What Is an AEO Audit? How We Score Your AI Visibility from 0 to 100

You might be the best dentist in Munich. The highest-rated contractor in Austin. The most awarded law firm in Geneva. But if you ask ChatGPT to recommend a business in your industry and your city, your name might never come up.

That gap -- between your actual reputation and what AI engines know about you -- is exactly what an AEO audit measures.

An AEO audit is the first step toward understanding your AI visibility. It answers a simple but critical question: when people ask AI for recommendations in your space, does AI recommend you?

This article explains exactly how we run AEO audits, how the 0-100 scoring system works, and what you can do with the results.

What an AEO Audit Actually Is

An AEO audit is a systematic test of your business's visibility across AI-powered answer engines. We query the major AI platforms with the kinds of questions your potential customers actually ask, and we document whether your business gets mentioned, recommended, or ignored.

It is not a website audit. It is not an SEO check. It is not a review of your Google rankings. Those are separate disciplines with separate tools.

An AEO audit focuses exclusively on what happens when someone asks an AI engine about your industry, your services, and your location. The output is a clear picture of where you stand in the AI recommendation landscape -- and a score that quantifies it.

What an AEO audit is not

Let's clear up a few common misconceptions:

• It is not an SEO audit. You can rank #1 on Google and score 0 on an AEO audit. We have seen this dozens of times. AI engines do not simply re-rank Google results.
• It is not a reputation audit. A low score does not mean your business has a bad reputation. It means AI engines lack the structured information they need to recommend you.
• It is not a one-time report. A single audit gives you a snapshot. Real value comes from tracking your score over time, which is why we recommend monthly audits at minimum.

The Five AI Engines We Test

AI is not one monolithic system. Different engines serve different audiences, pull from different data sources, and weight different trust signals. Testing only one gives you a dangerously incomplete picture.

We run every audit across five major AI answer engines:

• ChatGPT (OpenAI) -- The dominant player with 800+ million weekly users. Pulls from training data, web browsing, and plugin integrations.
• Gemini (Google) -- Deeply integrated with Google Search, Maps, and Business Profiles. Particularly influential for local business recommendations.
• Perplexity -- The research-oriented AI search engine that cites its sources, providing transparent attribution you can trace.
• Claude (Anthropic) -- Weighs authoritative, well-structured content heavily. A good barometer for content quality signals.
• Grok (xAI) -- Integrated with X, with access to real-time social signals. Surfaces businesses with strong social media presence.

Each engine has its own biases and blind spots. A business might score well on Gemini but poorly on ChatGPT. The audit captures these differences so you know where to focus.

How We Design the Queries

The queries we run are not generic. Asking "best dentist" tells you nothing useful. Real customers ask specific questions, and AI engines give different answers depending on how the question is phrased.

Industry-specific queries

We build query sets tailored to your exact industry and service offerings. For a dental practice, that might include "Best cosmetic dentist in [city]," "Who should I go to for dental implants in [city]?," and "Which dentist in [city] specializes in anxious patients?" We typically run 20 to 40 queries per audit, covering core services, specializations, and competitive positioning.

Location-specific queries

We test visibility at multiple geographic levels -- city ("best [service] in Munich"), neighborhood ("best [service] in Schwabing"), regional ("best [service] in Bavaria"), and comparative ("best [service] in Munich vs Berlin"). AI engines often give completely different recommendations depending on geographic specificity.

The multilingual dimension

This is one of the most overlooked factors in AI visibility. AI engines do not only search in English. A potential client in Geneva might ask in French. A prospect in Munich asks in German. A tourist in Barcelona asks in English, but a local asks in Catalan or Spanish.

We run queries in every language relevant to your market. For a business in Zurich, that might mean testing in German, French, Italian, and English. For a business in Dubai, we test in English and Arabic.

The results are often surprising. A business that appears consistently in English queries might be completely invisible in the local language -- the language most of their actual customers use. Multilingual testing is not optional. It is core to understanding real-world AI visibility.

The Scoring Methodology: 0 to 100

The AI visibility score is not a subjective rating. It is calculated from measurable data points across every query and every engine we test.

Here is how it works:

For each query on each engine, we record whether your business is:

• Mentioned by name as a recommendation (full credit)
• Referenced indirectly -- for example, described but not named (partial credit)
• Absent entirely from the response (zero credit)

The raw score is the weighted percentage of total possible mentions across all queries and all engines. We weight primary recommendations more heavily than secondary mentions, and we weight high-intent queries (like "who should I hire") more heavily than informational queries (like "what does this service cost").

The result is a single number from 0 to 100 that represents your overall AI visibility.

Score Ranges: What Your Number Means

Not every score tells the same story. Here is what each range means in practical terms, along with anonymized examples from real audits.

0-20: Invisible

What it means: AI engines do not know you exist, or they do not have enough information to recommend you. When potential customers ask AI about your industry in your city, your business never comes up.

Real example: An award-winning architecture firm in Turin with 15 years of experience scored 4/100. Despite features in Architectural Digest, none of the AI engines had enough structured data to connect the firm to relevant queries. Beautiful website, but minimal directory presence and no schema markup.

How common: Roughly 60% of businesses fall here on first audit. It does not reflect quality -- it reflects data gaps.

20-40: Emerging

What it means: You appear occasionally, usually on one or two engines for a narrow set of queries. AI has some awareness of your business but does not consistently recommend you.

Real example: A physiotherapy clinic in Lausanne scored 28/100. It appeared on Gemini (thanks to a strong Google Business Profile) but was invisible on ChatGPT, Claude, and Perplexity. It also disappeared entirely for French-language queries, despite serving a primarily French-speaking clientele.

How common: About 25% of businesses land here. They typically have one strong signal but lack the cross-platform consistency AI engines need.

40-60: Competitive

What it means: You appear regularly across multiple engines and multiple query types. AI considers you a legitimate option in your market. You are not dominant, but you are in the conversation.

Real example: A cosmetic dentist in Beverly Hills scored 52/100. The practice appeared on ChatGPT and Gemini for most queries and showed up on Perplexity with citations from Healthgrades and Yelp, but was missing from Claude and Grok. It performed well in English but scored lower for Spanish-language queries -- a significant gap given local demographics.

How common: About 10% of businesses reach this range, typically those with strong review profiles and some structured data.

60-80: Strong

What it means: AI engines consistently recommend you across most engines and most query types. You are a top recommendation in your market, though there may be specific query categories or languages where gaps remain.

Real example: A luxury real estate agency in Dubai scored 71/100. It appeared prominently on four of five engines, consistently placed in the top three recommendations, and performed well in both English and Arabic. The remaining gaps were niche queries where more specialized competitors appeared instead.

How common: Fewer than 5% of businesses reach this level -- typically brands with deliberate AEO strategies and strong citation networks.

80-100: Dominant

What it means: AI engines recommend you for nearly every relevant query across all engines and all tested languages. You own the AI conversation in your market.

Real example: We have not yet audited a business that scored above 85/100 on an initial audit. This range requires sustained optimization across every AI trust signal over three to six months minimum.

How common: Extremely rare. This is the goal, not the starting point.

What Factors Affect Your Score

Through hundreds of audits, we have identified six factors that matter most:

• Structured data and schema markup. Businesses with comprehensive JSON-LD markup (LocalBusiness, FAQ, Service schema) consistently score higher. This is often the single biggest lever.
• Third-party citations. Mentions in industry directories, review platforms, and publications act as trust signals. A single mention in a respected source can move your score significantly.
• Review volume and quality. AI engines read the actual text of reviews, not just star ratings. Detailed reviews that mention services by name carry more weight than generic praise.
• Content depth and structure. FAQ pages, detailed service descriptions, and educational content that directly answers common questions all contribute.
• Cross-platform consistency. Consistent name, address, services, and positioning across your website, directories, and social profiles help AI engines confidently identify your business as one entity.
• Multilingual presence. If your market includes non-English speakers, content and citations in those languages are essential. AI engines treat each language as a somewhat separate information space.

How to Interpret Competitor Comparisons

An AEO audit is more valuable in context. That is why every full audit includes a competitor benchmark -- we run the same queries for your top competitors and score them using the same methodology.

Three things to watch for:

• Relative gaps matter more than absolute scores. If you score 25/100 and your closest competitor scores 22/100, you are in a tight race. If they score 55/100, you have a significant disadvantage that is costing you leads right now.
• Engine-specific differences reveal opportunities. If a competitor dominates on Gemini but is weak on ChatGPT, you can leapfrog them on specific engines with targeted optimization.
• Query-specific gaps are actionable. If a competitor appears for "best [service] in [city]" but you appear for "affordable [service] in [city]," that tells you how AI engines perceive your positioning versus theirs.

The competitor comparison is not about vanity. It is about strategy -- showing you exactly where the opportunities are.

What Happens After the Audit

The audit gives you a diagnosis. What matters next is the treatment plan. After a full AEO audit, you receive:

• Your score and breakdown -- the overall number, plus per-engine and per-query-category scores showing exactly where you are strong and where you are invisible.
• Competitor benchmarks -- how you compare to the top competitors in your market across the same queries and engines.
• Priority recommendations -- a ranked list of actions specific to your situation. If structured data is your biggest gap, that goes first. If it is missing citations in French, that goes first.
• A monitoring baseline -- your initial score becomes the benchmark for tracking monthly progress.

Most businesses that take action on audit recommendations see measurable score improvements within four to eight weeks. The fastest gains come from structured data implementation and citation building, because those changes directly feed the information AI engines use to form recommendations.

Start With the Free Mini Audit

You do not need to commit to anything to find out where you stand. The free mini audit on aeomedia.ai runs your business through multiple AI engines and shows you whether they recommend you. It takes less than a minute.

If the mini audit shows gaps -- and for most businesses, it will -- that is the starting point for a full audit with competitor benchmarks and actionable recommendations.

Try the free mini audit at aeomedia.ai right now. Find out what AI engines say about your business before your competitors do.