How to Track Your Visibility in AI Search

Your customers are asking ChatGPT, Gemini, Copilot and Google’s AI Mode for recommendations. Some of those answers mention your business. Some mention your competitors. Most business owners have no idea which.

Until recently, tracking AI search visibility meant guesswork. That changed in 2026. Google, Microsoft and OpenAI now give site owners real data, and Google Analytics 4 sorts AI assistant visits into their own channel. You still need a manual check for brand mentions, but you can build a reliable picture with mostly free tools.

Here is a practical system, from free reports to paid tools, that a small marketing team can run in about an hour a month.

What “AI search visibility” actually means

It helps to separate 3 things, because each is measured differently:

• Mentions. An AI answer names your brand, product or business, with or without a link.
• Citations. An AI answer links to your website as a source.
• Referral visits. Someone clicks that link and lands on your site.

A business can be mentioned often but rarely cited, or cited often but rarely clicked. You need all 3 to understand what is happening. If you are new to the topic, our guide to answer engine optimization explains how AI answers choose their sources.

The free reports to set up first

Google Search Console

Google now reports AI search data in 2 places. Its AI features documentation says that sites appearing in AI Overviews and AI Mode are included in the regular Performance report under the “Web” search type.

There is also a dedicated generative AI performance report, which Google says was rolled out to all websites worldwide as of August 31, 2026. It shows impressions, meaning how many times links to your site were shown in a generative AI feature, and lets you group results by page, country, date and device. It may not appear if your site has not received enough AI impressions yet.

Use it to answer a simple question: which of my pages are being shown in AI answers, and is that number growing? Our Google Search Console guide covers the basics if you have not set it up.

Google Analytics 4

On May 13, 2026, Google Analytics added an AI Assistant channel to its default channel group. According to Google’s default channel group documentation, it covers traffic from sources like ChatGPT, Gemini, Deepseek, Copilot and Grok. It excludes Google’s AI Overviews and AI Mode, which are counted as Organic Search.

To see it, open Reports > Acquisition > Traffic acquisition and look for the AI Assistant row. From there you can check:

• Which landing pages AI visitors arrive on.
• How engaged they are compared with organic search visitors.
• Whether they convert: form fills, calls, bookings or purchases.

Google also added a Source Group field in June 2026 that consolidates spelling variants of common sources, including ChatGPT and Perplexity, which makes it easier to compare individual assistants. If you need a different grouping, GA4 lets you build custom channel groups using rules on source or referrer.

ChatGPT also tags many outgoing links. OpenAI’s publishers and developers FAQ says ChatGPT adds utm_source=chatgpt.com to referral URLs, so those visits are easy to identify, as long as you allow its OAI-SearchBot crawler. Our guide to UTM parameters explains how UTM tags appear in reports.

Bing Webmaster Tools

Microsoft launched an AI Performance report in Bing Webmaster Tools in public preview in February 2026. It shows when your site is cited in AI answers across Microsoft Copilot, AI-generated summaries in Bing and select partner integrations, along with total citations, cited pages and a sample of the grounding queries the AI used to find your content.

This is one of the few places where you can see citation counts directly from the AI provider, so it is worth setting up even if Bing sends you little regular traffic.

Key takeaway: Start with the free data from the AI providers themselves: Search Console for Google’s AI features, GA4’s AI Assistant channel for visits, and Bing Webmaster Tools for Copilot citations. Add manual or paid prompt tracking only after those are in place.

Manual prompt testing: the brand mention check

None of the reports above tell you what AI assistants actually say about your business. For that, you need to ask them. A simple monthly routine works well:

1. Write 15 to 25 prompts your customers would realistically ask. Mix 3 types: - Category prompts: “What are the best accounting firms for restaurants in Denver?” - Problem prompts: “How do I fix a leaking water heater, and who can do it near Tampa?” - Brand prompts: “What is [your business] known for?” and “[Your business] vs [competitor]”
2. Run them in the assistants your customers use, such as ChatGPT, Gemini, Google AI Mode, Copilot and Perplexity. Use a logged-out or fresh session where possible, because personalization and memory can change answers.
3. Record the results in a spreadsheet: date, assistant, prompt, whether you were mentioned, whether you were cited with a link, which competitors appeared, and whether the facts about you were correct.
4. Repeat on the same prompts each month so you can compare trends, not one-off snapshots.

AI answers vary from one run to the next, so don’t panic over a single result. Look for patterns over 2 or 3 months.

Pay special attention to accuracy. If an assistant lists the wrong hours, old prices or a service you don’t offer, trace it back to the source, often an outdated directory listing or old page, and fix it there. Our guide on getting your business mentioned in AI answers covers how to improve what assistants say.

Paid AI visibility tools

Once manual testing becomes too slow, paid tools can automate it. A growing category of AI visibility trackers, including add-ons from SEO suites such as Semrush and Ahrefs and newer specialized platforms, run large sets of prompts across assistants and report:

• Share of voice: how often you are mentioned compared with competitors.
• Citation tracking: which of your pages are linked as sources.
• Sentiment: whether mentions are positive, neutral or negative.
• Prompt and topic trends over time.

Before you buy, check these points:

Pricing varies widely and changes often, so compare current plans on vendor sites. For many small businesses, a spreadsheet and an hour a month is enough to start.

How to read the numbers together

The value comes from connecting the layers. Some common patterns and what they suggest:

• High Search Console AI impressions, few visits. You are being shown, but people get their answer without clicking. Make sure your cited pages offer something worth the click, such as pricing calculators, booking or detailed guides. Our piece on zero-click search and AI traffic covers this shift.
• Strong AI Assistant visits, low conversions. Check which landing pages they arrive on. AI tools often link deep pages, which may lack a clear next step.
• Frequent mentions, few citations. Assistants know your brand but use other sites as sources. Strengthen your own pages with clear, quotable answers and up-to-date facts.
• Competitors mentioned, you are not. Look at which sources the assistant cites for them: review sites, directories, comparison articles or their own pages. That shows you where to build presence.

Tie it back to business results. Our guide on how to measure marketing ROI explains how to connect channel data to revenue.

Where to start: a 30-minute setup

1. Verify your site in Google Search Console and open the generative AI report and the Performance report.
2. Open GA4 Traffic acquisition and confirm the AI Assistant channel appears. If you are new to GA4, start with how to set up Google Analytics 4.
3. Verify your site in Bing Webmaster Tools and check AI Performance.
4. Check your robots.txt to make sure you are not accidentally blocking the search crawlers of the AI assistants you want to appear in.
5. Create a prompt-tracking spreadsheet with 15 to 25 prompts and run your first round.
6. Set a monthly calendar reminder to update all 4 sources in one sitting.

The bottom line

AI search visibility is no longer a black box. Google, Microsoft and OpenAI now provide real data on impressions, citations and visits, and GA4 groups AI assistant traffic for you. Add a simple monthly prompt check for brand mentions, and you will know where you stand, where competitors are winning and which fixes are worth making. Start free, track consistently, and pay for tools only when the manual work outgrows your time.

Sources

• Google Search Console Help: Generative AI performance report — AI Overviews and AI Mode impressions; rollout to all sites as of August 31, 2026
• Google Search Central: AI features and your website — AI feature data included in the Performance report’s Web search type
• Google Analytics Help: Default channel group — AI Assistant channel definition and exclusion of AI Overviews and AI Mode
• Google Analytics Help: What’s new in Google Analytics — May 13, 2026 AI Assistant channel and June 11, 2026 Source Group field
• Bing Webmaster Blog: Introducing AI Performance in Bing Webmaster Tools — Copilot citation reporting in public preview
• OpenAI Help: Publishers and developers FAQ — utm_source=chatgpt.com on ChatGPT referral links