How to Measure AI Search Visibility: A Simple Monthly Method

In short: You can measure AI search visibility with a spreadsheet and an hour a month. Write 30 to 50 questions your customers really ask, run them on the AI systems they use, and record whether your business is mentioned, whether your site is cited, how you are described and who else appears. Because answers change between runs and platforms, repeat the exercise on a schedule and watch the trend, not any single result.

Key Takeaways

• Traditional rank tracking does not show whether an AI assistant recommends you. You have to ask the assistants.
• A prompt set is the unit of measurement: a fixed list of customer questions that you re-run each month.
• Record four things per answer: mentioned or not, cited or not, described accurately or not, and which competitors appear.
• Run each prompt more than once. Answers vary, so a single run can mislead.
• Google reports traffic from its AI features inside the Web search type in Search Console, not as a separate report, so Search Console alone cannot tell you your AI visibility.
• Use the results to find fixes: wrong facts, missing pages, absent third-party mentions.

Step 1: Build the Prompt Set

Aim for 30 to 50 questions, grouped by what a customer wants. Use their language, not yours.

• Discovery: "best dental clinic in Kathmandu", "who builds AI chatbots for real estate".
• Comparison: "[your company] vs [competitor]", "AI receptionist or answering service".
• Evaluation: "how much does X cost", "is X worth it for a small clinic".
• Brand: "what does [your company] do", "is [your company] good".

Include the questions that sales calls and support emails already contain. Freeze the list so you can compare month to month, and add new prompts in a separate batch.

Step 2: Choose the Systems

Test where your customers are. A reasonable set is ChatGPT (with search on), Perplexity, Google (AI Overviews and AI Mode where available), Gemini and Claude. Note the date, the system and, where the product allows, the mode. Test in a fresh session without a personal history that might skew results, and record the country and language you used, since answers can differ by location.

Step 3: Record the Results

For each answer, log:

Repeat each prompt two or three times. If you appear in some runs and not others, note that as partial visibility rather than a yes or no.

Step 4: Calculate Simple Scores

• Mention rate: answers that name you divided by answers collected.
• Citation rate: answers that cite your site divided by answers collected.
• Accuracy rate: mentions that are fully correct.
• Share of answer: your mentions as a proportion of all brands named for the same prompts.

These are your own measures from your own sample, not industry benchmarks. Keep the method fixed and the numbers will tell you whether things are moving.

Step 5: Turn Findings Into Work

• Not mentioned at all: check whether you have a clear page that answers the question, and whether third-party sources that the systems cite mention you.
• Mentioned but wrong: correct the source. Update your own pages first, then listings and directories that carry old facts.
• Competitors always cited: look at what they publish that you do not: comparison pages, pricing explanations, evidence.
• Cited but not described well: make the top of the cited page state who you are and what you do in a sentence or two.

Our post on where AI answers get their sources explains why third-party listings often decide the outcome.

Combine With Other Data

• Search Console: Google says AI features traffic is included in the Web search type. Watch impressions and clicks for question-style queries.
• Analytics: look for referrals from AI products and for direct traffic changes after you publish new content.
• Server logs: see which AI bots fetch your pages. Our guide to AI crawlers and robots.txt lists the user agents.
• Sales questions: ask new customers how they found you and whether an assistant was involved.

Common Mistakes

• Treating one run as the truth.
• Changing the prompt set every month so nothing is comparable.
• Testing only with your company name, which tells you little about discovery.
• Chasing mentions that do not bring buyers. Prioritise the questions that sit close to a purchase.
• Paying for a tool before you have a method. Tools can save time, but they work from the same idea.

The Short Version

Measure AI visibility by asking the assistants your customers' questions on a fixed schedule, logging mentions, citations and accuracy, and acting on the gaps. It is manual at first and valuable immediately, because it shows you what a buyer sees before they ever reach your site.

If you would like help running this or fixing what it uncovers, see our answer engine optimization service.

Related Insights

• What Is GEO? Generative Engine Optimization vs AEO vs SEO
• Where AI Answers Get Their Sources, and Why Listings Matter as Much as Your Site
• AI Overviews and Fewer Clicks: What Google Actually Says You Need to Do

Sources

• Google Search Central, AI features and your website, read November 3, 2026
• Zunkiree Labs, AI search playbook