How to Measure AI Visibility:
The Metrics That Actually Matter
By Lesli Rose · April 11, 2026 · 11 min read
You can't improve what you can't measure. That's true for AI visibility just like it's true for everything else in business. But here's the problem: the metrics most people are used to tracking -- keyword rankings, organic traffic, click-through rates -- don't capture AI visibility. You need a different measurement framework, and it doesn't exist as a standard yet. So I built one.
What follows is the system I use to measure AI visibility for every audit I run. It's not theoretical. It's what I check, how I score it, and what "good" looks like at different stages. You can do most of this yourself with free tools and a spreadsheet.
Why Traditional SEO Metrics Miss the Mark
Traditional SEO gives you clear numbers. You rank #3 for "best plumber in Dallas." You get 2,400 organic visits a month. Your click-through rate is 4.7%. These are useful metrics for search visibility. They tell you almost nothing about AI visibility.
The reason is simple: AI recommendations and search rankings are different systems. You can rank #1 on Google for your target keyword and be completely absent from ChatGPT's recommendations. I've audited businesses where this is exactly the case -- strong SEO, zero AI visibility.
AI visibility requires its own set of metrics. Here are the five that matter.
Metric 1: AI Citation Rate
This is the most direct measure of AI visibility: when people ask AI about your category, do you show up?
How to measure it
Write 5-10 prompts that a potential customer might ask. Examples: "Best [your service] in [your city]" and "Who do you recommend for [your category] near [your area]?"
Ask each prompt in ChatGPT, Perplexity, and Gemini.
Record whether your business appears in each response. Note your position (first mentioned, second, third, or not at all).
Calculate your citation rate: number of times you appeared divided by total number of prompts across all platforms.
A citation rate of 0% means you are invisible to AI. A rate above 50% across multiple platforms means you have strong AI visibility in your category. Most businesses I audit are between 0% and 20%.
Metric 2: Schema Completeness Score
Schema markup is the structured language your website uses to communicate with AI. Having some schema is not the same as having complete schema. Most websites have basic schema or none at all. The ones that AI recommends consistently have comprehensive, detailed schema.
Organization/LocalBusiness schema -- your entity identity. Name, address, phone, website, description, founding date, service area.
Service schema -- every service you offer, described individually with clear names and descriptions.
Review/AggregateRating schema -- your reviews, structured so AI can read your rating and review count directly.
FAQ schema -- questions and answers on your pages, marked up so AI can extract them.
Person schema -- the owner/founder, with credentials and expertise signals.
Score yourself: count how many of these you have and how complete each one is. Most businesses score 1 out of 5 or 2 out of 5. Getting to 4 or 5 out of 5 puts you in the top tier.
Metric 3: AI Crawler Access
This is binary: either AI crawlers can access your site, or they can't. But it's surprising how many businesses fail this check.
Check these three crawlers
GPTBot -- OpenAI's crawler for ChatGPT. If blocked, ChatGPT cannot update its knowledge of your site.
ClaudeBot -- Anthropic's crawler for Claude. Same principle.
PerplexityBot -- Perplexity's crawler. Perplexity actively crawls and cites sources, so blocking this one directly hurts your citation rate on that platform.
Go to yourdomain.com/robots.txt. Search for each of these bot names. If you see "Disallow: /" after any of them, that platform cannot crawl your site. The fix is straightforward -- update your robots.txt to allow these crawlers. But you have to know to check.
Metric 4: Earned Visibility Score
AI systems cross-reference your website with third-party sources. The more consistent, positive mentions you have across trusted platforms, the more confident AI becomes in recommending you. This is the metric most businesses underestimate.
Review count and rating -- on Google Business Profile and industry-specific platforms. AI systems can see this data and use it as a confidence signal.
Directory listings -- are you listed on the authoritative directories for your industry? Are the listings complete and consistent with your website?
Third-party mentions -- blog posts, roundups, listicles, news articles that mention your business. Each one is another signal node.
Social proof signals -- active, verified profiles on social platforms with consistent business information.
I score earned visibility on a simple scale: how many independent sources confirm your business entity with consistent information? Under 5 is weak. 5-15 is developing. Over 15 is strong. The key word is "consistent" -- conflicting information across sources actually hurts you.
Metric 5: Entity Consistency
This is the hidden metric that ties everything together. If your business name, address, phone number, description, and services are different across your website, Google Business Profile, directories, and social profiles, AI systems get confused. Confusion kills confidence. Low confidence means no recommendation.
The consistency check
Search for your business name in quotes on Google. Look at every listing that appears. Is the business name exactly the same everywhere? Is the address formatted the same way? Is the phone number consistent? Is the description of what you do aligned across all platforms? Every inconsistency is a signal to AI that your business entity is ambiguous, and AI does not recommend ambiguous entities.
Building a Simple AI Visibility Dashboard
You don't need software. A spreadsheet works. Here's the structure I recommend.
Row per month -- track each metric monthly so you can see trends.
Column for each metric -- AI Citation Rate (%), Schema Score (out of 5), Crawler Access (pass/fail for each bot), Earned Visibility Score (number of consistent sources), Entity Consistency (pass/fail).
Notes column -- record what changed, what you fixed, what competitors are doing.
Screenshot folder -- save screenshots of AI responses each month. This is your evidence trail.
Fifteen minutes a month with this system gives you more AI visibility insight than any automated tool on the market right now.
What "Good" Looks Like at Different Stages
Starting out (most businesses)
Citation rate under 10%. Basic or no schema. Some AI crawlers may be blocked without your knowledge. Fewer than 5 consistent directory listings. Inconsistent entity information across platforms. This is normal. Most businesses are here.
Building momentum
Citation rate 20-40%. Schema covers 3+ types. All AI crawlers allowed. 10+ consistent directory listings. Entity info mostly aligned. You start appearing in AI responses occasionally. This takes 2-4 months of focused work.
Strong AI visibility
Citation rate above 50%. Comprehensive schema (4-5 types). All crawlers allowed plus llms.txt file. 15+ consistent sources. Perfect entity consistency. AI recommends you regularly and by name. This is where your AI visibility becomes a real competitive advantage.
How Often to Measure
Monthly at minimum. AI systems update their training data and models regularly. What was true in March may not be true in April. Your competitors are also making changes that affect the competitive landscape.
Monthly checks take about 15 minutes once you have the system set up. Quarterly, spend an hour doing a deeper analysis: look at trends, compare to competitors, reassess your strategy. Annually, do a full audit to make sure nothing has drifted.
The businesses that measure consistently are the ones that improve. The ones that check once and forget about it eventually lose ground to competitors who are actively working on their AI presence.
Frequently Asked Questions
Can Google Analytics track AI visibility?
Not directly. Google Analytics tracks website traffic, not AI recommendations. It can show you if traffic is coming from AI platforms (look for referrals from chat.openai.com or perplexity.ai), but it cannot tell you whether AI is recommending your business to people who never click through. Most AI recommendations happen in a conversation where the user may call you directly or visit your site by typing your URL -- neither of which Google Analytics attributes to AI.
Is there a tool that measures AI visibility automatically?
As of 2026, no single tool measures all aspects of AI visibility automatically. Some platforms track AI citations, and schema validators check structured data, but no tool combines citation tracking, schema auditing, crawler access verification, and earned signal monitoring into one automated system. The most reliable approach is still a combination of manual checks and free tools, done consistently each month.
What AI visibility score should I aim for?
There is no universal AI visibility score yet. Focus on these benchmarks: appearing in at least one AI platform's recommendations for your primary category, having complete schema markup, confirming all major AI crawlers have access, maintaining 20+ reviews on Google Business Profile, and having consistent business information across at least 5 directories. If you hit all five, you are ahead of 90% of businesses.
How do I know if ChatGPT is recommending my competitors?
Ask it. Open ChatGPT and ask questions like "Who is the best [your service] in [your city]?" Record which businesses appear. Do this monthly. Also ask about specific competitors by name -- "What do you know about [competitor name]?" This tells you how much AI knows about them compared to you. If competitors appear consistently and you do not, they have stronger AI visibility signals.
Get Your AI Visibility Measured by a Pro
Stop guessing. Get a complete AI visibility audit with real metrics, real benchmarks, and a clear action plan.
Get Your AI Visibility Audit