The classic B2B funnel assumed a simple deal: you publish content, Google ranks it, a buyer clicks, and you capture them. AI Overviews break the middle of that sentence. Google now answers many research queries directly on the results page, and that is why AI Overviews are rewriting the marketing funnel from the top down. Buyers still research, but they increasingly do it without ever visiting your site. If your reporting still treats organic sessions as the main proof of top-of-funnel health, you are measuring a shrinking slice of what is actually happening.
What the Data Actually Says
Be careful with headline numbers here, because studies disagree and methodologies differ. Still, the direction is consistent. Field studies reported by outlets like Search Engine Journal and PPC Land have found that AI Overviews cut organic clicks on affected queries by roughly 38 to 40 percent. SparkToro data discussed in industry coverage suggested that around two thirds of US Google searches in early 2026 ended without a click to the open web. Your own numbers will vary by query type: informational and definitional searches are hit hardest, while high-intent commercial searches are affected less.
The practical takeaway is not “SEO is dead.” It is that impressions can rise while clicks fall, and that a flat traffic line may hide a real change in how buyers find you.
How the Funnel Is Compressing
Three shifts matter for B2B teams.
First, awareness is happening inside the answer. A buyer asks “what is a customer data platform” or “best cold email tools” and gets a synthesized summary with a handful of cited sources. If you are cited, you get brand exposure without the click. If you are not, you are invisible at the moment the shortlist forms.
Second, the shortlist forms earlier and out of sight. Buyers arrive at vendor sites already holding opinions, which is why direct and branded traffic often quietly grows while non-branded organic shrinks. This is the same dark-social effect that has been eroding attribution for years, now accelerated by AI search and assistants like ChatGPT, Perplexity and Gemini.
Third, the click that does arrive is more qualified. Reports from several teams suggest that visitors coming through AI citations convert at a higher rate than generic organic traffic, though sample sizes are small and you should check your own analytics before believing it.
Adjustment 1: Write to Be Cited, Not Just Ranked
AI systems favor content that gives a clear, specific, attributable answer. That means direct definitions near the top of a page, concrete numbers with sources, and original data or firsthand testing that cannot be summarized from ten other pages. Generic listicles built from vendor homepages are exactly what gets compressed away. If you are unsure whether your pages are being picked up, we walked through how to check in our guide to whether your content is getting cited by ChatGPT.
Adjustment 2: Measure Visibility, Not Just Visits
Add a second layer of reporting that tracks how often your brand appears in AI answers. Tools like Profound, Peec AI and Otterly were built for this, and we compared them in our AI search visibility tools roundup. Pair them with Google Search Console impressions, branded search volume and a simple “how did you hear about us” field on demo forms. None of these is precise on its own. Together they give you a usable picture.
Adjustment 3: Rebalance Toward Bottom-Funnel and Owned Channels
If top-of-funnel clicks are less reliable, the content that converts existing demand becomes relatively more valuable. Comparison pages, pricing explainers, integration guides and case studies are harder for AI to replace because buyers need the details and proof. At the same time, invest in channels you control: email newsletters, communities, LinkedIn and webinars. These give you a direct line that does not depend on a search engine deciding to send you a visitor.
Adjustment 4: Rethink Which Keywords Deserve Content
Not every informational query is worth chasing anymore. If the entire value of a post is a definition that an AI summary will fully satisfy, the traffic may not justify the effort. Prioritize topics where you can add something a summary cannot: experience, benchmarks, opinionated frameworks, or tools and templates people actually want to use. Review your existing library too. Pages that lost clicks but still earn impressions are candidates for a refresh with fresher data and clearer answers.
What Not to Do
Do not panic-rewrite your whole site, and do not buy into any consultant promising a guaranteed “AI Overview hack.” Google’s systems change often, and no one outside the company has a stable playbook. Do not abandon classic SEO either: the pages AI Overviews cite are still largely pages that already rank well. And do not stop publishing for humans, because the brand recognition built through real expertise is what makes buyers type your name into a search box later.
A Simple 30-Day Plan
In week one, pull Search Console data and sort queries by impression change versus click change to find where AI Overviews are biting. In week two, run your top 20 commercial and category queries through Google, ChatGPT and Perplexity and note who gets cited. In week three, refresh the five most important pages with clearer answers, original data and updated sources. In week four, add a visibility metric to your monthly report next to organic sessions. Our post comparing Perplexity, ChatGPT and Gemini as research assistants is a helpful companion for understanding how buyers might be asking those questions.
The funnel is not disappearing. It is moving, and a lot of the movement is happening where your analytics cannot see it. The teams that adapt fastest will be the ones that stop equating traffic with influence.