What Buyers Actually See When AI Tells Your Story

A buyer asks an AI model for help choosing an IT service management platform. Here's what they get:

The buyer sees each platform described by what it does and where it fits.

Now notice what isn't there. No analyst rankings. No customer logos. No awards. No satisfaction scores. ServiceNow was named a Gartner Magic Quadrant Leader for nine consecutive years. Jira Service Management's site declares "60,000+ customers trust Jira Service Management." None of it appears.

B2B positioning works in layers. At the top, emotional: brand voice, vision, aspiration. Below that, reassurance: the proof that makes a decision defensible. And at the base, structural: what the product does, where it sits in the market, what makes it different.

Structural claims are the only type of positioning that has any chance of reaching the buyer. Not a single reassurance or emotional claim appears anywhere in the study's AI responses. Across every industry, every vendor, every AI model.² Zero.

It took me a while to see why, and then it was obvious. These claims all exist for the same reason: to help a human decide. They create aspiration, a sense of something bigger. Or they create confidence, proof and validation that make a decision defensible. They give a person something to feel. But AI isn't scared of getting fired for a bad recommendation. It has no feelings to appeal to, no gut to validate. So it drops everything designed to move a human and keeps only what helps it explain the category.

Companies haven't been describing themselves wrong. They built their positioning for a human audience. Now there's a non-human intermediary in the loop, and it has its own logic for what gets through.

And getting through the filter is not the end of the story. Even a structural claim that reaches the buyer doesn't always arrive as sent. Here's the same buyer, in the same purchase, asking about Salesforce in two different ways:

Same buyer. Same purchase. Same AI model. The frame shifts from help me choose to help me assess risk, and the positioning inverts. Customization becomes complexity. Scale becomes lock-in. Ecosystem breadth becomes vendor dependency. The claim doesn't disappear; it backfires. This is a pattern I saw across the entire study. What gets through depends not just on the claim but on how the buyer asks and where they are in the purchase.

Turning a rival's strength into a risk is a standard play in B2B selling. What's new is who's running it: not a competitor's rep the buyer knows to discount, but an assistant the buyer trusts to be neutral.

Remember the buyer's IT service management query? The full response named seven platforms: ServiceNow, Jira Service Management, BMC Helix, Freshservice, Ivanti, ManageEngine ServiceDesk Plus, Cherwell Service Management.

Zendesk wasn't there.

Zendesk is a $2 billion company that has spent years pushing into IT service management, selling an employee service product to exactly this buyer. And in this buyer's conversation, it doesn't exist. If your first reaction is that Zendesk is a customer service company, not an ITSM vendor: that's exactly the answer Zendesk has been spending all that effort to change. For this buyer, it hasn't changed.

That's not random. When a vendor is absent from an AI response, the space doesn't stay empty. 78% of the time, a competitor fills it. The buyer doesn't notice you're missing. They see a recommendation for your competitor.

That's competitive replacement.

It works in the other direction too: when a buyer asked AI about one vendor by name, the rest of the category's positioning almost never entered the answer.

In observability (the tools engineering teams use to monitor their software systems), a widely covered category, competitive replacement is total. Every time a vendor is missing from a response, a competitor takes the space. And company size doesn't determine who gets through. Dynatrace is a publicly traded company with over $2 billion in annual revenue and one of the longest track records in the category. It appears in 19% of buyer conversations. Honeycomb is a fraction of the size. It appears in 86%.

The difference is what each company's positioning gives AI to work with. Honeycomb maps cleanly to a slot AI recognizes: the observability company built around debugging complex distributed systems. Dynatrace's broader "software intelligence" positioning gives AI no single role to place it in. Scale doesn't decide who fills the space. Clarity does.

Cyber Resilience tells the opposite story. There's less public coverage of the category, so AI has less to draw on. Even Rubrik, the vendor AI mentions most in this category, appears in only 35% of conversations. When a vendor is missing, a competitor takes the space about half the time. The other half, AI gives generic advice without naming anyone. In a quiet category, that's often the outcome: not replacement, just silence. Veeam, a company with over 550,000 customers and over $1.5 billion in annual recurring revenue, didn't appear in any query responses whatsoever.

In this study, visibility and accuracy aren't separate problems. The vendors AI mentions most are the ones it describes most accurately. The vendors it rarely mentions get described wrong on the rare occasions they come up. The root cause is the same: whether your positioning gives AI a clear role to assign you.

The amount of content a company has online shows no correlation with how its positioning fares. This surprised me, so I double-checked. Kyriba, a treasury management vendor with one of the smallest content footprints in the study, about 3,400 pages indexed, gets its positioning through to buyers intact more than 10% of the time. Salesforce, with 7.8 million pages, gets 2.5%. Two thousand times the content, a quarter of the result.

More content gives AI more of your words to work with. It doesn't give AI a clearer idea of what you are. Content is how you say it. Positioning is what holds up when someone else says it for you.

Take Motive. They started in trucking, building the electronic logging devices that keep commercial fleets compliant with federal safety regulations. Then they kept going. AI dashcams. Vehicle diagnostics. Spend management. Workforce tools. More than $600 million raised. When Motive's claims do reach a buyer through AI, they arrive more faithfully than almost any other company's in this study. That should be a success story. It isn't.

Here is what Motive says about itself, and here is what a buyer sees when they ask AI for help choosing a fleet management platform.

Notice the shape of Gemini's description. A platform built around the compliance device they started with. Compliance as "their foundation." Samsara's biggest competitor, a company described by reference to someone else. The claim that gets through most faithfully is the one that defines where AI places them. Claude does the same thing: "best value for compliance-heavy operations." The AI dashcam story, the spend management expansion, the $600 million raised, none of it changes the position AI has assigned them.

That's the ceiling. The claim AI preserves most faithfully about you is the one that caps what else can get through.

And the position is sticky. Once AI has assigned it, even basic facts about you struggle to update. The company rebranded from KeepTruckin to Motive in 2022. Four years later, GPT-4o still uses the old name.

Motive is not the only company capped by its best claim. HubSpot appears in nearly every CRM conversation AI generates. It's not absent. It's not ignored. It built a full CRM suite, AI tools, an operations hub, a commerce platform, and arguably the largest content and education ecosystem in B2B. But no matter the question, AI files it in the same position: "suitable for small to medium businesses focusing on inbound marketing." That described HubSpot accurately years ago. Everything it has built since hasn't moved it. Ask about Revenue AI instead, and HubSpot still appears, "known for inbound marketing software." A different question, the same answer: inbound marketing. The ceiling held.

Somewhere right now, a buyer is asking AI about your market. I've read thousands of these answers, so here's what's likely happening to you. Everything you built to make choosing you feel safe is not in the answer. If your positioning hasn't made it to the response, odds are a competitor is standing in your space and the buyer has no idea. The claim AI repeats most faithfully is the slot it has placed you in.

Whether AI includes you comes down to a position that maps to a slot AI recognizes, and enough public coverage of that slot for AI to draw on. In a quiet category, the gap is coverage: more independent writing about your market would change what AI can construct. In this study, quiet categories were the exception. In most markets, AI already has plenty to say: the gap is positioning, and more content won't fix it.

This is not the first time your story is being retold. Your champion pitched you to their CFO and lost half. The analyst compressed it into a quadrant. Your sales rep tells a version of their own. Every retelling is a prism bending the story somewhere you mostly can't watch.

AI changes two things. First, timing: this retelling comes first, before the demo, before the first call, before you know this buyer exists. They're already asking. Second, legibility: a buyer asks, AI answers, and what your positioning became is sitting right there on the screen.

You can probably already name the role AI has assigned you. What's holding that story in place? What would move it?