AI visibility audit

AI visibility audit for your brand

An AI visibility audit shows what ChatGPT, Perplexity, Gemini and Claude say about your brand, whether they searched the web or answered from memory, and which sources shaped the answer.

  • 6 AI models
  • Web and memory
  • Fixes as tasks

What it checks

What an AI visibility audit checks

Two parts: how often AI names you on the questions buyers ask, and what it says when it does.

  • Named on buyer questions How often each AI engine names you for the questions your buyers ask. Your first visibility report covers this, free.
  • Which models mention you 6 models, asked who you are and what your reputation is.
  • What they say Key themes and sentiment, model by model.
  • Misleading sources Third-party pages that spread wrong information about you, and inconsistencies on your own pages.
  • Known or only found Whether a model knows you from training or only finds you with web search.
  • The sources behind it Every analyzed third-party page, with an action: request a link, a mention or an update.
  • Negative sources The pages pulling your sentiment down.
  • Your own site How often AI cites your pages when it describes you, and which of your registered names it used.
What's Included

What we analyze
about your brand

AI Perception

AI perception analysis

See exactly how AI systems describe and understand your brand across platforms.

  • ChatGPT "adidas makes performance running shoes, from the Adizero racing line to Supernova daily trainers" Positive
  • Perplexity "adidas is a sportswear brand. I am not sure which of its running shoes suit high mileage" Neutral
  • Gemini "Runners pick the adidas Ultraboost for cushioning and the Adizero line for race day" Positive
Example audit

Your brand at a glance

The numbers each audit reports, across the six models.

4 / 6
Models that mention you
Positive
Overall sentiment
3
Negative sources
18%
Self-citation rate
Cited pages

What to do with each source

Third-party pages the audit analyzed, how accurately each one describes you, and the action it suggests.

  • runnersworld.com Accuracy 88 Request link
  • runrepeat.com Accuracy 41 Request update
  • reddit.com Accuracy 67 Monitor
Over time

Compare audits

Re-run the audit and compare it with the last one, or with another country.

  • Mentioned 83%
  • Positive 71%
  • Self-citation 18%

Has AI learned you, or only found you?

We ask with web search and without it, ChatGPT and Gemini both ways, so you know whether a model has learned your brand or only found it.

ChatGPT With web search

adidas Supernova Rise 2 is a daily trainer for high mileage, positioned against the Brooks Ghost 17 on durability and the ASICS Gel-Nimbus 27 on weight.

Named, three sources read

ChatGPT From memory, no search

For high mileage I would usually name the Brooks Ghost 17 and the ASICS Gel-Nimbus 27. I am not certain which adidas shoe fits.

Not named, nothing retrieved

Named only with search means the model has not learned you yet. Missing with search means your pages are not being retrieved. The two need different work.

How to run one

How to run an AI visibility audit

Four steps in Rank Prompt, from setup to fixes.

  1. 01

    Set your brand facts

    Rank Prompt reads your site and fills them. The audit scores other pages against them.

  2. 02

    Run your first report

    Free. It asks AI engines your buyers' questions and shows how often you are named.

  3. 03

    Run the Brand Audit

    From 10 credits. 2 prompts on 6 models, for the country or city you pick.

  4. 04

    Fix by cause, then re-run

    Turn each finding into a task. Re-run in 2 to 4 weeks and compare.

Before & After

See the impact of optimization

How brand perception changes after implementing audit recommendations

Before optimization 40%
  • ChatGPT 35
  • Perplexity 52
  • Claude 28
  • Gemini 45
  • Missing product facts AI knows the Ultraboost but not the current Adizero and Supernova running shoes
  • Wrong classification AI files adidas running shoes under lifestyle sneakers, not performance running
  • Your site never cited AI never references adidas.com when it describes the running range
After optimization 82%
  • ChatGPT 88
  • Perplexity 91
  • Claude 76
  • Gemini 73
  • Complete product knowledge The Adizero, Supernova and Ultraboost lines, as adidas.com describes them
  • Correct categorization AI describes adidas as a performance running brand for racing and daily training
  • Your own pages cited AI now cites adidas.com product pages and running guides as sources

FAQ

AI visibility audit FAQ

Still have questions? Contact our team

What is an AI visibility audit?

An AI visibility audit checks what AI assistants say about your brand and where that comes from. Rank Prompt's Brand Audit asks 6 models 2 questions each, one about who you are and one about your reputation, then reports which models mention you, the sentiment, the sources they cite, and the pages that describe you inaccurately.

Which AI models does the audit test?

Three answer with web search: ChatGPT, Perplexity and Gemini. Three answer from what they learned in training: ChatGPT, Claude and Gemini. Comparing the two shows whether a model already knows your brand or only finds it when it searches.

Is an LLM visibility audit the same thing?

Yes. LLMs are the large language models behind ChatGPT, Claude, Gemini and Perplexity, so an LLM visibility audit and an AI visibility audit check the same answers.

Is there a free AI visibility audit?

Your first visibility report is free: it runs the questions your buyers ask across AI engines and shows how often you are named. The Brand Audit costs from 10 credits on Starter and up. For a quick check of one question with no signup, use the free AI visibility checker.

What does the audit report?

Which models mention you, overall sentiment, key themes, negative sources, how often AI cites your own site, which of your registered names it used, and which models started or stopped mentioning you since the last audit. Analyze the cited pages and each third-party page gets an accuracy score and an action (request a link, a mention or an update, or keep monitoring), while your own pages get their consistency issues and missing elements.

How is it different from an SEO audit?

An SEO audit checks your own site: crawling, speed and markup. An AI visibility audit checks the answers AI gives about you, which lean on other sites as much as yours. Rank Prompt runs both, and its SEO audit also checks whether AI crawlers can reach your pages.

How do I fix what it finds?

Generate the action plan and each recommendation lands on your Tasks board as a suggested task, with its priority. Inaccurate third-party pages: request an update. Consistency issues on your own pages: edit them and your brand facts. Negative review sites: respond and earn new reviews. Models that do not know you from memory: work on the pages web-search models read, since training catches up in months.

How often should I run an AI visibility audit?

Monthly as a routine, 2 to 4 weeks after major content changes, 1 to 2 weeks after press coverage, and right away after an incident. If you sell in several countries, run it per country.

Can I audit different markets?

Yes. Pick a country or city for each audit, then compare audits of the same market over time.

Can I share audit results?

Yes. Every Brand Audit can be shared via a public link with a read-only view, so stakeholders or clients see the findings without a Rank Prompt account. On Agency plans the shared view carries your white-label branding.

What are Brand Facts?

The key facts about your company: name, category, products, markets and what sets you apart. Rank Prompt reads your site to fill them, and the audit uses them to judge how accurately other pages describe you and how closely AI matches your own description.

Start your free 7-day trial

Run your AI visibility audit

See what AI says about your brand, and what to fix first.

Models asked
6
Prompts per model
2
Credits to run the audit
10