Articles on: Getting Started

AI Analysis — AI Presence Score

The AI Analysis section measures your brand's ability to be recognized, associated with the right domain, and recommended by artificial intelligence models (ChatGPT, Gemini, etc.). It shows you where you stand and what to do to improve your AI visibility.


The AI presence score

A score from 0 to 100 provides an overall assessment of your visibility in AI responses. It is calculated from six weighted dimensions and updated with each analysis.

Level

Score

Meaning

Critical

0–30

Very low AI visibility, foundations to fix as a priority

Weak

31–50

Foundations present but still insufficient

Fair

51–70

Workable base with clear optimizations

Good

71–85

Good level, proofs and recommendations need strengthening

Excellent

86–100

Very strong AI visibility with a solid and lasting dynamic


Click "Score details" to see the full breakdown by dimension, with the weight and contribution of each one.


How is the score calculated?

Six dimensions are scored out of 10, then weighted to form the final score:

  • Web findability (25%) — the AI agent finds the brand and can respond with sources
  • Domain association (20%) — the agent associates the brand with the right market
  • Spontaneous recommendation (20%) — the brand comes up when not mentioned in the question
  • Assisted recommendation (15%) — the agent can recommend the brand when named
  • Competitive comparison (15%) — the brand can be positioned against alternatives
  • Citable content (5%) — AI models find usable text on the website


The diagnostic

A summary text generated by the AI agent describes the overall perception of your brand by AI models: how they describe it, which domain they associate it with, and what recommendations they make spontaneously. Click "See more" to read the full diagnostic.


AI visibility levers

Nine levers are analyzed individually, each scored and rated (Excellent, Good, Fair, Weak):

  • Brand clarity — does the AI understand what your brand does and which domain to associate it with?
  • Website & web proof — is your site detected and linked to the right domain?
  • Web coverage — how many independent web sources mention your brand?
  • Recommendation potential — is your brand featured in comparisons?
  • Trust signals — do independent external proofs strengthen your credibility?
  • Market positioning — is your competitive positioning clear to AI models?
  • Recommendation risk — are negative signals holding back your recommendation?
  • Supplier consistency — do your partners and suppliers reinforce your image?
  • Citable content — does your site contain short, well-structured answers that AI models can easily quote?

Click "View details" on each lever to access a detailed sheet including: what is working, what is limiting the score, and the recommended priority action.


Why this score — the test table

The "Why this score" section lists the 14 tests actually sent to AI models during the analysis. For each test, you can see:

  • The test type (brand awareness, domain association, spontaneous recommendation, market comparison, independent proof, public credibility, recommendation barriers…)
  • The question sent to the AI model
  • The status: Present, Partial, or Absent
  • The number of web sources returned
  • The score obtained

Expand each test to see in detail: the question sent, the AI response returned (GPT / Gemini), the interpreted result, and the sources retrieved.


The AI agent automatically generates action plans ranked by impact, each presenting:

  • The detected issue
  • Why it matters for your AI visibility
  • A concrete action plan
  • The estimated effort and suggested owner (e.g. Marketing)


Take action — Turn analysis into community actions

The analysis identifies missing signals. The Studio lets you address them directly by activating your community:

  • Create visible proof — launch quests to collect reviews, testimonials, use cases, or result screenshots from your members
  • Increase interactions — activate your members with simple missions: comment, reply, share an experience, or relay an announcement
  • Feed AI visibility — turn the best feedback into FAQs, posts, proof pages, client case studies, or content that AI engines can cite

Private interactions are mainly useful for steering. To strengthen web and AI visibility, also prioritize public or citable proof: social posts, reviews, FAQ pages, testimonials, and client case studies.


Scan status
  • Status — Synced or pending
  • Last updated — date and time of the last analysis Frequency — on demand (manual launch from the dashboard)
  • Analysis reliability — confidence indicator on the collected data (e.g. 100% reliable)

Updated on: 09/07/2026

Was this article helpful?

Share your feedback

Cancel

Thank you!