← Back to blog

What is an AEO audit? What it checks and can ChatGPT do it?

Search used to give you a list of ten links. Now a buyer asks ChatGPT or Perplexity a question and gets one written answer with three or four brands in it. If your brand is missing from that answer, no amount of ranking data will show you why.

An AEO audit fills that gap. It is the starting point for answer engine optimization, and it is what we deliver in our AI Visibility Audit. This post explains what the audit checks, what you get at the end, and where a general chatbot can and cannot help.

What is an AEO audit?

An AEO audit is a structured test of how AI assistants describe your brand when buyers ask questions in your category. It measures whether you are named, whether you are cited as a source, whether the facts are right, and what stops you from showing up more often.

A classic SEO audit looks at your website: crawl errors, page speed, keywords and backlinks. An AEO audit starts from the answers instead. It asks: when a buyer types the question that should lead to you, what does the AI say, and who does it trust? Then it works back to your site, your content and the third-party sources the engines lean on.

If you are new to the term, our post what is AEO covers the basics. The short version is that SEO wins the click and AEO wins the answer.

The 6 checks in an AEO audit

Every audit we run covers the same six checks. You can use this list to judge any audit you are offered.

  1. Prompt set. We build a list of the questions your buyers actually ask, grouped by stage: category questions ("best payroll software for startups"), comparison questions ("X vs Y"), and problem questions ("how do I reduce failed payments"). The prompts come from keyword data, sales calls and People also ask results, not from guesswork.
  2. Share of answers. We run each prompt through the engines that matter to you, usually ChatGPT, Perplexity, Google AI Overviews and Gemini, and record which brands appear. Your share of answers is the percentage of prompts where you are named, compared with each competitor.
  3. Cited sources. AI answers lean on sources. We log every page each engine cites: your site, competitors, review sites, Reddit threads, directories, news and Wikipedia. This shows which third-party pages shape the answer in your category.
  4. Entity accuracy. We check what the engines say about you when they do mention you. Wrong pricing, an old product name, a missing market or a confused competitor are common. Errors here can cost more than not being mentioned at all.
  5. Technical access. We check whether AI crawlers such as GPTBot, OAI-SearchBot, PerplexityBot and ClaudeBot can reach your pages, whether key content renders without JavaScript, and whether your structured data is valid. Our technical AEO page covers this layer in depth.
  6. Content gaps. Finally, we map which prompts you lose and why. Usually the cause is one of three things: there is no page that answers the question directly, the answer exists but is buried, or the engines trust third-party sources that do not mention you.

What you get at the end

The output is a report, not a dashboard login. It includes the prompt set, share of answers by engine and by competitor, the top cited sources in your category, a list of factual errors with the source of each, technical issues, and a ranked fix list with effort and expected impact.

[PROOF NEEDED: sample report screenshot, anonymised, with caption]

Turnaround: [PROOF NEEDED: audit turnaround time in working days]. Price: see AEO pricing [PRICING NEEDED].

What is an AEO assessment?

An AEO assessment is usually a lighter version of an audit: a quick read on whether AI assistants mention your brand at all, based on a small set of prompts. It tells you if there is a problem. A full audit tells you why, and what to fix first.

Agencies and tools use the two terms loosely, so ask what is included before you buy. Here is the practical difference.

An assessment is a fair first step if you only want to know whether AI search is a problem for you yet. Our free AI visibility checker is built for that: you enter your brand and we email your score.

An audit is the right step when you already suspect a gap, when a competitor keeps showing up in AI answers, or when you need a plan you can hand to your content and dev teams.

How to tell if an audit is real

  • It names the exact prompts it tested, and the date each was run. AI answers change, so undated results are hard to trust.
  • It covers more than one engine. ChatGPT, Perplexity and Google AI Overviews often cite different sources for the same question.
  • It shows competitors on the same prompts, so you can see share of answers, not a score in isolation.
  • It explains the cause of each gap, and ties every fix to a prompt you are losing.

We explain our own method in detail on how we measure AI visibility.

Can ChatGPT do an SEO audit?

ChatGPT can help with parts of an SEO or AEO audit, but it cannot run a reliable one on its own. It can review a page you paste in, suggest schema, or spot thin content. It cannot measure how often you appear across engines, track changes over time, or see what other users are shown.

What ChatGPT does well in an audit

  • Reviewing a single page for clarity, missing answers and weak headings.
  • Drafting FAQ questions from a topic, which you then check against real search data.
  • Writing or checking JSON-LD for Organization, FAQPage or Product schema. Our free schema generator does this without a chatbot.
  • Summarising a crawl export or a list of errors into plain English.

Where it falls short

  • One answer is not a measurement. Asking ChatGPT "who are the best agencies for X" once tells you what it said to you, that time. Answers vary by wording, location, account history and model version. An audit runs each prompt repeatedly and records the spread.
  • It cannot see other engines. ChatGPT cannot tell you what Perplexity or Google AI Overviews say about you.
  • It is not a neutral judge of itself. It can guess why it cites a source, but it cannot show you its ranking logic. The cited-source log in a real audit gives you evidence instead of a guess.
  • It can be wrong about you. If its facts about your brand are out of date, it will audit you against those facts.

So use ChatGPT as a helper inside the audit. Do not use it as the audit.

When to get an AEO audit

Get an audit when AI search starts to show up in your pipeline: prospects say they "asked ChatGPT", a competitor keeps appearing in AI answers, or organic traffic drops while rankings hold. It is also the right first step before you sign any AEO retainer, because it sets the baseline you will measure progress against. Our post on AEO audit vs retainer covers that choice.

For an example of what we found and fixed for a client, see [PROOF NEEDED: Dhan Saarthi case study results, with disclosure] on our case study page.

Ready to see where you stand? Book an AI Visibility Audit. We run your buyers' prompts across the main engines, show you who wins each answer, and hand you a ranked list of fixes.

Need GTM creative shipped at AI speed?

One team for content, decks, landing pages, SEO/GEO, and video.

Get started