Last time, one laptop brand vanished between two ways of asking the same question. That work started with a media partner whose vendors kept asking for help with AI visibility and selection. Since then I've run the same kind of analysis in three more B2B categories, for agencies whose clients sell to banks, payments teams and security buyers. The pattern held in every one.
This one is about the step before selection: being in the answer at all.
Before AI can select you, it has to include you
When a buyer asks an AI tool who they should consider, it hands back a shortlist: the handful of vendors it recommends. The buyer still makes the call, but they make it from that list. And AI can only shortlist the brands it has already put in its answer.
That's Inclusion, the first layer of The Selection Layer. Positioning, Preference and Selection all depend on it. If you're not in the answer, you're not framed, prioritised or chosen. You simply aren't there.
Inclusion is decided one question at a time
In September I tracked the questions UK banks ask when they're choosing a partner to modernise their payments systems. Two of them ask for the same thing: "Who can help meet the ISO 20022 deadline?" and "Best partner for ISO 20022 migration".
One of the world's largest IT services firms was named in 67% of answers to the first. In answers to the second, it wasn't named once.
Nothing about the firm changed. Only the buyer's wording did. AI makes the inclusion call separately for every question it's asked, and your buyers don't all ask the same way.
Three things decide it
AI builds its answer from sources it trusts. Whether your name makes it in comes down to three boxes, and you have to tick all three.
Does AI know you exist as a player? I'll start with us. When a question names Decoding Growth, AI includes us about two-thirds of the time. When it doesn't, we turn up in roughly one answer in 200. Decoding Growth is a new name, so none of that surprised me, but it's exactly why we track it.
How you'd spot it: AI names you when a buyer asks for you by name, but not when they describe exactly what you sell.
Can it find and read what you do? One of the big card networks is named in 84% of AI answers about card payouts, and in 14% about cross-border and treasury payouts. On those cross-border answers, the models drew on its own website 4% of the time. Explainer pages from Freysoft, a small fintech and payments software engineering firm, were used four times as often. Size doesn't decide this. Pages that answer the buyer's actual question do.
How you'd spot it: The pages AI cites for your buyer's question belong to someone else, and you have nothing that answers it directly.
Does anyone else vouch for you? I learned how this works at Spiceworks. IT pros came to ask each other what they'd bought and what had gone wrong, because they didn't trust what vendors said about themselves. They wanted an impartial view. AI does exactly the same - at speed.
How you'd spot it: Your pages get cited, but your name doesn't appear in the answer.
Being read isn't being named
In enterprise security, one vendor's own website fed dozens of AI answers about how to solve a problem, and the vendor was named in about one in six of them. Gemini answered a question on securing SaaS applications from another vendor's glossary, cited it five times, then recommended two competitors.
The obvious response is to publish more. These vendors already had. Their content was being read and used, and it still didn't get their name into the answer.
It isn't a quirk of our sample. A Semrush and Growth Memo study found that in six out of ten cases where a site appeared in an AI answer, the brand behind it wasn't mentioned. Explainer content gets cited. Comparison content is far more likely to get you named.
That's why the third box is the hardest to tick. You can fix the first two on your own site. The third you have to earn somewhere else.
You can't see it happening, so test it
There's no Search Console for AI answers. You can't see what buyers type, every tracking tool (including the one I use) samples rather than sees, and buyers rarely click through. They take the answer and act on it somewhere else. The only way to know where you stand is to ask the way your buyers do.
Run the test on your own brand.
Write down five questions your buyers actually ask, in their words - including the specific ones with a constraint, an integration or a deadline in them, and at least one pair that asks the same thing two ways.
Open a private window and log out of every AI tool, so your history and memory don't colour the answer.
Run each question on ChatGPT, Gemini, Perplexity and Google's AI Mode, and note who's named and which pages are cited.
Run them again next week. AI tools won't give you the same answer twice, so one run tells you very little.
Check what you find against the three "How you'd spot it" signs above. It won't give you a score, but it will show you which box you're not ticking.
Time needed: 30 minutes now, 10 minutes next week.
If you want to know where you stand today, the AI Shortlist Scan is the place to start: how AI represents you, how you compare, and how it arrives at the preference it hands your buyers. £1,995, delivered in seven days, money-back guarantee.
The full piece, with every source and the method, is here: The Inclusion Problem
If you run the test, reply and tell me what you found. I read every one.
John
Decoding Growth (formerly Get2Growth)