its response, for the queries that matter most to your business? What is your Citability Score across the semantic field your brand operates in? These are the questions that will determine organic visibility in the AI era — not keyword rankings, and not impression counts that no longer translate into reach.
The organisations that begin measuring and optimising citation probability now are building an asset that compounds. The ones that continue optimising for a metric that is structurally decoupling from real exposure will find themselves with strong analytics and declining relevance.
Understanding the AI Overview effect today is not about adapting to a new trend. It is about recognising that the infrastructure of search visibility has already changed — and acting accordingly.