Find out whether your buyers use AI search at all, what they ask it, and how consistently your brand is described across the sources those engines read.
Most businesses being sold AI search optimisation have never been shown evidence that their own buyers use AI search. The AI Visibility Diagnostic establishes that first: whether the people who sign your contracts ask an answer engine about your category, what they ask it, and what it currently says about you. Where the honest finding is that your market has not moved yet, that is the finding, and we will say so rather than sell you a build.
Do your buyers actually use AI search?
This is not the same question as "is AI search growing". It is growing. Whether it is growing among your buyers is a separate matter, and the answer varies enormously by sector: construction, manufacturing and mortgage broking behave very differently from software and professional services.
So the diagnostic starts with evidence rather than with the headline. We look at your own analytics for sessions referred by ChatGPT, Perplexity, Google AI Overviews, Microsoft Copilot and Claude, at how your buyers describe their own research when asked, and at whether your category is one those engines are being asked about at all.
What does the diagnostic answer?
1. **Do your buyers use AI search, and at which point in the decision?** Early orientation and late verification are different problems with different answers.
2. **What prompts do they actually run?** Not the keywords you rank for. The questions people type, in their words.
3. **What do the engines say when we run them?** Where you appear, where a competitor appears instead, and where the answer names nobody at all.
4. **Which sources are those answers drawn from?** Answer engines cite as they go, and the citations are the map of what you would have to influence.
5. **Where is your brand described inconsistently?** A different service list on a directory entry, an old company name on a profile, two versions of what you do.
6. **Is there anything worth building yet?** And if so, what, in what order, and roughly what it would take.
Why will you not quote the build work first?
Because we would be guessing. The work that makes a brand visible to an answer engine depends entirely on which sources that engine is drawing from for your category, and that is not knowable in advance. An agency quoting a content programme before it has looked is quoting a template.
It also means we can be wrong cheaply. Seven hundred and fifty pounds is a small amount to spend establishing that a much larger programme is not yet worth commissioning.
What happens if the answer is no?
You get the report, we bill you for stage one, and we recommend you wait. That is a real outcome, not a hedge, and for some of the sectors Flowbird works in it is the likely one today.
The report is still worth having. It gives you a baseline to re-run in a year, it usually surfaces brand inconsistencies worth correcting for ordinary search reasons, and it means the next person who pitches you AI search optimisation can be answered with evidence.
How is this different from SEO?
Search engine optimisation is about ranking a page. Answer engine optimisation is about being the source an answer is assembled from, which is a different mechanic: the engine reads, summarises and cites, and the page it cites is often not yours. So the work is as much about what third parties say about you as about what you publish.
They overlap, and a site that is already well structured has a head start. But a strong search ranking does not guarantee you are in the answer, and that gap is what the diagnostic measures.
This is also the outward-facing half of a larger question. What AI can do with the records you already hold is the inward-facing half, and that is what AI readiness is about. They are separate purchases and neither one depends on the other.
What happens after stage one?
If there is something worth building, we scope and price it separately, against what the diagnostic found. If there is not, we say so and you have lost the cost of finding out rather than the cost of a programme.
Either way the report is yours. There is no obligation to do the next piece of work with Flowbird, and no part of the diagnostic is withheld to make you.
What does the diagnostic cost, and what do you get?
Small enough to decide on your own authority, and specific enough to act on - including on the occasions when what it tells you to do is wait.
Where to start when you are not sure what to fix first. Fixed price, a written report in two to three weeks, and the report is yours to keep whether or not anything follows. Prices exclude VAT.
AI Visibility Diagnostic
£750
Research before build. Whether your buyers use AI search at all, what they actually ask it, and where your brand is described inconsistently across the sources those engines read.
Stage one on its own. We do not quote the work that follows until this has established whether there is anything worth building, and in some markets the honest answer is not yet.
Whether your buyers use AI search, evidenced from their behaviour rather than assumed from the headlines
The prompts they really run, tested against ChatGPT, Perplexity, Google AI Overviews and Microsoft Copilot
Where you appear in the answers, where a competitor appears instead, and where nobody does
The sources those answers are drawn from, and which of them describe you wrongly, partially or not at all
A consistency check across your own site, your CRM-published content, your directory entries and your third-party profiles
A written report you keep, with a recommendation that may well be to do nothing yet