If you want to know why your competitors keep showing up when someone asks ChatGPT to recommend a tool in your space — and you're not — you don't need an expensive platform to find out. You need a systematic process, a few browser tabs, and about two hours. Here's exactly how to do a thorough competitor AEO audit without spending a cent.
What a Competitor AEO Audit Actually Tells You
Before jumping into steps, let's be clear about what you're measuring. An AEO audit maps which brands get cited by AI answer engines (ChatGPT, Claude, Perplexity, Gemini) when users ask questions relevant to your market. It tells you:
- Which competitors are consistently named and recommended
- What question types trigger those citations
- What content formats and sources AI engines are pulling from
- Where the gaps are that you could fill
This is different from a traditional SEO audit. You're not just tracking rankings — you're tracking narrative authority. AI engines don't just index pages; they form opinions about which sources and brands are trustworthy and relevant. Understanding how your competitors earned that trust is the whole game.
Step 1: Build Your Competitor List and Question Map
Start with a clean spreadsheet. List your top five to eight competitors in column A. Then, in column B, write out the ten to fifteen questions a potential customer might ask an AI assistant when they're looking for what you sell.
Think about question intent in three layers:
Discovery questions — "What's the best [product category] for [use case]?" Comparison questions — "What's the difference between [Competitor A] and [Competitor B]?" Validation questions — "Is [Competitor Name] worth it?" or "What are the downsides of [Competitor Name]?"
Be specific. If you sell project management software for construction firms, your question map shouldn't say "best project management tool." It should say "best project management software for small construction companies" and "what do general contractors use to track job site tasks."
The more specific your questions, the more useful your data. Vague questions produce generic AI answers. Specific questions reveal who actually owns authority in your niche.
Step 2: Run Your Prompts Across Multiple AI Engines
Open four tabs: ChatGPT (using GPT-4o), Claude (Sonnet or Opus), Perplexity, and Google's AI Overview (just run a search in Google and look at the AI-generated summary at the top).
Run every question from your map in each engine. Here's the important part: record the raw output verbatim, not just a summary. Paste it into your spreadsheet or a notes doc. You'll be comparing patterns, and small wording differences matter.
For each response, note:
- Which brands are named
- What order they appear in
- Whether the AI gives a recommendation or just lists options
- What specific claims are made about each brand
- Whether a source URL is cited (Perplexity is especially useful here — it shows its sources)
Do this for all your questions across all four engines. Yes, it takes time. But this raw dataset is genuinely valuable and most of your competitors haven't done it.
A practical tip: In ChatGPT and Claude, preface your question with "Please answer as if you're advising a [specific persona, e.g., small business owner who isn't very technical]." This surfaces more opinionated, citation-rich answers than bare questions.
Step 3: Build Your Citation Frequency Map
Now turn your raw data into a simple frequency count. For each competitor, count how many times they were cited across all your questions and all engines. Create a table that looks something like this:
| Competitor | Total Citations | ChatGPT | Claude | Perplexity | Google AI |
|---|---|---|---|---|---|
| Competitor A | 38 | 11 | 9 | 12 | 6 |
| Competitor B | 22 | 6 | 7 | 5 | 4 |
| Your Brand | 4 | 1 | 2 | 1 | 0 |
This is your visibility gap in black and white. It's not always comfortable to look at, but it's the most honest starting point you'll find.
Pay special attention to which engine is doing the most citing. Perplexity tends to favor brands with strong, recently updated web content. ChatGPT and Claude lean on training data and tend to favor brands with broad third-party mentions — reviews, press, community discussions. Google AI Overviews favor brands with strong traditional SEO signals. The engine breakdown tells you where to focus your fix.
Step 4: Diagnose Why the Top-Cited Competitors Are Winning
This is where the audit gets interesting. Pick the competitor with the highest total citations and investigate what's actually driving their AI visibility. You're doing this with entirely free tools.
Check their content footprint with Google. Search site:competitorname.com to see how many pages they have indexed. Then search "competitor name" -site:competitorname.com to see how many external pages mention them. A high ratio of external mentions to owned pages is a strong signal of earned authority.
Look at their backlink profile with Ahrefs' free tier or Moz's Link Explorer free version. You don't need a paid account to see the top linking domains. Are they getting links from industry publications, review platforms, community sites, YouTube channels? AI engines absorb these signals.
Check review platforms. Search for the competitor on G2, Capterra, Trustpilot, Reddit, and Product Hunt. Count the number of reviews and look at the recency. AI engines — especially models with live web access like Perplexity — weight recent, high-volume review presence heavily.
Read what people say about them, not just the star rating. Look at the actual language reviewers use. If five different G2 reviews independently describe your competitor as "the easiest option for teams without a dedicated IT person," that phrase is likely getting absorbed into AI training and inference. This is your competitive intelligence for messaging.
Look at their structured content. Do they have a FAQ page? A glossary? A "how it works" explainer? These formats are AI citation gold because they directly answer the kinds of questions users ask AI engines. If your competitor has a detailed FAQ and you don't, that's a concrete gap you can close.
Step 5: Run the Comparison Prompt Test
Here's a tactic most people miss. Go back to your AI engines and run deliberate comparison prompts that put your brand head-to-head with the top competitor.
"Compare [Your Brand] and [Competitor A] for [specific use case]." "What are the pros and cons of [Your Brand] versus [Competitor A]?"
Record the full responses again. A few things to look for:
- Does the AI have enough information about your brand to form an opinion, or does it hedge and say it doesn't have reliable data?
- What specific claims does it make about each brand?
- Is the information about your brand accurate?
If the AI says something factually wrong about your brand — which happens more than you'd think — that's an actionable finding. It means your own content isn't clearly communicating those facts, and you need to publish clearer, more direct pages that state your positioning explicitly.
If the AI simply doesn't have much to say about you compared to a competitor, that's a content volume and third-party mention problem. You need more published content and more external references.
Step 6: Map the Source Gap
For this step, Perplexity is your best friend because it shows sources. Take your ten highest-priority questions and run them in Perplexity. For each answer that cites your competitors but not you, click through the cited sources.
Build a list of the types of sources being cited:
- Blog posts from the competitor's own site
- Review platform profiles
- Industry publication features
- YouTube video transcripts
- Reddit or community forum threads
- Comparison articles from third-party review sites
This list tells you exactly what content and mentions you need to earn to close the gap. If Perplexity keeps citing a TechCrunch article about a competitor, you know you need press coverage. If it keeps citing a Capterra comparison page, you need more verified reviews on Capterra.
Create a priority-ranked action list from this source gap analysis. Not every gap is equally important — focus first on the sources that appear across the most questions and the most AI engines.
Step 7: Document Your Baseline and Set a 30-Day Re-Test Date
An audit is only useful if it becomes a benchmark. Before you close your spreadsheet, record today's date and the exact citation counts. Then put a reminder in your calendar to re-run the same 15 prompts across the same 4 engines in 30 days.
AI visibility shifts faster than traditional search rankings — sometimes dramatically within weeks of publishing new content or earning new mentions. Tracking your baseline helps you understand which actions actually move the needle.
A simple version-controlled spreadsheet with a new tab for each month works perfectly. You don't need software for this part of the process.
FAQ: Competitor AEO Audits
How often should I run a competitor AEO audit?
Once a month is a good cadence for most SMBs. If you're in a fast-moving space or actively publishing new content, every two weeks lets you see what's working faster.
Does this work for local businesses too?
Yes, with an adjustment. Add geographic specificity to your question map: "best [service] in [city]" and "who should I call for [service] near [neighborhood/region]." Local AI citation patterns are increasingly important, especially in Perplexity and Google AI Overviews.
What if my competitors all have much higher citation counts than me?
That's normal in the early stages of building AI visibility, and it's exactly the kind of gap this audit is designed to surface. The source gap analysis in Step 6 gives you a concrete to-do list. Brands that are consistently cited today usually built that visibility through a combination of strong FAQ/glossary content, review platform volume, and third-party mentions — all things you can replicate with a clear plan.
Are AI citations stable, or do they change constantly?
They change — but not randomly. They tend to shift in response to new content being published, new third-party mentions accumulating, and model updates. Running prompts at different times of day can produce slightly different results, so take single-query snapshots as directional data rather than precise measurements.
Is this the same as tracking AI rankings?
Not quite. Traditional rankings are binary — you're on page one or you're not. AI citations exist on a spectrum of prominence and context. A brand can be cited with a strong endorsement, a neutral mention, or a cautionary note. The context matters as much as the citation count, which is why recording verbatim responses matters.
What to Do With Your Audit Results
A solid manual audit gives you a clear picture of where the AI citation gaps are and why they exist. The next step is plugging those gaps — publishing content in the formats AI engines favor, accumulating reviews, earning third-party mentions, and structuring your own pages to directly answer the questions you mapped.
If you want a faster starting point for your own brand's AI visibility (not just your competitors'), AEO Juice's free 26-check AEO report runs a systematic analysis of your site's current AI visibility signals in a few minutes. Think of the manual audit you just learned as the competitive intelligence layer — and the free report as your brand's personal baseline. Run them together and you'll have a genuinely clear picture of where you stand and what to do next.
The gap between AI-visible brands and invisible ones is real — but it's mostly a content and authority infrastructure problem, not a mystery. Map it clearly, and it becomes solvable.