A VP of revenue operations at a 1,800-person SaaS company opens ChatGPT. She asks: “best revenue intelligence platforms for mid-market sales teams.” ChatGPT names six. The brand her CMO had been planning to evaluate, one with strong analyst coverage and an aggressive LinkedIn presence that week, isn’t among them. Perplexity, Claude, and Gemini converge on different shortlists. The brand doesn’t show up in those either.
That company spends $4 million a year on content. They rank page-one for thirty commercial keywords. None of it shows up in the answers the buying committee is actually reading.
This is the part of the buyer journey that’s broken for most enterprise brands.
The Convergent Finding
Three independent 2026 audits all landed in the same neighborhood. The 2026 Fuel AI Index, based on an audit of 1,000 enterprise domains across SaaS, finance, retail, and legal, found 62% of enterprise brands technically invisible to generative AI, with citation failure rates of 81% on unbranded queries about their own core services.
The Pulp Strategy GEO Benchmark Index, based on 408,000 prompt simulations across 350 brands, found 68% of brands don’t appear in AI-generated lists in their own categories. The 2X AI Innovation Lab AI Visibility Index, released in April, found only 4.3% of B2B companies maintain visibility across the early-stage discovery queries buyers actually run.
Three different methodologies, three different sampling frames, the same conclusion.
These studies don’t measure AI visibility in identical ways. That’s the point. The pattern keeps showing up across different datasets, categories, and definitions: brands that look strong in search are often weak in AI-mediated discovery.
Why Your SEO Authority Doesn’t Carry Over
The first instinct is to assume this is solvable with more of what worked before. The evidence argues against that.
Profound’s analysis of 50,000-plus prompts across major AI platforms found a correlation of 0.05 between organic search traffic and AI citation. Backlinks fared worse, at 0.038. Domain Rating and URL Rating, the metrics most SEO programs orient around, didn’t predict AI visibility in a meaningful way. Semrush data found 90% of ChatGPT citations come from pages ranking position 21 or lower in Google.
The mechanism is structural. AI citation behavior runs on different signals than organic search ranking, and the brands that look strong in one channel can be entirely missing from the other. The same enterprise SaaS company can dominate page-one for its category keywords and never get cited by ChatGPT when a buyer asks “what should I use for X.” Both are true. They’re measuring different things.
Ghost Citations are Doing Most of the Damage
There’s a second layer to the invisibility problem that’s easier to miss, because it doesn’t look like invisibility in a dashboard.
Profound’s analysis of AI citation behavior found 62% of citations are what they call ghost citations: the brand’s URL appears in the source list at the bottom of the AI’s response, but the brand name never appears in the answer text itself. From the model’s perspective, your page was useful. From the buyer’s perspective, you weren’t in the conversation. They read the answer. They don’t click through to the footer of citation links.
AirOps’ analysis of 45,000-plus citations found brands appearing in both the body text and the source list were roughly 40% more likely to resurface across repeated runs of the same query. The mention-and-citation pair carries forward. The citation alone often doesn’t.
A brand can register as “cited” in a tracking tool and still be functionally absent from the conversation.
Platform Divergence is Structural
The other instinct is to treat AI as a single channel. The data argues against that too.
Growth Unhinged found only 11% of domain citations overlap between ChatGPT and Perplexity. That means 89% of AI visibility opportunities are platform-specific. The brands winning on ChatGPT aren’t the same brands winning on Perplexity, and the content patterns that work on one don’t transfer cleanly to the other. Microsoft Copilot favors business publications. Google AI Mode favors first-party sources and YouTube. Claude favors technically precise reference content.
A buyer running the same question across three AI platforms can read three substantively different sets of recommended vendors. The brand that’s invisible on whichever platform the buying committee happens to use is, from that committee’s perspective, invisible everywhere.
The Diagnostic, Not the Playbook
The temptation in this category is to publish a ten-step remediation roadmap. The honest starting point for an enterprise marketing leader is shorter.
Run the audit your team probably hasn’t run yet: where does your brand appear, by buyer journey stage and by platform, across the queries your category actually faces. Not branded queries, where every brand looks fine. Unbranded discovery and comparison queries, where the gap lives.
One performance management software company Resonate Labs audited had 81% visibility on comparison-stage queries, which looked encouraging until we looked at problem-identification queries. Their presence rate at the earliest stage of the buyer journey was 8%. Competitors were framing the category problem for buyers before this brand ever entered the conversation.
That’s the cost of AI invisibility for enterprise brands. It happens before any analytics surface a single signal. The pipeline gap shows up six months later as a win-rate question nobody can answer.
The audit is the place to start.
Shane H. Tepper is the author of Cited: How B2B Brands Win in the Age of AI-Generated Answers and co-founder of Resonate Labs, a B2B GEO consultancy.
