AI search is not destroying B2B marketing; it is redistributing traffic to reveal which content actually converts. When impressions fall by two-thirds but clicks remain flat, the real problem is not the search engine,it is the content strategy. New data from real-world B2B sites shows that AI answer engines like Perplexity and Google's AI Mode are filtering out low-value awareness traffic while preserving high-intent buyer queries, forcing companies to rethink what they publish and how they measure success.

What Happened to B2B Search Traffic When AI Arrived?

In July 2026, one B2B technology company pulled its Google Search Console data and found impressions had fallen by roughly 68 percent over a 28-day window. The reaction was panic. Then the company looked closer at the numbers. Clicks stayed flat at around 635 to 638. Click-through rate nearly tripled.

The pattern became clear when the company broke down which pages lost impressions. The vanished traffic came almost entirely from blog posts about other companies' AI models, funding announcements, and benchmark comparisons. These are the kinds of queries where readers want a news summary, not a vendor's perspective. An AI answer box can settle these questions in two sentences, so the reader never needs to click through. The company was never going to win those clicks anyway.

There is an important caveat: Google confirmed a logging error that inflated Search Console impressions from May 2025 through April 2026. Part of any year-over-year impression drop businesses are seeing includes a reporting correction, not lost demand. Before rewriting strategy based on impressions alone, companies should verify which months their data covers.

Why Are Clicks Actually Getting Better, Not Worse?

The real insight is that remaining clicks are higher quality. When people do click from search results that include AI summaries, they spend more time on the destination site, according to Google's own analysis. This matches what the B2B company observed: fewer visits, but visits from people closer to a buying decision.

Independent research from Pew Research Center tracked 900 U.S. adults in March 2025 and found that users who saw an AI summary clicked a traditional search result in only 8 percent of visits, compared to 15 percent when no summary appeared. They clicked links inside the summary itself in just 1 percent of visits. However, this data represents general-public search behavior across all query types. A procurement manager comparing three data integration vendors behaves very differently from someone checking a sports score. The queries where B2B buyers are close to a decision are exactly the ones an AI summary cannot resolve, because buyers need pricing, case studies, named contacts, and proof the company exists.

What Does Google Say About Optimizing for AI Search?

Google's official guidance is straightforward: there are no special requirements to appear in AI Overviews or AI Mode, and no unique optimizations are necessary. All existing SEO fundamentals continue to work. What changes is which pages deserve investment and which metrics matter most.

The terminology itself has shifted. Google's Search Generative Experience (SGE) was a Search Labs experiment. On May 14, 2024, Google announced AI Overviews, which roll out as a summary box above traditional results. A year later, on May 20, 2025, Google added AI Mode, a separate conversational interface that behaves much like ChatGPT or Perplexity. These are now two distinct surfaces, and any B2B marketing strategy still built around the old SGE terminology is working from an outdated map.

How to Adapt Your B2B Content Strategy for AI Search

• Audit Your Traffic Mix: Split your organic clicks into pages that could lead to a sale and pages that cannot. If most traffic goes to comparison posts, news pieces, or model explainers, AI search will make that imbalance visible faster. The fix is not a new optimization technique; it is changing what you publish.
• Shift from Volume to Intent: Stop chasing impressions on low-commercial-intent queries. Focus on the decision-stage content that AI summaries cannot answer: pricing, case studies, implementation timelines, and customer testimonials. These are the pages that convert.
• Monitor Click Quality, Not Impressions: Impressions are now a poor proxy for B2B SEO health. They include a year of inflated reporting and are most exposed to answer boxes on non-commercial queries. Track clicks and what those clicks do next: time on page, pages per session, and conversion rate.
• Publish Decision Content Over News Content: One B2B company found that 68 percent of its search clicks went to posts about other companies' AI models, while service pages earned only 23 clicks in 28 days. The company shifted to fewer news pieces and more decision-stage content, the kind a buyer reads when weighing how to choose a vendor rather than which model topped a leaderboard.

Are AI Crawlers Now More Important Than Google?

The data suggests a quiet shift in traffic sources. On September 1, 2026, one B2B site's Cloudflare dashboard showed OpenAI's crawler making 544 requests against 135 from Googlebot over the same period. By mid-July, the site's Google Analytics was already reporting an "AI Assistant" channel with 109 sessions. For this company, the AI side is now the busier reader.

This does not mean Google is irrelevant. It means that for B2B companies, the distribution of traffic is changing. Content that appears in AI answer engines reaches a different audience than content that ranks in traditional search results. Both matter, but they serve different purposes in the buyer journey.

The uncomfortable truth is that most B2B SEO problems are not caused by AI search. They are caused by publishing the wrong content. AI search simply makes that mistake visible faster. Companies that built their organic strategy around awareness-stage content, model comparisons, and news coverage are now seeing those pages lose impressions. Companies that focused on decision-stage content, buyer education, and conversion-oriented pages are seeing their clicks hold up or improve. The shift is not a crisis; it is a correction.