How B2B buyers use AI to find suppliers.

Two thirds of senior decision-makers now use AI tools during procurement. The shortlist is being formed before any sales call happens. Most B2B companies are not on it.

That last sentence should concern you. Not because AI search is new or speculative, but because it's already the default for a growing majority of buyers and the companies being recommended are getting the calls while everyone else wonders why inbound dried up.

This is the upstream shift that everything else follows from. Before AI search visibility matters, before GEO matters, before content strategy matters, this is the fact that makes all of it relevant: B2B buyers have changed how they find suppliers, and most suppliers haven't noticed.

How are B2B buyers using AI to research suppliers?

Buyers are not typing brand names into ChatGPT. They are describing problems and asking for solutions.

A typical query looks like this: "Which construction consultancies in the Midlands specialise in pre-construction planning?" Or: "What are the best AI training providers for professional services firms?" Or: "We need a new ERP system for a 200-person manufacturing company. What should we consider?"

AI platforms respond with a shortlist. Usually three to five companies, sometimes with brief explanations of why each was recommended. The buyer reads the list, clicks through to a website or two, and makes contact.

The entire top-of-funnel process that used to take days of searching, reading reviews, and asking contacts now happens in a single conversation. The buyer walks away with a shortlist, a rough understanding of each provider's strengths, and often a recommendation on which to contact first.

This is not a niche behaviour. It's mainstream procurement. And it's reshaping which companies get considered and which get ignored at a speed that traditional marketing channels can't match. The signals AI platforms use to recommend businesses explain exactly how these recommendations are formed.

What percentage of decision-makers use AI in procurement?

66% of UK senior decision-makers now use ChatGPT, Claude, Copilot, or Perplexity when researching suppliers.

Source: Sopro, 2025

That's not a projection. It's survey data reflecting current behaviour.

85% of buyers aged 25 to 34 use AI for supplier research. Among those over 55, it's 23%.

Source: Sopro

The generational split makes the trajectory unmistakable. As younger professionals move into purchasing roles, AI-assisted procurement becomes the default, not the exception.

The conversion data reinforces why this matters commercially. AI visitors convert at 2 to 4.4 times the rate of traditional organic search visitors. Ahrefs reported that their own AI search visitors convert 23 times better than organic. The reason is intuitive: by the time a buyer asks an AI platform "Who should I work with for [service]?", they're past the browsing phase. They're making a decision. Being cited at that moment is worth more than any keyword ranking.

They're optimising for Google while ignoring the platforms that an increasing share of their buyers use first.

Which AI platforms do B2B buyers use most?

Six platforms matter. Each one draws from different sources and recommends businesses using different criteria.

ChatGPT is the most widely used. It draws heavily from Bing's index (87% correlation with Bing's top 10 results) and weights referring domains as its strongest signal. When a buyer asks ChatGPT for a supplier recommendation, the businesses with the strongest backlink profiles and Bing visibility tend to appear first.

Claude is growing rapidly in professional and technical contexts. It emphasises methodology transparency, primary sources, and well-attributed claims. Businesses that publish detailed process documentation and original research tend to perform well here.

Perplexity is the research-focused platform. It weights content freshness at approximately 40% of its ranking signal and leans heavily on Reddit, Wikipedia, and YouTube as sources. For buyers doing deep comparison research, Perplexity is often the first stop.

Google AI Overviews and Gemini appear above traditional search results, making them the first thing many buyers see. They favour multi-modal content (text plus images plus video) with a correlation of r=0.92 and require strong E-E-A-T signals. 96% of AI Overview citations come from sources with verified authority.

Microsoft Copilot matters for organisations in the Microsoft ecosystem. It requires Bing indexing as a hard prerequisite and decays citations by 97% over 91 days.

Grok (X's AI) cross-references web content with X posts. Most relevant for industries with active X presence.

Buyers using different platforms get different recommendations, which means businesses need visibility across the full set.

What does AI search tell buyers about your business?

When a buyer asks an AI platform about your business by name, the response reveals how the AI understands you. And that understanding shapes the buyer's perception before you've had any contact with them.

AI platforms construct their responses from everything available about your business across the web. Your website content. Directory listings. Review sites. LinkedIn. News mentions. Forum discussions. Industry publications. If these sources tell a consistent story, the AI presents your business clearly and confidently. If the sources conflict (outdated directory listings, old company descriptions, name variations across platforms), the AI hedges or gets it wrong.

Brand entity confusion is a real and underappreciated problem. A business might be described as "a Dubai-based Web3 agency" on Tracxn while positioning as "a UK AI readiness firm" on its own website. The AI encounters both descriptions and has to decide which to trust. The result is often uncertainty, which means the AI is less likely to recommend you with confidence.

AI platforms give 6.5 times more citation weight to what others say about your business than what you say about yourself. 82% of AI citations originate from earned media.

Source: Muck Rack

If your business has no presence on review platforms, industry directories, or third-party publications, the AI has very little external validation to work with.

How AI search changes the B2B buying journey

The traditional B2B buying journey looked like this: the buyer identifies a need, searches Google, clicks through several websites, reads case studies, asks colleagues for recommendations, compiles a shortlist, and contacts three to five providers.

The AI-influenced buying journey is compressed. The buyer identifies a need, asks an AI platform, receives a shortlist with explanations, visits one or two websites to confirm, and contacts the provider the AI recommended most strongly.

The middle of the funnel collapses. The comparison phase, the review-reading phase, the colleague-asking phase. AI does all of that in one response. The shortlist that used to take a week to build now takes 30 seconds.

This has three consequences for B2B companies:

First, the companies that appear in AI responses get disproportionate attention. The buyer doesn't see 20 options and pick 5. They see 3 to 5 options that the AI already filtered from thousands. Being in that filtered set is the new version of ranking on page one.

Second, the content that influences buying decisions has shifted. It's no longer just your website. It's every source the AI draws from: directories, reviews, forums, industry publications. Your off-site presence is now part of your sales funnel whether you manage it or not.

Third, speed of discovery has increased. Buyers who previously took weeks to build a shortlist now have one in minutes. The businesses that are visible to AI get contacted sooner. The ones that aren't visible get contacted later, or never.

What happens when your competitors are visible and you are not?

The maths is straightforward. If an AI platform recommends three businesses for a query and your competitor is one of them while you're absent, your competitor is being introduced to buyers you'll never know existed.

This isn't theoretical. Run a query like "best [your service] companies in [your region]" through ChatGPT, Claude, and Perplexity right now. See who appears. If your competitors are listed and you're not, those platforms are actively routing buyers to your competition.

The compounding effect is what makes this urgent. AI platforms learn from engagement patterns. When buyers click through to a recommended company's website, that engagement reinforces the AI's confidence in the recommendation. The companies that get recommended early get recommended more often. The gap between visible and invisible widens over time, not narrows.

Early SEO offers a useful parallel. The companies that understood Google in 2005 built positions that took competitors years to match. The same dynamic is playing out with AI search, except faster. AI adoption among buyers is moving quicker than Google adoption did. The window to build an early advantage is shorter, which means the cost of waiting is higher.

How do buyers verify AI recommendations?

AI recommendations are not taken at face value. Buyers use them as a starting point, then verify through a predictable sequence.

First, they visit the website. If the site is thin, outdated, or doesn't match what the AI described, the buyer moves to the next recommendation. The website needs to confirm and expand on what the AI told them.

Second, they check third-party sources. LinkedIn, Google reviews, Clutch, G2, industry directories. They want to see that real clients have worked with the business and had positive experiences. This is where testimonials, case studies, and review platform presence become critical.

Third, they may ask the AI a follow-up question. "Tell me more about [company name]." "What do clients say about [company]?" "Is [company] good for [specific requirement]?" The AI's response to these follow-ups depends on the depth and consistency of information available across the web.

The businesses that perform well through this verification process are the ones with consistent information across all platforms, published case studies with specific results, active review profiles, and content that demonstrates genuine expertise rather than generic marketing claims.

What should B2B companies do to be found by AI-using buyers?

The shift is happening now. Waiting for it to "mature" before acting means building authority in a space that's already been claimed by competitors who moved first.

• 01
  Understand your current positionTest your visibility across AI platforms with real buyer queries. If you're not appearing, you need to know. If competitors are appearing, you need to know who and why. 1DOT's audit covers this across six platforms, scored Strong, Mixed, or Weak across nine dimensions of commercial AI search visibility health on a live call.
• 02
  Fix the basicsAI crawler access in robots.txt. Schema markup on every key page. Consistent business information across directories, LinkedIn, and review platforms. Server-side rendered content so AI crawlers can read it. These are prerequisites, and where the page roles, schema, and entity structure are unclear, a Visibility Rebuild puts that structural foundation in place.
• 03
  Build content depthOne blog post and a few service pages is not enough for AI platforms to recognise you as an authority. Topical authority is the strongest predictor of AI citation and requires sustained depth. Articles over 2,900 words average 5.1 citations compared to 3.2 for articles under 800 words.
• 04
  Publish consistentlyAI platforms actively favour recent content. Pages updated in the last three months earn roughly twice as many citations as older material. A quarterly blog post doesn't cut it. Weekly at minimum, daily during a launch phase. Each new piece of content is a new surface for AI to cite.
• 05
  Build your off-site presenceWhat others say about you matters 6.5 times more than what you say about yourself. Industry directories, client reviews on Clutch and G2, LinkedIn thought leadership, Reddit and forum contributions. These are the sources AI platforms actually read when deciding whether to recommend you.
• 06
  Structure content for AI extractionSelf-contained passages that lead with a direct answer to the buyer's question. FAQ sections with schema markup. Statistics with source attribution every 150-200 words. Question-based headings. Content that an AI can extract and cite without needing to read the entire page.
• 07
  Monitor and maintainAI citation volatility averages 40-60% month to month. Only 20% of brands appear consistently when the same prompt is repeated. This is not a set-and-forget exercise, which is why ongoing Growth work sustains visibility across content, citations, and measurement in 90-day cycles.