Where this actually shows up
These are the surfaces that now answer questions directly instead of handing over a list of links. Each one picks a small number of businesses to name.
There is no sixth place
On a results page, ranking sixth still gets you seen. In a generated answer there is no sixth. There is one paragraph naming a handful of businesses, and everyone else is invisible for that question.
That is a harder game, but a more winnable one for a small business than you would expect. These systems are not ranking by domain authority or backlink count. They are looking for clear, checkable facts they can lift into an answer. A one-person business that publishes real prices and answers questions plainly is easier to quote than a large competitor whose site says "award-winning solutions tailored to your vision".
Read the full guideWhat we actually change
Six things, in roughly this order of impact. None of them are tricks; they are the things that make a business easy for a machine to describe accurately.
How the work runs
Three stages. You see the before-and-after in the assistants themselves, not in a made-up score.
Baseline: what the assistants say today
We ask ChatGPT, Claude, Gemini and Perplexity the questions your customers actually ask, and write down whether you are named, in what position, whether the description is right, and who gets named instead. That competitor list is the most useful research in the whole engagement.
Fix the inputs
Content restructured so it answers first, prices published, schema added and validated, robots.txt opened to the AI crawlers, llms.txt written, entity details made consistent everywhere you appear.
Re-test and keep going
We re-run the same prompts monthly and track the direction of travel. Retrieval surfaces move within days; training-data answers take longer. You get the actual before-and-after, not a dashboard of invented scores.
Honest about what this is
Nobody can guarantee that ChatGPT will recommend you, and you should walk away from anyone who says otherwise. There is no ranking dashboard for AI answers, no console, no verified position. What exists is a set of inputs the models read, and a set of reasons they skip businesses.
Our job is to remove every one of those reasons and then measure it the only way that is currently honest: by asking the assistants the same questions every month and writing down what they say.
It also does not replace classic SEO. Google Maps and organic search still send most of the work for a local business, which is why this usually runs alongside getting found on Google rather than instead of it.
Who it is worth it for
Businesses where people ask for a recommendation before they buy: trades, dental practices, law firms, agencies, specialist services, anything with a considered decision behind it.
It is worth less if you sell on impulse or purely on price, and worth more if your competitors are still writing brochure copy. Most of them are.
If you are not sure which side you fall on, ask us. We will tell you if we think it is too early, which happens more often than you would think.
Quoted per project
The starting point differs too much for a list price. A site with clean structure and schema already in place is a small job. One with hidden prices, no schema and blocked crawlers is a bigger one. Either way you get a fixed price before anything starts, and it can be folded into a Stay Visible plan if you want it maintained.
Common questions
What is AI visibility?
Whether your business gets named when someone asks an AI assistant like ChatGPT, Claude, Gemini or Perplexity for a recommendation. Unlike a results page with ten links, an AI answer names two or three businesses, so you are either in it or absent entirely.
How is it different from SEO?
They overlap, but the goal differs. SEO competes for a position in a ranked list. AI visibility competes to be one of the few names inside a single generated answer. Extractable facts, published prices, structured data and third-party mentions matter more; keyword density matters less.
How long does it take?
Retrieval-based surfaces such as Perplexity, ChatGPT Search and Google AI Overviews can reflect changes within days of a recrawl. Answers that lean on model training data move over weeks or months, because they only change when the model is updated.
What does it cost?
Quoted per project, because the starting point varies so much. You get a fixed price before we start, and it can be folded into a monthly plan if you want it kept up.
Can you guarantee ChatGPT will recommend my business?
No, and anyone who guarantees it is selling you something. We control the inputs the models read: your content, your structured data, your crawler access, your consistency across the web. We cannot control the model. What we can do is remove every reason for it to skip you.
Ready to be part of the answer?
Tell us what you sell and who buys it.