There's no Google Search Console for AI. No dashboard that shows you how often ChatGPT recommends your business. This guide explains exactly how to fill that gap — what to measure, how to measure it, and how to turn the data into action.
AI engines are increasingly the first stop for commercial research. Buyers ask ChatGPT "who's the best [service] in [city]" and act on whatever it says — without clicking through to a website, reading reviews, or running a Google search first.
If your brand appears in those answers, you're winning customers you never knew you could have. If you don't appear, those customers go to whoever does — and you'll never see it in your Google Analytics.
The problem: there's no native analytics for AI search. ChatGPT doesn't send you referral traffic you can track. There's no impression count. No position report. The only way to know where you stand is to ask the AI directly — systematically, at scale, across all the commercial queries relevant to your business.
That's what AI visibility tracking is: a structured, repeatable process for measuring your brand's presence in AI-generated answers before your competitors figure out it matters.
What percentage of relevant commercial-intent prompts include your brand in the response? This is your top-line AI visibility score.
When you are mentioned, are you first, second, or buried in a list of five? Position one carries far more conversion weight than position four.
When mentioned, is the description positive ("highly rated"), neutral ("an option"), or negative ("some complaints")? Negative AI sentiment is worse than not being mentioned at all.
How does your mention rate compare to the 2–3 competitors you actually lose deals to? Your absolute score means less than your relative position.
These four metrics give you a complete picture: are you showing up, where are you showing up, how are you described, and are you winning or losing against the competition?
The foundation of AI visibility tracking is a consistent set of commercial-intent prompts — the questions buyers actually ask AI when they're making purchase decisions in your category. Think: "best [your service] in [your city]", "who should I use for [problem you solve]", "[your category] vs [competitor]". Aim for 20–50 prompts covering your main service lines, target locations, and buyer personas. These need to stay consistent across tracking periods so your results are comparable.
Track across ChatGPT (GPT-4o), Claude, and Gemini at minimum. Each has a different user base and different retrieval logic — your scores will vary between them, and that variation tells you where to focus your content efforts. Don't just test one platform and assume the results apply everywhere.
Submit each prompt to each platform and record: (a) whether your brand appears in the response, (b) at what position if it's a list, (c) the sentiment of the description, and (d) which competitors appear alongside you or instead of you. At 25 prompts across 3 platforms, you're scoring 75 responses per tracking period. Manual scoring works but takes 2–3 hours. Automated tools reduce this to minutes.
Run the same prompt set with your top 2–3 competitors in mind. A prompt like "best digital marketing agencies in Chicago" will likely mention several businesses — record who appears for each prompt, not just whether you appear. This gives you a share-of-voice picture that puts your score in context.
Mention rate = prompts where you appeared / total prompts. Average position = sum of positions / prompts where you appeared. Sentiment score = (positive mentions − negative mentions) / total mentions. Competitive gap = your mention rate − competitor mention rate.
For every prompt where you don't appear: why not? Usually it's one of three reasons — (a) you don't have content that addresses that question; (b) your content exists but isn't structured as a clear, direct answer; (c) your brand isn't mentioned in the third-party sources the AI is drawing from. Each gap maps to a specific content or citation fix.
Use the same prompt set every period. Track your mention rate, average position, and competitive gap over time. Month-over-month improvement is your proof that the content investments are working.
AI visibility benchmarks vary by industry and competition level. As a rough guide based on current data across established businesses:
Most businesses that haven't invested in AEO score C or D on their first audit. That's not a disaster — it's a baseline. The point is to establish where you are now so you can measure improvement.
More important than your absolute score: how you compare to your competitors. If you're scoring 45% (Grade C) but your top competitor is scoring 30%, you're winning AI search even at a C. If your competitor is scoring 65%, that's the gap to close.
The goal is to make AI visibility tracking as routine as checking your Google Analytics. Here's a simple monthly workflow:
The prompt set in visibilityaudit.io's Prompt Library saves your approved prompts so re-running is a one-click process. The shareable report link means you can drop the results directly into a client report or Slack update without any additional formatting.
Run commercial-intent prompts across AI platforms (ChatGPT, Claude, Gemini) and score whether your brand appears, at what position, and with what sentiment. Track four metrics: mention rate, citation position, sentiment, and competitive gap. Tools like visibilityaudit.io automate the full process and return scored results with competitor benchmarking in minutes.
Monthly is the recommended baseline. Track every 2 weeks if you're actively publishing new content. Track quarterly if AI visibility is a lower priority. Always re-run after a major website change or service launch.
A mention rate above 60% on commercial-intent prompts is strong. 35–55% is average. Below 35% indicates significant gaps. More important than the absolute score is how you compare to your top 2–3 competitors — that's the gap that costs you actual deals.
The four that matter: (1) mention rate — percentage of prompts that include your brand; (2) citation position — are you first or fifth; (3) sentiment — positive, neutral, or negative description; (4) competitive gap — your mention rate vs your top competitors.
Run a full baseline audit across ChatGPT, Claude, and Gemini. Get your mention rate, competitor comparison, and a list of specific content fixes. From $5.
Run Your AI Visibility Audit → Results in under 5 minutes · Save your prompts for monthly re-runs