The Buyer Channel Most B2B Brands Are Not Measuring
Forty percent of all information-seeking queries now begin in an AI interface rather than a Google search box. That stat comes from Gartner, and it lines up with what we see in client deployments every week. B2B buyers are doing meaningful chunks of their vendor research inside ChatGPT, Perplexity, and Claude before they ever visit a website.
If those engines are not citing your business when a buyer in your category asks the question, you are invisible to a meaningful slice of your pipeline. Google Analytics will not tell you this is happening. Your SEO rankings will look stable. Your impressions will look fine. And your pipeline will slowly, quietly, soften.
We covered the mechanics of HubSpot's new AEO tool in our earlier post on the public beta launch. This post is a different question: how do you know if you are invisible in the first place? Here are five signs, plus a two-minute test you can run right now.
Sign 1: You Have Never Actually Tested It
The fastest diagnostic is also the one nobody runs. Open ChatGPT and ask: "Best [your category] providers for B2B." Read the answer. Count how many vendors get named.
Is your brand in the list? Where does it appear? First, third, buried at the bottom? Does the AI describe what you do accurately, or does it confuse you with a competitor?
If you have never run this query in the actual engines your buyers use, you are guessing at your AI visibility. And in our experience baselining clients, the gap between what operators assume and what the engines actually return is usually large. Most operators believe they are showing up reasonably well in AI answers. Most are not.
Sign 2: Your Google Traffic Is Stable, but Your Pipeline Is Soft
This is the silent killer. Across 50 B2B SaaS keywords in Q1 2026, pages holding top-three organic rankings experienced CTR declines of 18% to 34% once AI-generated answers appeared above the fold. Rankings did not change. Impressions did not change. Clicks dropped anyway.
If your team is reporting flat or improving SEO metrics, but pipeline from organic search is softening, this is the gap. AI answers are absorbing the click before the buyer reaches your result. Your beautiful ranked blog post is still ranked. Nobody is reading it.
The fix is not more SEO content. The fix is content shaped for the AI extraction layer, which is a different discipline with different structural rules.
Sign 3: Competitors Appear in AI Answers and You Do Not
Run a second query: "What is the difference between [your brand] and [competitor]?" Now run: "[Your category] vs [adjacent category]." Now run: "Top [your category] vendors for [your ICP]."
If competitors are appearing in answers where you do not, that is a leading indicator of pipeline you are losing without ever seeing it in your CRM. The buyer asked the AI which vendors to consider, the AI named three competitors and did not name you, and you never get the form fill.
Otterly AI and HubSpot AEO both surface this competitor displacement automatically. If you are not running either, you are not seeing it in the dashboard. You only see the downstream effect: pipeline that should have closed, conversations that never happened, and a slow erosion of category mindshare.
Sign 4: You Have Never Tracked Share of LLM
Share of LLM is the AEO equivalent of share of voice. The calculation is simple: of the queries you care about, what percentage produce an AI answer that mentions your brand?
If you cannot answer that question with a number right now, you have never baselined. And if you have never baselined, you cannot tell whether your content investment over the last year is paying off in the AI layer, or whether it is going entirely to a Google index that fewer buyers visit every quarter.
In our deployment data, the brands that score under 5% Share of LLM are functionally invisible to AI search. The brands at 15% to 30% are established players. The brands above 30% are category leaders. The number is easy to measure if you have the tool. Most B2B operators do not have the tool yet.
Sign 5: Your Content Is Optimized for Google Ranking, Not for Extraction
Look at your top-performing blog post. Does the H1 ask a specific question? Does the first paragraph answer that question directly in plain language? Does the page include a comparison table, a clear list of bullet points, or a data-dense section with at least three citable numbers per 500 words?
If the post is a 2,500-word essay with the answer buried in the middle, it is optimized for time-on-page and Google ranking signals. It is not optimized for AI extraction. Perplexity pulls 71% of its citations from the first 200 words of a page. Google AI Overviews pull 55% of citations from the first 30% of content. If your answer is not at the top, the AI does not see it.
The shift from SEO content to AEO content is not bigger content. It is differently structured content. Question in the H1. Answer in the first 100 words. Data tables and structured sections in the body. The page reads less like a blog post and more like a well-organized reference document.
The Two-Minute Test
Run these three queries in ChatGPT, Perplexity, and Claude (nine total searches, about two minutes of work):
- "Best [your category] providers for B2B"
- "What is [your specific service or category]?"
- "Top [your category] vendors for [your ICP segment, e.g. mid-market SaaS]"
Count how many of the nine answers name your brand. Note the position (first mention, third, buried). Note whether the AI describes you accurately.
If your brand appears in fewer than three of the nine answers, you are in the "invisible" bucket. If it appears in three to five, you are emerging. If it appears in six or more, you are established and the next move is defending your position.
Whatever your number is, that is your starting line. The full playbook for moving from invisible to established lives in the AEO Field Guide we just published, which walks through the 30/60/90 plan we use with every client and on Flywheel itself.
What to Do With This
Pick one of two paths.
If you are on HubSpot Marketing Hub Professional or Enterprise: enable the HubSpot AEO beta this week. The setup takes under 10 minutes. Define 15 to 25 tracked prompts. Run your first baseline. The mechanics are in our earlier HubSpot AEO post.
If you are not on HubSpot: install OpenLens (it is free), define the same prompt list, and run a baseline this week. Both tools will tell you within an hour whether you are invisible, emerging, or established.
The field guide covers what to do after the baseline. The two-minute test above is enough to know whether you need the rest.