Courtyard · AI Visibility Index
The State of AI Visibility2026 · for small businesses
AI will happily name your business. Understanding it is another matter. We measured how well ChatGPT, Gemini, Perplexity, and Google AI Mode understand local businesses across the questions customers ask, from services to availability to price.
Edition 01 · July 2026
Being named isn’t being understood.
How well AI understands each, scored 0 to 1. It has the surface; the specifics fall away.
The depth of understanding
The deeper the question, the thinner AI's understanding
AI can tell a customer what a business does. But as the questions get more specific, the ones that actually decide who gets chosen, its understanding falls away.
Measured across the buyer’s journey, understanding drops from confident on services to almost nothing on price. The specifics a customer weighs in between, whether you’re available and what sets you apart, never reach the confidence of the basics.
In their own words
What that looks like when a customer asks
The scores come from answers like these, taken straight from the assistants with the business names removed. AI handles what a business does. The moment a customer needs a specific, a price, an opening, what sets one place apart, it hedges.
“What's your service-call fee?”
I could not find a published flat diagnostic or service-call fee on the company's website. It advertises free estimates for system replacements, but that is different from a diagnostic visit to troubleshoot a repair.
“What's on the class schedule this week?”
I don't have live access to the gym's current class schedule right now. I can help you find it quickly if you'd like, or I can summarize typical class types they offer based on public listings.
“What's the diagnostic fee?”
The company does not publicly publish a flat service-call or diagnostic fee on their website. Rates vary depending on your building type, such as high-rise co-ops or commercial properties.
“What makes this practice different?”
I can't reliably answer this because the practice's name appears to be used by multiple dental practices, and without the exact clinic page I can't confirm which specific credentials or technologies they claim.
Answered, but noncommittal
Six in ten answers hedge instead of answering
When AI does engage a specific question it usually will not commit, a tell that it doesn’t understand the business well enough to be sure. Sixty percent of answers hedged: “prices aren’t listed,” “you’d have to check with them,” “it may vary.” Only 40% gave a direct answer.
The surprise is what it almost never does: send the customer away. Fewer than 1% of answers deflected outright. AI would rather answer vaguely than admit it doesn’t know, which is harder for a business to notice and easy for a customer to bounce off.
- “prices aren't listed”
- “you'd want to check with them directly”
- “it may vary”
- “I couldn't find a specific figure”
- “appears to offer”
The price question
Ask what it costs, and AI comes up empty for three in four businesses
Pricing is where AI is weakest, and it is not close. For roughly three out of four businesses, assistants could not produce a specific price, even for the businesses they recommend most.
This is not a claim that these businesses hide their prices. It is that the assistant, standing in for a customer, could not find or state one. The rate varies by category but never drops below 60%.
No single answer
No two assistants see your business the same way
There is no single “AI understanding” of a business. Run the same questions across all four assistants and no two land in the same place. Understanding ranges from 65 out of 100 (Google AI Mode) to 52 (ChatGPT), and they share one soft spot: price.
| Assistant | Services | Availability | Pricing | Fit | Standout | Overall |
|---|---|---|---|---|---|---|
| Google AI Mode | 0.84 | 0.68 | 0.38 | 0.74 | 0.67 | 65 |
| Gemini | 0.85 | 0.70 | 0.26 | 0.75 | 0.69 | 63 |
| Perplexity | 0.76 | 0.58 | 0.53 | 0.66 | 0.49 | 61 |
| ChatGPT | 0.74 | 0.57 | 0.27 | 0.58 | 0.48 | 52 |
They don’t just score differently, they contradict each other. Ask the same question and each has its own facts, and sometimes a flat disagreement.
Everywhere at once
It doesn't matter where you are
This is not a big-city advantage or a small-town gap. Across all 50 markets, understanding lands in a narrow band. No metro rises above 52 out of 100, and the US and Canada are effectively tied. Wherever your business is, AI understands it about as poorly.
Where AI learns about you
Discovery is sourced from others. Depth comes from you.
So why does understanding thin out? Because of where AI gets it. When AI recommends who to use, it reads the reputation layer, the sites that talk about businesses: Google, Reddit, directories, local critics. Ask about a specific business and the mix flips: the business’s own site, absent from discovery, becomes 39% of citations, while Google falls from 14% to 5% and Reddit from 4.6% to 1.7%.
It’s the one source that’s genuinely yours, and AI leans on it hard: it cites the business’s own site in 72% of depth answers, and in 42%it’s the only source, with no directory to fall back on. If that site is thin or hard to read, AI has little to work with.
How AI finds local businesses: the full source breakdown, including the sites AI reads in each industry and why reaching your site isn’t the same as reading it.
What it means
The gap is consistent, and it’s closeable
Across every category the finding repeats: AI can name a business, but it doesn’t understand it deeply enough to answer for it. As a customer’s questions get specific, the price, the availability, what makes you different, the answers thin out. The businesses that get chosen will be the ones AI understands well enough to speak to.
That understanding is something you can give it. Courtyard publishes and maintains a live knowledge base of a business, its services, hours, pricing, and availability, in a form ChatGPT, Gemini, Perplexity, and Claude can use. It raises the odds you’re understood and recommended accurately. No one can guarantee a given answer, but AI can’t put you forward for what it doesn’t understand.
FAQ
AI visibility, in brief
- What is AI visibility?
- AI visibility is how well assistants like ChatGPT, Gemini, Perplexity, and Google AI Mode can find, understand, and accurately describe your business when a customer asks. Being named in a list is only the surface. Real visibility means the assistant can answer the specific questions that lead to a booking.
- How is AI's understanding of a business measured?
- Each week we put 24 real customer questions to the four major assistants across 50 US and Canadian markets, then score how well each carries a buyer's journey (services, availability, pricing, fit, and what sets a business apart), not just whether it responds.
- Why does AI hedge instead of answering?
- When an assistant lacks an accurate, current source for a specific detail, it avoids committing. In this edition, 60% of answers hedged (“prices aren't listed,” “you'd have to check”) rather than giving a direct answer.
- What can a business do about it?
- Give assistants an accurate, current source to read. A live knowledge base of your services, hours, pricing, and availability raises the odds AI understands and recommends you correctly. No one can guarantee a given answer, but AI can't put you forward for what it can't understand.
Methodology
How this was measured
The questions
Each week we put 24 real customer questions to ChatGPT, Gemini, Perplexity, and Google AI Mode, asked three times each, for 288 answers per market across 50 US and Canadian markets.
Depth
For the businesses each market recommends most, we run a buyer's journey (offerings, availability, pricing, fit, differentiation) and score how well AI can carry someone toward booking, not just whether it responds.
Posture
Every answer is classified as direct, hedged, or deflected. Pricing is checked for a specific figure that holds up across assistants.
This edition
A point-in-time snapshot from the week of July 10, 2026. The live, continuously updated view is the AI Visibility Index.
These figures are an observed measurement of what third-party AI assistants say, gathered by Courtyard. They are not a rating, certification, or endorsement of any business. Full methodology on the Index.