Read your AI Visibility score
What does my AI Visibility score actually measure?
AI Visibility is a rank score. Synup asks ChatGPT, Gemini and Perplexity a set of discovery questions about your category and city, then scores each answer on whether you were named and how near the top. Nothing else feeds it: not your rating, not your review count, not your directory listings.
That is the most useful thing to know about this page, because the Visibility factors card underneath the score measures different things.
Key facts
| Where it lives | AEO → Overview, for one location |
| What it scores | Whether and how high each engine names you across the discovery questions, 0–100 |
| What it ignores | Reviews, ratings, listings and anything else that is not a rank in an answer |
| The label | Excellent, Good, Average, Below average or Low, from the score bands below |
| How it moves | It changes when the location's report updates, on its schedule or when refreshed by hand |
Steps
Open AEO, choose the client and location, and stay on Overview.
Read the score and its label first, in the gauge.
Read the five cards across the top (see below), then the prompt grid.
Read Visibility factors last, as a list of places to investigate.
What the Overview is showing you
The Overview has three parts that read together. The row of five cards gives the quick read, the gauge gives the score and its label, and the engine cards beside it show how that score splits by engine.
Each engine card shows that engine's own score and an arrow with a percentage, which compares it with the previous stored report. The engine cards are colour-coded on the same 60 and 30 scale as the gauge. An engine that is not available in your account shows Add the API key to track followed by the engine name instead of a score.
Why a score can sit next to a weak factor
Here is the Visibility factors card for a location scoring 49:
Only one of those five is inside the score.
| Factor, and what it measures | Feeds your AI Visibility score? |
|---|---|
| Relevance: how visible you are across the discovery questions | Yes, this is essentially the score itself |
| Reviews: your rating and review-volume signal | No |
| Engagement: the share of answers that mention you | No |
| Freshness: how recently the listing was updated | No |
| Citations: directory citation coverage | No |
In this example Freshness is a perfect 100 while the score is 49, and Reviews is 38. Neither is a contradiction. The score records what happened in the tracked AI answers, and the factors report separate signals. The factors are measured when the report is generated, so they can differ from what Reviews or Listings show right now.
Read the factors as places to investigate, not as inputs to today's number or proof of why an engine ranked you. A low factor can help you choose what evidence to inspect next, but this report does not establish that changing it caused, or will cause, a change in visibility. Beside each factor is a link to the area that moves it: Target prompts, View reviews, Update listing or Fix citations.
The five cards across the top
The row of cards is the fastest read on the page, and each answers a different question.
| Card | Example above | What it tells you |
|---|---|---|
| AI Visibility | 81/100 | How often and how highly you are named |
| Mention rate | 91% | The share of answers you appear in at all |
| Avg position | 2 | Where you land when you appear. Lower is better |
| Prompts covered | 15/16 · 1 excluded | Questions where at least one engine named you, out of all questions. An excluded question is one where the engines only asked for your location, and it is not counted as a miss |
| Sources cited | 19 | A coverage total: directory sources found, other sources detected and key directories still missing |
Each card also shows how it changed since the previous stored report, and the AI Visibility and Sources cited cards add a small trend line once there are two months of reports.
Mention rate and Avg position are the pair to watch together. A high mention rate with a poor average position means you appear often but usually lower in the answer.
Sources cited is not a count of sources citing you. It is a coverage total, so it can be well above zero when your visibility is very low.
This location scores 7/100, is named in 13% of answers, and still shows 16 sources cited. That does not mean 16 sources cite the business. It shows that source coverage and business visibility answer different questions. Its chips under the page title say the rest: Absent on 11 prompts and #3 of 3.
The bands, and what they're called
| Score | Label the product shows |
|---|---|
| 81–100 | Excellent: "Strong visibility, well positioned when customers ask an AI assistant." |
| 61–80 | Good: "Good visibility, with room to improve." |
| 41–60 | Average: "Moderate visibility, with clear opportunities to improve." |
| 21–40 | Below average: "Limited presence. AI assistants rarely bring you up." |
| 0–20 | Low: "AI assistants almost never mention you yet." |
The boundaries are tighter than they look: 80 reads Good and 81 reads Excellent. Don't let the label do more work than the number.
There is a second scale on the same page. The gauge colour and the Strong, Fair and Weak chips on the roll-up use 60 and 30: green from 60, amber from 30 to 59, red below 30. So a score of 65 is labelled Good and shown in green, while a 55 is labelled Average and shown in amber. The label and the colour use different cut-offs, so read both.
Sydekick's read, and when to trust it
Above the page title you may see a Sydekick's read bar, or a line saying there is no read for the page yet with a Write one button. It produces a one-sentence summary of what Sydekick sees. It depends on Sydekick being available in your workspace.
Treat it as a prompt to look, not as a finding. It is a one-sentence summary and can overweight a single absent prompt. The numbers on the page are the authority.
What you'll see when it's done
- A score and label you can place in a band
- A clear picture of which of the five cards and which engine explains it
- A short list of factors to investigate, kept apart from the score
Watch out for
- A factor is not an input. Only Relevance tracks the score.
- Sources cited is not a count of mentions. It stays above zero at zero visibility.
- Label and colour use different cut-offs. Read both.
- The numbers change with the report. Compare your own location from one report to the next, not with an example.
Where to go next
- How do I read the prompt coverage grid? — the grid behind the score, one prompt and one engine at a time
- How do I generate an AEO report, and what happens next? — when reports are generated, what the cost line means and how the schedule works