Search still has a public scoreboard: rankings, impressions, and clicks. AI recommendations do not. The Scope AI Visibility Index is our working scoreboard for that second channel — how often a business is named when customers ask ChatGPT, Claude, Gemini, Perplexity, and the other engines we monitor.
This page is the hub. It explains what the Index measures, how to read it against industry benchmarks, and how to run the same method on your own business with the free AI visibility checker.
What the Index is (and is not)
The Index is not a Google ranking report. It does not replace Search Console. It answers one question:
When a buyer asks an AI who to hire, buy from, or book, does the model name this business — and who does it name instead?
A high Google rank can still produce a low Index if the model cites directories, review sites, or a competitor’s entity data and never mentions you.
How a score is built
Scope’s visibility score is documented in the data methodology. In short:
- Prompts, not keywords. We store the questions customers actually ask (“best family dentist in Austin that accepts new patients”), not just head terms.
- One check, every included engine. A check evaluates one prompt across the engines on the plan. Starter includes the four core engines; Growth includes the full set.
- Mention quality. Appearing as the recommendation is not the same as a passing mention or a hallucination.
- Citations. We record which sources the model leaned on, because that is usually the fix list.
The Index uses that same method. Industry pages on Scope (dentists, contractors, SaaS, and 100+ verticals) exist so commercial intent and informational guides are not stuffed into one URL.
How to read 2026 benchmarks
Our 2026 industry benchmark report is the numbered companion to this Index. Read it with three rules:
- Compare inside a category. A 45 for a law firm is not the same as a 45 for a restaurant. Category baselines differ because citation graphs differ (Healthgrades vs Yelp vs G2).
- Treat the panel as an optimizing panel. Businesses that already use a GEO tool are not a random census of every shop in America. Use the numbers as a directional baseline, not a census.
- Re-run after you change the entity. Schema, GBP categories, and directory NAP fixes move AI answers. A single score without a second scan is a screenshot, not a trend.
What we publish so others can repeat the work
If you want to reproduce the Index without our product:
- Ask the same prompt on at least four engines, on the same day.
- Record: named / not named, rank in the answer, and cited URLs.
- Keep the prompt text stable for 30 days.
- Split branded prompts (“Is Scope a good AI visibility tool?”) from non-branded (“best AI visibility platform for dentists”).
That last split is the same discipline SEO teams already use in Google Search Console. We will apply it to GA4 and GSC as well once those properties are queried for this program.
Where to go next
- Measure your business: AI visibility checker
- Industry commercial pages: all industries
- Informational guides: Learn
- Category definitions: GEO, AI visibility
- Tool comparisons: Scope vs Profound, Scope vs Peec, Scope vs Scrunch
Q: Is the Index a ranking factor Google uses? A: No. It is Scope’s measurement of generative-engine recommendations. Google Search Console remains the source of truth for Google search.
Q: How often should I refresh an Index score? A: Weekly for competitive local categories. Monthly is enough for slower B2B categories unless you just shipped schema or citation changes.
Q: Can agencies use the Index with clients? A: Yes. Run a baseline scan, pick 10–20 non-branded prompts, and report the score plus citation gaps. That is the retainer artifact — not a keyword spreadsheet alone.