Honest About The Limits
We report what we can observe and test, and say plainly where the evidence stops. No guaranteed placements in AI answers.
GEO / AI-Visibility Audit
How ChatGPT, Perplexity, and AI Overviews see your brand.
A one-off audit of how AI systems read, describe, and cite your business — and what to change in your entity data, structured markup, and source coverage so the answers get it right.
$500–1,500One-off audit, priced by brand footprint. Not a retainer.
Early-mover playAn advantage while the space is forming — not a mandatory line item.
2 weeksFrom access to a documented set of recommendations.
Why it matters
A growing share of research now happens inside an answer rather than a results page. When someone asks an assistant to shortlist vendors in your category, something gets said about your business — assembled from your markup, your site, and whatever third-party sources the model happened to ingest. Nobody at your company approved that description.
We’ll be straight about the state of this: measurement is young, the systems change monthly, and no one can promise you placement in an AI answer. What an audit does is make sure the machine-readable version of your brand is accurate and consistent — work that is worth doing on its own terms, and that pays off further if these surfaces keep growing.
Being described wrongly is the default
Outdated pricing, a discontinued product, or a competitor’s framing — models fill gaps with whatever they can find.
Entity clarity is the lever
Consistent structured data and third-party corroboration are what let a system state facts about you with confidence.
Early, cheap, low-risk
Almost everything the audit recommends also improves classic SEO — so the downside if AI search plateaus is close to zero.
[Why Choose Us]
Four reasons to have us run this rather than the first agency selling “AI SEO”.
Proven Process, Not Guesswork
A fixed prompt set run across several assistants, repeated identically each time — so results are comparable, not anecdotes.
Full-Stack Team
Developers implement the structured data, so recommendations become shipped markup instead of a document nobody actions.
Grounded In Classic SEO
Ten years of technical search work behind every recommendation — entity clarity is an old discipline with a new audience.
[What You’ll Get]
A documented read on how machines currently describe you, and what to change so they describe you correctly.
[ 01 ]
Entity Audit
How your brand exists as an entity: name, category, products, people, and locations — and where those facts contradict each other.
[ 02 ]
Schema Review
Every structured data type on the site checked for validity, completeness, and the connections between entities.
[ 03 ]
AI Citation Snapshot
A fixed prompt set run across ChatGPT, Perplexity, and AI Overviews, recording whether you’re mentioned and what’s said.
[ 04 ]
Accuracy Report
Every factual error we find in AI descriptions of your business, traced back to the source that likely caused it.
[ 05 ]
Source Coverage
Which third-party sources models draw on in your category, and where you’re absent from the ones that matter.
[ 06 ]
Competitor Comparison
Who does get cited for your core questions, and what their footprint has that yours doesn’t.
[ 07 ]
Structured Data Roadmap
Prioritised markup and content changes, written so your developers can implement them without further interpretation.
[ 08 ]
Re-Test Baseline
The prompt set and results saved, so the same test can be re-run in three months and compared like for like.
[Our Process]
Two weeks, five stages, and a baseline you can measure against later.
[ 01 ]
Question Set
We agree the questions your buyers actually ask — the prompts the audit will be run against.
Days 1–2 [ 02 ]
Assistant Testing
The prompt set run across several assistants and AI Overviews, with every response recorded.
Week 1 [ 03 ]
Entity & Schema Audit
Your markup, site facts, and third-party sources reviewed for consistency and gaps.
Week 1–2 [ 04 ]
Diagnosis
Each inaccuracy traced to a likely source, and each gap to the change that would close it.
Week 2 [ 05 ]
Roadmap & Walkthrough
Prioritised recommendations, a saved baseline for re-testing, and a call to go through both.
Week 2[Who Needs It]
For brands willing to move before the category settles. If you need a guaranteed return this quarter, start with a classic SEO audit instead — we’ll tell you so.
SaaS
Buyers increasingly ask assistants to shortlist tools. Being described accurately in that shortlist is the whole game.
Considered Purchases
High-value products researched over weeks, where an early AI summary shapes the entire shortlist.
Professional Services
Firms whose expertise should be citable — and whose credentials are currently scattered across inconsistent profiles.
Brands Being Misdescribed
You’ve already asked an assistant about your company and didn’t recognise the answer.
Post-Rebrand
New name, new positioning, and models still confidently repeating the old one.
In-House SEO Teams
You need a defensible baseline and a technical roadmap to take to your engineering team.
[Pricing]
A one-off audit, priced by how large your brand footprint is. This is an early-mover investment, not a mandatory line item — if classic SEO fundamentals are still broken, spend the money there first.
Single Brand
$500–750one-off
One brand, one market, a focused product or service line.
- Entity and schema audit
- Prompt set across three assistants
- Accuracy report with source tracing
- Prioritised structured data roadmap
- Saved baseline for re-testing
Multi-Product
$900–1,200one-off
Several products, categories, or buyer segments to test separately.
- Everything in Single Brand
- Prompt set per product or segment
- Competitor citation comparison
- Source coverage analysis for your category
- Implementation specs for developers
- Walkthrough call with the analyst
Enterprise / Multi-Market
$1,300–1,500one-off
Multiple markets or languages, or a group with several brands.
- Everything in Multi-Product
- Testing per language and market
- Cross-brand entity consistency review
- Knowledge graph and profile reconciliation
- Roadmap phased across quarters
Re-testing against your saved baseline is $300 per run, whenever you want to see whether the changes moved anything. No retainer, no obligation to repeat.
Curious what the assistants say about you?
Send us your brand and category. We’ll run three questions your buyers would ask and send you the raw answers — free, and often uncomfortable reading.
Be described accurately
Two weeks to a documented baseline and a roadmap your developers can ship — most of which strengthens classic SEO regardless of where AI search goes.
[FAQ]
Generative Engine Optimization — making your business legible and accurately described to systems that answer questions rather than list links. In practice it’s entity clarity, structured data, and source coverage.
No, and be sceptical of anyone who does. Model behaviour changes constantly and citation isn’t a rankable slot. What we can do is remove the reasons you’re currently absent or described wrongly.
Possibly, for some businesses — and we’ll say so. If your technical SEO is broken, fix that first. This is an early-mover play that happens to also improve classic search foundations.
A fixed prompt set, run identically across several assistants, with every response recorded and saved as a baseline. It won’t be perfect, but it is repeatable and comparable over time.
ChatGPT, Perplexity, and Google AI Overviews as standard. Others can be added if they matter in your market.
Ship structured data changes, correct inconsistent facts across your profiles and site, and close gaps in the third-party sources models rely on. Your team can do it, or we can.
Yes, though they overlap. A classic audit asks whether pages can rank. This asks whether machines can state facts about you confidently and correctly.
Expect meaningful drift within a quarter — these systems change fast. That’s why we save the baseline, so re-testing is cheap.
No. This is a standalone audit. If you already work with us, the recommendations fold into your existing roadmap at no extra cost.
Then you have documented evidence that your entity data is clean, plus a baseline to measure against later — and we’ll tell you honestly that there was little to fix.