AEO Audit: Is AI Citing Your Business? How to Check
RedHub AI Editorialupdated August 17, 20267 min read
Jump to a section9
TL;DR
- What it is: An AEO audit tests whether AI answer engines — ChatGPT, Claude, Perplexity, Gemini — name your business when buyers ask, and scores how citable your pages are.
- Who it's for: Operators, founders, and agencies who suspect AI is recommending competitors — see the AEO Citation Audit & Optimizer Kit.
- How it works: Build a buyer-intent query set, run it across the four engines, score the answers on a 100-point rubric, fix the weakest signals, then re-audit with the same queries.
- Bottom line: You cannot fix what you have not measured. Baseline first. Fix second. Re-audit third.
What is an AEO audit?
An AEO audit is a structured test of your visibility in AI answers. You run a fixed set of buyer-intent questions through ChatGPT, Claude, Perplexity, and Gemini, record what each engine says, and score the responses on a rubric — whether you were named, how prominently, how accurately, how freshly, and in what tone. The result is a hard number you can improve against, instead of a vague worry that "AI never mentions us."
Best for: anyone who can edit their own site's content and schema — the AEO Citation Audit & Optimizer Kit ($79) ships the full rubric, query generator, schema library, and 30-day fix plan.
When a buyer asks an AI assistant for a recommendation, the engine writes one answer and names a handful of brands. If your business is not in that short list, you never enter the conversation — and you never see it happen, because there was no click to track and no ranking report to read. An AEO audit is how you make that invisible loss visible.
The audit does one job: it turns "are we showing up in AI answers?" from a feeling into a number. Then it tells you which specific signal is dragging that number down, so the fix work is targeted instead of guesswork.
Why your analytics can't see this problem
Traditional search leaves a paper trail. Rankings, impressions, clicks — you can watch a page climb or slip. AI answers leave almost nothing. The buyer asks, the engine answers, the buyer acts. No impression shows up in your dashboard when an engine recommends your competitor instead of you.
That is why most businesses find out the hard way: a prospect mentions "ChatGPT suggested three other companies," or a slow quarter arrives with no obvious cause. The only way to see the problem early is to go ask the engines yourself, the way your buyers do — systematically, with the same questions, on a schedule.
Key insight: most businesses missing from AI answers don't have a content problem. They have an entity problem — the schema, canonical naming, and citation signals that tell an engine what the business is are incomplete or inconsistent. Those are auditable, and most of them are fixable in days, not months.
What an AEO audit actually measures
A useful audit scores more than "did they say our name." Being named last in a list of eight, described wrong, or framed with skepticism are all different problems with different fixes. A five-axis rubric separates them:
| Axis | The question it answers | What a low score points to |
|---|---|---|
| Presence | Does the AI name your brand at all across your query set? | Weak entity signals; engines don't know you exist |
| Prominence | When named, are you first, middle, or buried? | Thin authority; competitors carry stronger citations |
| Accuracy | Does the AI describe your offer and category correctly? | Conflicting or outdated descriptions across the web |
| Freshness | Do citations reflect your current positioning or stale data? | Old pages outweighing new ones; no recent signals |
| Sentiment | Does the AI frame you positively, neutrally, or skeptically? | Review and reputation signals pulling against you |
Score each axis 0–20 and you get a 100-point Citation Index — one number for the whole picture, five sub-scores that name the weak spot. We break the scoring down, with a working calculator, in your AI citation score: five axes, one honest number.
The audit-and-fix loop
The audit is not a one-time report. It is the first lap of a loop: measure, fix, re-measure. Here is the whole cycle at a glance.
- Build the query set. Generate 30–50 buyer-intent questions for your niche across four funnel stages — discovery, comparison, decision, post-purchase. These are the questions your buyers actually ask an assistant.
- Run the baseline. Put every query through ChatGPT, Claude, Perplexity, and Gemini. Save the responses as evidence — answers shift day to day, so the saved day-1 responses are your proof of where you started.
- Score it. Rate the responses on the five axes and record your day-1 Citation Index. Now you know which axis is dragging you down — and that axis picks your fix order.
- Fix the signals. Foundation first (entity hygiene and schema — it moves the needle most), then answer-first content, then amplification: freshness, reviews, citation sources.
- Re-audit at day 30. Run the identical query set again and re-score with the same rubric. Compare like-for-like. The difference is your measured lift — not a vibe.
Each stage of that loop has its own guide below. The step-by-step baseline run is in the AI citation audit checklist. The schema fixes are in schema markup for AI search. The full fix sequence is in the 30-day AEO plan.
The honest frame: answer engines are non-deterministic. The same question can surface different brands on different days. No audit — and no tool — can guarantee a citation. What an audit does is measurable: it improves the signals you control and gives you a repeatable way to see whether they moved. Anyone promising more than that is selling fiction.
Audit first, strategy second
An AEO audit is the execution layer: test, score, fix, re-test. It sits underneath a broader question — how AI visibility fits your whole marketing motion, which content to build for it, and how it relates to your SEO investment. That strategy layer is its own discipline. If you want it, the GEO / AI Visibility Playbook covers the full strategic framework, and our generative engine optimization pillar is the companion read.
But strategy without a baseline is planning in the dark. Whichever direction you take it, run the audit first — it is 90 minutes, and every decision after it gets easier.
Run your baseline audit this week
The AEO Citation Audit & Optimizer Kit ($79, one-time) ships the query generator, the 100-point five-axis rubric, 12 paste-ready JSON-LD schema templates with 8 answer-first content patterns, and the 30-day playbook — ending in the day-30 re-audit that measures your actual lift. Single-operator license, unlimited projects, 30-day refund.
Get the AEO Audit Kit — $79 →Where to go deeper
This pillar is the map. Each stage of the loop has its own guide:
- The AI citation audit checklist — the 90-minute baseline run, step by step, across all four engines.
- Your AI citation score — how the five-axis, 100-point rubric works, with a calculator to score yourself.
- Schema markup for AI search — the entity signals engines actually read, and the audit-first order to fix them.
- The 30-day AEO plan — Foundation, Content, Amplification: one task a day from baseline to re-audit.
Decision Guide
Run an AEO audit if: buyers in your market ask AI assistants for recommendations, you can edit your own site's content and schema, and you want a number instead of a worry.
Skip it if: your customers genuinely never use AI assistants to find businesses like yours, or you can't change your own website.
Best first step: pick your ten most valuable buyer questions and ask ChatGPT and Perplexity today. If your name doesn't come up, run the full audit.
FAQ
What is an AEO audit?
It's a structured test of whether AI answer engines name your business. You run a fixed set of buyer-intent queries through ChatGPT, Claude, Perplexity, and Gemini, then score the responses on five axes — Presence, Prominence, Accuracy, Freshness, and Sentiment — for a 100-point Citation Index.
How long does an AEO audit take?
About 90 minutes for a full baseline: generate the query set, run it across the four engines, save the responses, and score them. The fix work afterward is a separate 30-day effort at roughly 30–60 minutes a day.
Do I need paid AI subscriptions to run it?
Usually not. The free tiers of ChatGPT, Claude, Perplexity, and Gemini are typically enough for a baseline audit. You bring the engines; the audit kit brings the queries, rubric, and playbook.
Is an AEO audit the same as an SEO audit?
No. An SEO audit measures where your pages rank in a list of links. An AEO audit measures whether engines name your brand inside a generated answer — a different surface with different signals, like entity schema and answer-first structure. They complement each other; neither replaces the other.
Can an audit guarantee AI will cite my business?
No, and be wary of anyone who says otherwise. Answer engines are non-deterministic — the same query can return different brands on different days. An audit improves the signals you control and measures whether they moved. It raises the odds; it cannot promise the citation.
How often should I re-run the audit?
Re-audit at day 30 with the identical query set, so you compare like-for-like against your baseline. After that, a monthly or quarterly re-run is enough to catch drift — the discipline matters more than the frequency.
What tool do I need to run this?
The AEO Citation Audit & Optimizer Kit ($79) contains all four pieces: the query generator, the 100-point scoring rubric, the 12-template JSON-LD schema library with 8 content patterns, and the 30-day playbook. For the broader strategy layer, pair it with the GEO / AI Visibility Playbook.
Stop guessing whether AI is recommending you
Run the audit, get your Citation Index, work the 30-day playbook, and re-audit to measure the lift. $79 once — usable on every business you ever touch.
Get the AEO Citation Audit & Optimizer Kit →