Why the gap matters now
For many B2B buyers, the AI answer is now the first draft of the shortlist. G2's March 2026 survey of 1,076 B2B software decision-makers found that 51% started their research in an AI chatbot more often than in Google, up from 29% in G2's April 2025 survey of 1,169 buyers (G2, 2026; G2, 2025). Both surveys come from a software review platform and cover software buyers, so other sectors may be moving more slowly, but the speed of change is the point.
The shortlist matters because preference forms early. A 2024 vendor study by 6sense found buyers were nearly 70% through their purchase process before engaging sellers, and 81% had a preferred vendor at first contact (6sense, 2024). Forrester's 2024 research found that 86% of B2B purchases stall at some point and that 13 people are typically involved in a buying decision (Forrester, 2024). A large buying group researching separately, much of it through AI tools, is exactly the situation in which a vague or invisible supplier is dropped without anyone telling it.
The good news is that the usual causes are fixable. Most come down to three things: vague positioning, content locked behind forms, and thin proof.
Reason one: vague positioning and jargon
B2B websites often describe capability in terms only the team understands, such as "integrated solutions" or "end-to-end excellence". An AI tool has nothing specific to repeat, so it moves on to a competitor who says what they make and who they make it for.
Replace the abstract line with a plain one. For a manufacturer, that means the product, the materials, the sectors served and the lead times. For a professional firm, it means the service, the client type and the result.
Research on generative engines supports this. In a paper presented at the KDD conference in 2024, computer scientists tested ways of rewriting content and found that adding statistics, quotations and citations to sources raised visibility in AI-generated answers by up to around 40%, while stuffing pages with keywords did not help (Aggarwal et al., 2024). Specific, evidenced text gives a model something to repeat. A quick test for any key page:
- Could a stranger say what you make or do after reading the first two sentences?
- Does it name the sectors, applications or client types you serve?
- Does it include at least one number a buyer would care about, such as tolerances, capacity, lead time or turnaround?
- Would the page still make sense if your logo were removed?
Reason two: content locked behind forms
Gated white papers and guides are useful for lead capture, but AI tools generally cannot read what sits behind a form. If the only description of your expertise is gated, it cannot be cited.
Google's guidance for its AI features says a page must be indexed and eligible to show a snippet before it can appear as a supporting link, and that important content should be available in text form (Google Search Central, 2026). A PDF behind a form fails both tests, and chatbots that browse the web face the same barrier.
Buyers also want to serve themselves for longer. G2's 2025 survey found that nearly two-thirds of buyers preferred to engage vendor salespeople only in the later stages of the buying journey (G2, 2025). If your best explanation of the problem sits behind a sales conversation or a form, the buyer's research will happen somewhere else.
Keep a public summary on every gated asset, with the key findings and the questions it answers. Keep the full download behind the form if you need it for lead capture.
Reason three: thin proof
Claims without evidence are easy to ignore, for people and for AI tools. Case studies with named outcomes, certifications, customer quotes and clear project scopes give AI tools something specific to cite. Our manufacturing work shows how a sector page can carry this proof without becoming a brochure.
Third-party proof matters as much as your own pages. The Ahrefs analysis of 75,000 brands found that brand mentions across the web correlated with appearing in Google's AI Overviews at 0.664, far ahead of backlinks at 0.218, and that 26% of the brands studied were not mentioned in AI Overviews at all (Ahrefs, 2025). It is a correlation study from a tool vendor, but it suggests many brands are invisible because too few credible sources talk about them.
Human buyers respond to the same signals. The 2024 Edelman and LinkedIn study of nearly 3,500 managers found that 55% associated the best thought leadership with strong data and research, and 86% were more likely to invite firms producing high-quality thought leadership to bid (Edelman and LinkedIn, 2024). Evidence that persuades a buying committee is also evidence an AI tool can cite. Proof that works for both includes:
- Case studies with the sector, the problem, the scope and a measurable outcome.
- Certifications and accreditations, named and dated.
- Technical data sheets published as web pages, not only as PDFs.
- Named experts with their roles and experience.
- Listings and reviews on the directories your sector uses.
How to measure whether the fixes work
Pick 20 to 30 questions a qualified buyer would ask, and test them across the main AI tools each month. Record whether you are named, how you are described and which sources are cited. Our B2B AI visibility plan sets out how to structure this.
Watch traffic quality as well as volume. Clicks from search are falling on informational queries: Seer Interactive's analysis of 3,119 queries across 42 organisations found organic click-through rate on queries with an AI Overview fell from 1.76% to 0.61% between June 2024 and September 2025, but brands cited in the overview earned a 35% higher click-through rate than those not cited (Seer Interactive, 2025). In the UK, Ofcom reports that around 30% of searches now show an AI summary (Ofcom, 2025). Being the cited source is how you keep a share of the clicks that remain.
Finally, ask new enquiries how they found you, and add "an AI assistant" as an option on forms and in sales notes. Over a few quarters, that gives you a direct line between visibility work and pipeline.
How to gain the advantage
Once the three gaps are closed, the work is steady rather than dramatic. For example, an industrial coatings supplier might start by rewriting its five sector pages in plain terms, adding public summaries to its two gated guides and publishing three case studies with measured results, before turning to trade press. None of that requires a new website or a large content budget; it requires someone to own the list and work through it.
What good looks like after two or three quarters: you are named for most of your priority questions, the description matches your own positioning, and the cited sources include your pages and at least a few independent ones. The ongoing habits look like this:
- Publish clear answers to the questions your buyers ask, starting with the ten most common.
- Name the people behind the expertise, with their roles and track record.
- Keep key pages current and show the date they were last reviewed.
- Earn mentions in the trade press and industry directories your buyers read.
- Link related pages together so the relationship between topics is clear.
What to ask a supplier helping you with this
If you bring in outside help, ask how they will work, not only what they will deliver. A supplier who cannot answer these questions clearly is likely selling activity rather than outcomes.
Our GEO service and our page on AI search for manufacturers explain how we approach this work.
- Which questions will you track, and how did you choose them?
- How many times do you run each question, and on which platforms?
- How will you separate mentions from accurate descriptions?
- Which third-party sources do you expect to influence, and how?
- How will results connect to enquiries and pipeline?
Put this into practice
Book a discovery call and we will review your channels and results first, then agree what to test.
Book a discovery callQuestions and answers
Should we remove forms from our content?
Not all of them. Keep the form for the full asset, but publish a public summary so the substance can be found, read and cited.
Which pages should we fix first?
Start with the pages that a qualified buyer would land on: the core service pages, the sector pages and the two or three guides that answer the most common questions.
Does AI visibility matter if most of our sales come from referrals?
Yes. Referred buyers still check suppliers out, and many now ask an AI tool what it knows about a firm before a first meeting. If the answer is vague or wrong, it weakens the referral. Visibility also matters for members of the buying group who were not part of the original recommendation.
Do we need to publish lots of new content?
Usually not at first. Most B2B brands gain more by making existing pages clearer, publishing summaries of gated material and adding proof than by producing more articles. Add new content where buyers ask questions you do not yet answer.
Do we need special files such as llms.txt to appear in AI answers?
Google says no new machine-readable files, AI text files or special markup are needed to appear in its AI features (Google Search Central, 2026). Focus on indexable pages, clear text and structured data that matches what is on the page. Other AI platforms may treat such files differently, but they are no substitute for clear content.
References
- G2, "The Answer Economy: How AI Search Is Rewiring B2B Software Buying", 2026
- G2, "New G2 Research: How AI Is Redefining the Buyer Journey in 2025", 2025
- 6sense, "6sense Launches 2024 Buyer Experience Report Unveiling Global B2B Buyer Trends" (press release via Silicon UK), 2024
- Forrester, "To Master B2B Buying Mayhem, Providers Must Prioritize Buyers' Needs" (The State of Business Buying, 2024), 2024
- Aggarwal et al., "GEO: Generative Engine Optimization" (KDD 2024), 2024
- Google Search Central, "AI features and your website", 2026
- Ahrefs, "An Analysis of AI Overview Brand Visibility Factors (75K Brands Studied)", 2025
- Edelman and LinkedIn, "2024 B2B Thought Leadership Impact Report" (summary by Curzon PR), 2024
- Seer Interactive, "AIO Impact on Google CTR: September 2025 Update", 2025
- Ofcom, "From apps to AI search: how the UK goes online in 2025", 2025
This article is adapted from ContxtAI's article on why B2B brands struggle with LLM visibility, rewritten for Insight Labs readers.


