Measuring AI visibility

How to measure and improve AI search visibility for your website.

Instly Scout uses two instruments to measure AI visibility: an 8-dimension AI readiness audit (0-100) and an AI answer check that runs your buyer questions across ChatGPT, Perplexity, Gemini and Claude (named in X of 20). Most tools measure only readiness. Both numbers are needed because they measure different things.

Measure your site The 8-dimension framework
Two instruments

Why one score is not enough.

A site can be technically perfect and invisible to AI engines at the same time. These are our own numbers, published so the gap is visible.

Instrument 1: AI readiness score
90/100

Scout Full report, 29 Sep 2026. 8 technical dimensions: AI visibility, buyer intent, trust, SEO/AEO, technology, security, answer coverage, conversion path.

Tells you: CAN the engine cite you?

≠
Instrument 2: named in AI answers
0/20

AI answer check, 28 Sep 2026. 5 buyer questions across 4 engines. instlyscout.com named in 0 of 20 answers.

Tells you: DOES the engine cite you?

These are instlyscout.com's own numbers. We publish them as our before state. The work is to close the gap.

The 8 dimensions

What the AI readiness score measures.

Instly Scout's readiness audit covers 8 dimensions, each contributing to the 0-100 score. A score below 70 on any dimension is a specific fix that can be named, costed and tested.

AI VisibilityCan machines resolve who you are and extract your claims?
Schema, entity clarity, structured data
Buyer IntentAre cost, comparison, eligibility answered on-page?
Buyer question coverage
Trust and CredibilityIs proof machine-verifiable?
Reviews, cases, credentials with schema
SEO and AEODoes structure serve ranked search and answer extraction?
Headings, FAQs, crawlability
Technology StackDoes the tech layer help or hide content?
Rendering, JS reliance, sitemap
Security and PrivacyDo browser-visible signals undermine trust?
HTTPS, privacy policy, cookie handling
Answer CoverageDo pages answer buyers' questions?
FAQ and answer blocks, FAQPage markup
Conversion PathIs the next step clear once a buyer arrives?
One clear call to action per key page
What actually improves the number

Four moves that improve AI visibility.

Ordered by impact on the named-in-answers count, based on what we have measured across client sites.

1. Publish one page per buyer question

Each question your buyers ask AI engines needs one page on your site that answers it directly, with your brand name in the first sentence. A homepage that mentions all of this in passing is not the same as a dedicated answer page. AI engines quote specific pages, not whole sites.

2. Allow all AI crawlers

GPTBot (ChatGPT), PerplexityBot, ClaudeBot and Google-Extended (Gemini) must be allowed in robots.txt. A training-crawler block is a warning in Instly Scout's full report. If they cannot read the site, they cannot cite it.

3. Add pages to sitemap and llms.txt

Every answer page must appear in sitemap.xml (with a real lastmod date) and in llms.txt. llms.txt is the AI-native discovery file that tells engines what your site covers and where the key pages are.

4. Get into the round-up articles AI reads

82% of the sources read in our own AI answer baseline (28 Sep 2026) were independent articles, not vendor sites. Being listed in "best [category] tools" articles on sites AI engines already trust is the highest-authority signal. This is off-site work: pitching article authors directly.

Get Instly Scout to measure your site
Frequently asked

AI search visibility: questions answered.

What is AI search visibility?

AI search visibility is whether AI answer engines name your brand when buyers ask questions relevant to your product. It is measured as a count: named in X of 20 answers (5 questions across ChatGPT, Perplexity, Gemini and Claude). It is separate from and not predicted by technical readiness scores.

Can I improve AI visibility without touching the site's code?

Yes. The main levers are content (publishing answer pages), robots.txt (allowing AI crawlers), and sitemap/llms.txt (listing those pages). robots.txt and sitemap.xml are usually edited without touching the core codebase. Off-site, getting listed in round-up articles requires no site change at all.

How does Instly Scout produce a before-and-after proof?

Instly Scout saves the baseline (named in X of 20 answers, with a date and the source list from each engine) before changes are made. After changes go live and have been indexed, a second run is saved. The two are compared in a dated proof document. Without a saved baseline, AI answer checks overwrite each other and no comparison is possible.