Here’s an exercise worth doing — but only if you do it correctly. Go to ChatGPT, Perplexity, or Gemini and type the kind of category-level prompt a buyer who has never heard of you would type: “What are the best platforms for [your category]?” or “Which companies are considered leaders in [your space]?” Read the answer carefully. Note every brand that appears. Note how yours is characterized, if it appears at all. Note what the AI says about your competitors.
The catch — and it’s an important one — is that you need to run this test from a genuinely cold session. If you’ve ever discussed your company in that AI session before, if you’re logged in to an account with a history of brand-related conversations, or if the AI has learned your preferences and interests over time, you may get a distorted result. Your personalized AI experience is not your buyer’s AI experience.
To get a genuinely neutral read, you need one of three approaches: open a fresh incognito browser window where you’re not logged in to any AI service; ask a colleague whose account has no connection to your brand to run the same prompts; or, best of all, ask someone entirely outside your organization — a friend, an advisor, someone who genuinely has no prior association with your company — to run the prompts and report back what they see.
What you’re testing in this exercise is not whether AI knows you exist. It almost certainly does. What you’re testing is whether a buyer who has never encountered your brand before would have it surfaced in their research — and how it’s characterized relative to the competitors who appear. That is a fundamentally different question from “does AI know about us?” and it requires a genuinely uncontaminated session to answer.
The reason this audit matters so much is the same reason we’ve spent the first two posts in this series establishing the stakes: a growing share of B2B buyer research now begins in AI assistants before it ever touches a vendor’s website. The buyer who types “what are the leading vendors for [your category]” into ChatGPT isn’t doing idle research. They’re forming a consideration set. They’re deciding which companies deserve a closer look, which ones are worth booking a demo with, and which ones they can safely ignore.
If your brand isn’t in that initial AI answer — or if it’s in the answer but characterized less favorably than competitors you know you outperform — you’ve lost a deal before anyone on your team knew the deal existed. The prospect has moved on, already primed to evaluate your competitors more favorably, before your sales team has had any opportunity to intervene.
This is the invisible phase of the buying journey, and it’s operating at scale right now in virtually every B2B category. Understanding where your brand stands in it isn’t optional. It’s the baseline for any serious AI-era marketing strategy.
AI systems don’t pull vendor recommendations from a curated database or a sponsored list. They synthesize recommendations from the vast landscape of digital content they were trained on, weighted heavily by the authority and independence of the sources they’ve ingested. And in systems built on Retrieval Augmented Generation, they also pull from real-time web content at the moment of the query.
This means your visibility in AI category recommendations is a direct and fairly accurate reflection of your visibility in the sources AI trusts most: authoritative press coverage in respected trade publications, reviews on structured peer validation platforms, analyst mentions in Gartner or Forrester reports, citations in industry research, and the overall volume and quality of independent, authoritative content that discusses your brand substantively.
Critically, AI is not looking at your homepage, your marketing materials, or your own blog posts when it forms these recommendations. It’s looking at what independent sources say about you. This is the core insight behind why search presence is one of the five pillars of the TRUST framework: the signals that drive Google rankings — authoritative backlinks, domain authority, EEAT signals, earned editorial coverage — are largely the same signals that feed AI recommendation systems. Building genuine search authority and building AI visibility are, in most respects, the same investment.
AI isn’t applying a complex proprietary algorithm that you need to reverse-engineer. It’s applying the same basic credibility logic that thoughtful human buyers apply: independent, authoritative validation carries more weight than self-description. The company that has been written about substantively in respected publications, reviewed positively by verified buyers, and recognized by independent analysts is a more credible recommendation than the company whose primary presence is its own marketing content. AI has learned this from human-generated content because humans have always known it.
When you run those category queries and read the results carefully, you’re looking at a snapshot of your competitive landscape as AI currently understands it. The companies that appear most prominently and most favorably in AI category recommendations tend to share a specific set of characteristics, and it’s worth understanding what they are.
Depth of media coverage in authoritative publications is typically the most reliable predictor of AI recommendation visibility. Companies that have been covered substantively and repeatedly in respected trade outlets, technology publications, and business media have built a permanent, densely linked record that AI draws on when answering category queries. A single feature story in a high-authority publication is worth more to AI visibility than dozens of press release pickups on low-authority syndication sites.
Active, well-maintained profiles on relevant review platforms are the second major predictor. For B2B software and services, platforms like G2, Capterra, TrustRadius, and the vertical-specific equivalents carry significant weight in AI category synthesis. Understanding what makes buyers trust online reviews — volume, recency, specificity, response rate — is directly relevant to AI visibility because AI is using those same signals to assess the quality and authenticity of your review presence.
Analyst recognition is the third major factor. Even a brief mention in a Gartner Magic Quadrant or a Forrester Wave sends a powerful categorical signal to AI: this brand belongs in the consideration set for this market segment. Analyst firms are among the most authoritative sources AI knows about, and their coverage carries correspondingly significant weight.
Thought leadership that gets cited — original research, substantive expert perspectives, proprietary data that others reference — rounds out the picture. Brands that have positioned themselves as knowledge sources in their category tend to appear more prominently in AI recommendations because they’ve built exactly the kind of cited, authoritative presence that AI retrieval systems are designed to surface.
One of the most clarifying — and sometimes frustrating — things the AI category audit reveals is that AI recommendation visibility doesn’t correlate perfectly with product quality. There are brands that appear prominently in AI answers for their categories that you may know, from direct competitive experience, aren’t actually the best products in the space. And there are genuinely excellent products that don’t appear at all.
This gap exists because AI is recommending based on external validation, not direct product evaluation. AI has never used your product. It has never run a head-to-head comparison. It’s synthesizing the credibility signals in its source material, and those signals reflect brand-building effort as much as they reflect product quality. The brand with five years of consistent earned media investment and a rich review platform presence will typically appear in AI recommendations over the brand with a better product but a thinner external validation profile.
This isn’t a bug in the AI system. It’s a feature — the same feature that has always made independent validation valuable in B2B markets. The trust deficit that most businesses carry without realizing it — the gap between how trustworthy they believe they are and how trustworthy buyers actually perceive them — has a direct AI-era expression: the gap between how visible brands believe they are in AI recommendations and how visible they actually are. Both gaps exist for the same underlying reason: not enough independent, authoritative third parties have weighed in on the brand’s behalf.
The practical implication is important. If your brand is absent or poorly represented in the AI category audit results, the solution isn’t to optimize for AI directly. There are no AI visibility tricks or shortcuts that work the way keyword stuffing briefly worked for early SEO. The solution is to build the external validation foundation that AI draws on — which is the same foundation that builds genuine, durable brand credibility with human buyers.
Beyond simply noting whether your brand appears, the category audit is worth doing more carefully. Read the AI’s characterization of each brand it mentions, including yours. How does it describe what you do? What does it say your strengths are? What limitations or considerations does it mention? How does your characterization compare to how your competitors are described?
Common issues that surface in this more careful reading: category misclassification (AI describing you as competing in a slightly different segment than you actually occupy), outdated positioning (AI describing a product version or company positioning from two or three years ago), competitor conflation (AI blurring the lines between you and a competitor in ways that create confusion), and capability gaps (AI underrepresenting what your current product actually does because the most recent coverage predates your most significant developments).
Each of these issues is fixable, but each requires a different type of investment. Category misclassification is fixed by generating more current, consistent content across your owned and earned channels that clearly and repeatedly signals your actual category. Outdated positioning is fixed by a sustained fresh media presence that reflects your current state. Competitor conflation is fixed by sharpening your differentiation language and getting it into authoritative third-party sources. Capability gaps are fixed by ensuring your most significant product developments get substantive coverage, not just press release pickups.
The audit also tells you something valuable about your competitive set: which brands AI considers your primary competitors, regardless of how you define your competitive landscape. If AI is consistently grouping you with companies you don’t consider direct competitors, that’s a signal about how your brand is being categorized in the source material. If AI is consistently elevating competitors you know you outperform technically, that’s a signal about the external validation gap between you.
A one-time audit gives you a baseline. A quarterly cadence gives you a directional read on whether your investments in external validation are moving the needle. The queries to run regularly should cover three categories: generic category queries (“what are the best tools for [your category]”), problem-framed queries (“how do companies solve [the problem you solve]”), and competitive queries (“how does [your company] compare to [your key competitors]”).
Run these across at least three AI systems — ChatGPT, Perplexity, and Gemini are the minimum set worth checking — because the results can vary meaningfully between systems depending on their training data, their real-time retrieval sources, and the way they’ve been fine-tuned for different use cases. A brand that appears prominently in ChatGPT responses may be weaker in Perplexity, which relies heavily on real-time web retrieval and tends to surface more recent content.
Document the results systematically. Screenshot the answers. Note which competitors appear and how they’re characterized. Track changes quarter over quarter. This data becomes the most direct measure of whether your external validation investment is producing AI visibility improvement.
You can’t directly optimize for AI recommendations the way you’d optimize a Google page for a specific keyword. What you can do is build the kind of authoritative, third-party-validated digital presence that AI systems are specifically designed to recognize and trust. The investment required is the same investment that has always built genuine brand authority, but with a clearer understanding of why each component matters.
Earning coverage in publications your buyers read isn’t just a PR tactic. It’s a direct investment in AI recommendation visibility, because that coverage becomes permanent source material that AI draws on every time it answers a category query. Maintaining an active, well-managed presence on the review platforms that matter in your space isn’t just good for direct buyer research. It’s a continuous feed of peer-validated social proof that AI incorporates into its characterization of your brand. Publishing original research and insights that earn citations isn’t just thought leadership for its own sake. It’s the creation of authoritative, linkable assets that AI retrieval systems are specifically designed to surface.
The review platforms like Capterra and others relevant to your market are worth specific attention in this audit because they’re among the most directly AI-readable sources of structured buyer feedback. AI systems parse review platform data in a very literal sense: they read the aggregate ratings, the review volume, the specific language reviewers use to describe your product’s strengths and limitations. That language feeds directly into how AI characterizes your brand in category recommendations.
The specific tactics for building each of these signal types are what the rest of this series covers. But the starting point — the foundation for everything that follows — is doing the audit described in this post. Knowing exactly where your brand stands in AI category recommendations today, what your competitors look like from AI’s perspective, and where the gaps in your external validation profile are most significant: that’s the prerequisite for building a strategy to improve it. You can’t close a gap you haven’t measured. Start there.