The End of Google Search and the Future of B2B Strategy

Google announced at I/O yesterday that the ten blue links at the top of the search page are over. Google Search will now route users into AI-powered experiences and conversational interfaces rather than a ranked list of websites, with background information agents that can track market movements and changes continuously on a user’s behalf. The company describes it as the biggest change to Search in 25 years. As a B2B agency powered by the Google ecosystem, this one lands close to home. The channel that once drove buyers to your website is now an answer engine, like all the others. And most current B2B AI strategy is pointed at the wrong part of the buying journey regardless.

The 6sense 2025 Buyer Experience Report, drawing on nearly 4,000 buyers across North America, EMEA and APAC, found that in 95% of deals, the eventual winner was already on the buyer’s shortlist. The same study found 97% of buyers have prior personal experience with at least one vendor on that shortlist, and the average buyer has been through 8 to 9 prior purchase journeys in the same solution category. The shortlist is not being assembled by AI sweeping the market. It’s being assembled from memory. Memory built years before the brief lands, in rooms buyers walked through, on booths they spent time at, in conversations they had standing next to a product they could touch. That changes what the actionable strategy looks like. “Be visible in AI answers” only matters if AI is generating new names, but for most established B2B categories, it isn’t. The strategic priority is to be in the buyer’s experience set before they even have a problem to solve. This is the work brand experience has always done, on the exhibition floor, the senior leadership event, the immersive showcase. These are how the experience set gets built two or three years before there’s a brief to write.

There’s an exception worth being precise about. In fast moving software categories, where buyers have less accumulated vendor experience, AI is genuinely introducing new names. G2’s March 2026 survey of 1,076 B2B software buyers found 69% chose a different vendor than initially planned based on AI chatbot guidance, and one in three purchased from a vendor they’d never heard of before. This framework distinguishes three tracks: for software, AI discovery is real. For physical goods and services such as physical brand experiences, prior relationships still dominate. The shortlist is built from memory, and memory is built from what people have seen, touched, and stood inside.

The second misread sits around when buyers are actually using AI. The same 6sense data shows LLM usage peaks in the middle of the buying journey, not at the start. Buyers are not going to ChatGPT to discover their category, because they already know it. They’re using AI to compare vendors already on their shortlist, draft requirements, simulate cost of ownership, generate early-stage RFPs, etc. AI is a great synthesiser, but not so much a discoverer. The implication is operational: content optimised for AI citation matters most at the comparison stage, when buyers are validating choices, not when they are building awareness.

The third misread, and potentially the most consequential, is the reason buyers are engaging vendors earlier than they used to. The standard story is that AI tools have compressed the research phase, so buyers reach sellers sooner. The data points elsewhere. 6sense found 58% of buyers engaged with vendor representatives earlier specifically because they needed to evaluate how AI was implemented in the solutions they were considering. Capabilities, pricing, security, implementation timelines, all for the AI components of the product. Vendor websites do not answer those questions, and so buyers are much more likely to enquire directly.

The brands that win the next few years are the ones who can articulate the AI inside their product and services clearly, defensibly, and verifiably. Not the ones who optimise content for being cited by AI.

The trust dynamic that comes up in many AI related conversations is very real, but most B2B strategy treats it as a brand problem, rather than an accuracy problem. G2 found 64% of B2B buyers encounter inaccurate AI chatbot recommendations often or very often. Gartner found only 33% of buyers rank supplier-provided generative AI as trustworthy. Forrester’s State of Business Buying 2026 puts it as: AI answer engines often deliver incomplete or unreliable information, and buyers compensate by seeking validation from trusted sources. That’s a rational correction for a known failure, not a sentimental preference for human contact. Buyers use peer reviews, analyst reports, and practitioner voices to verify what AI has told them. Most B2B brand strategy still treats third-party validation as a nice-to-have. It isn’t. It’s the correction infrastructure the AI-mediated buying journey now depends on.

Forrester’s 2026 data, drawn from nearly 18,000 global buyers, found the typical B2B decision now involves on average 13 internal stakeholders and nine external influencers. Twenty-two people per decision, rising for more strategic purchases. Procurement is now a decision-maker in 53% of cycles, engaging from the start, evaluating features and functions rather than just price, and interacting with sales reps more frequently than other personas. Rather than a procurement check at the end, It’s a parallel evaluation track with its own criteria and its own information needs. Most B2B content is still aimed at champions which leaves the procurement track mostly unsupported. That gap has existed for a while, but AI-mediated research is making it more visible, because procurement evaluation is the part of the journey that benefits most from structured, comparable, machine-readable information. Procurement is the persona AI is actually best suited to serve.

“Be legible to AI” is doing more work in current strategy decks than the data supports. Loganix’s March 2026 analysis of 680 million AI citations found only 11% of domains are cited by both ChatGPT and Perplexity. A separate study of 15,000 queries found only 12% of cited sources match across ChatGPT, Perplexity, and Google AI. ChatGPT leans on directory listings and authoritative domains. Perplexity draws heavily from Reddit and review platforms. Microsoft Copilot, deployed inside enterprise Microsoft 365 environments, leans on LinkedIn for B2B queries.

Google now needs to be understood as a distinct surface in its own right, and a different one than it was a week ago. AI Overviews are already used by 2.5 billion monthly users; AI Mode, Google’s conversational search, has 1 billion. Traditional organic links are increasingly an afterthought in those results. The decade of B2B SEO investment built around ranking for category terms is structurally less effective than it was, and the I/O announcements accelerate that. Google’s visibility is now an AI Overviews and Gemini citation problem, not a keyword ranking problem.

The more forward-looking signal from I/O is the information agents. Google is introducing background tools that can track market movements, supplier developments, and sector changes continuously on a buyer’s behalf, synthesising and alerting rather than just surfacing links. Albeit in early form, they’re pointing at something that will matter within two to three years: the B2B buying cycle trigger becoming a rolling, ambient research process rather than a discrete moment. Brands absent from that continuous background layer won’t reach the moment of active comparison at all.

That is four or five different strategies, not one. Which platform your buyers are using depends heavily on category and seniority. Treating AI visibility as a single workstream produces generic, low-conversion presence everywhere instead of concentrated presence on the platforms that actually matter.

Most companies have not caught up yet, but they will, and faster than you’d expect. AI-referred traffic converts at 14.2% versus Google organic’s 2.8%, but only 22% of marketers currently track AI visibility and fewer than 26% plan to develop content specifically for it. That gap will close. The development and implementation timeline is faster than most strategy cycles account for.

So, taking this all into account, the strategic priorities therefore, in order, are these:

1. Be in the buyer’s experience set before the cycle starts. Memory is built in rooms, not in retargeting.
   
2. Articulate the AI inside your product or service more clearly than your competitors can.
3. Treat third-party evidence as primary infrastructure, not as an earned bonus. Support the procurement track explicitly.
4. Build platform-specific AI visibility while the cost of entry is still low. 

The brands that will be strong in B2B in 2030 are the ones that build this stack now. The ones spending their AI strategy budget on showing up in ChatGPT for their category will find that they showed up, and it didn’t matter, because the shortlist had already closed.