How to Use Google Search Console to Track AI Traffic Before Your Competitors Do

Last month I sat with a studio owner who had spent two years building her Google rankings, and she was frustrated because her new student inquiries had quietly started to plateau. Her Google Business Profile was fully optimized, her class schedule ranked in the local pack, and her reviews were strong. But when we typed "best yoga studio for beginners with intro offer" into ChatGPT, a newer studio two blocks down was the one it named. She had no idea that was happening, because she had no way to see it. Google Search Console AI traffic data is the first place you can look to figure out whether this shift is already affecting your bookings, and you can check it right now for free.

The fear for any independent studio, salon, or clinic right now is becoming invisible while someone else captures the recommendation. You have spent years winning Google Maps and collecting five-star reviews, and that work still matters. But buyers are increasingly asking conversational questions to AI assistants instead of typing keywords into Google, and those assistants pull from a different mix of sources. If you cannot see which queries are bringing people to your site, you cannot tell which ones AI engines are starting to intercept. Google's own documentation on AI features confirms that appearances in AI Overviews are now counted within your overall Search Console performance, but you have to know how to isolate them.

Can I see if ChatGPT or Perplexity is sending traffic to my studio website?

You cannot see ChatGPT or Perplexity traffic inside Google Search Console, because Search Console only tracks what happens on Google itself. To see referrals from chatgpt.com, perplexity.ai, or gemini.google.com, you need Google Analytics 4, which tracks your incoming referral traffic from any source. What you can do in Search Console is find the conversational queries that AI engines typically surface, and then check whether your booking pages are getting picked up by Google's own AI Overviews.

The distinction matters because studio owners conflate the two platforms. Google Search Console is a search performance report: it shows you the queries people typed into Google, the impressions your pages earned, and the click-through rate on those results. It does not show you a session-by-session log of every visitor who arrived from an AI assistant. Google's documentation on the Performance report makes this explicit, describing it as aggregated data for search results, not a referral log. When you want to know if ChatGPT specifically sent someone to your intro offer page, that is a GA4 question. When you want to know if Google's AI Overviews are showing your studio for "yoga classes near me with drop-in availability," that is a Search Console question.

Find AI-Driven Queries in the Search Console Performance Report

Open your Search Console and head to the Performance report for Search results. The default view shows you clicks, impressions, and average position for your top queries, which is useful but noisy. Most studios see a mix of branded searches, local intent searches, and a handful of generic terms. What you are looking for are the conversational, question-shaped queries that signal someone asked an AI assistant a full question rather than typing a keyword. These are queries like "best yoga studio for beginners in [your neighborhood]," "who has the best intro offer for yoga near me," or "affordable massage therapy for back pain walk-in."

To isolate these, you use the Regex filter on the Queries tab. Click the filter icon at the top of the queries table, select "Custom (regex)" from the dropdown, and enter a pattern that matches conversational phrases. A practical starting regex looks like this: best|near me|intro offer|how much|should i|vs|recommend. This pattern catches queries where the searcher is asking for a recommendation, a comparison, or a specific local solution, which are the exact query types that AI assistants intercept because they pull from third-party reviews, forum threads, and directory listings rather than just your website.

When you apply that filter, you will see a much shorter list of queries, and the impressions on those queries tell you something specific. If you see 200 impressions for "best yoga studio for beginners [your city]" but only 3 clicks, that gap is exactly where AI Overviews sit. Google shows your page in the results, but the AI answer block above your link answers the question directly, so the searcher never clicks through. Google's documentation on AI features confirms that when your link appears inside an AI Overview, it counts as an impression in Search Console, and clicks on external links in AI Mode count as clicks. The problem is that the click-through rate collapses because the AI summary satisfies the user.

I have watched this pattern across the dozens of brands I have run growth for, and the conversational queries almost always show the widest impression-to-click gap. That gap is not a content problem. It is a distribution problem. Your page is visible, but the AI layer is absorbing the click. When you know which queries have that gap, you know exactly where your visibility is being intercepted, and you can start doing the work to get named inside the AI answer rather than just ranking beneath it.

Check Your Search Appearance for AI Overviews

Google has introduced dedicated generative AI performance reports for Search and Discover, which means you can now see specifically how often your pages appear in AI Overviews and AI Mode. In your Search Console Performance report, look at the filter bar and click "Search appearance." If the new AI report is available for your property, you will see a filter for AI Overviews. Selecting it narrows your entire performance view to only the impressions and clicks that came from Google's AI-generated answer blocks.

What this metric means for a local studio is straightforward. If your booking page or class schedule page shows impressions under the AI Overviews search appearance filter, Google's AI is picking up your page and showing it inside an answer block when someone asks a relevant question. If you see zero impressions there, your pages are not yet being pulled into AI answers, even if you rank well in standard search results.

There is a critical measurement detail to understand here. For AI Overviews, Google counts an impression only when the link is scrolled or expanded into view. If a user searches, sees the standard results, and never expands the AI Overview block, your link inside it does not count as an impression. This means Search Console will undercount your AI visibility to some degree, because it only registers impressions for users who actually engaged with the AI answer block. Practitioner coverage of AI Overview tracking highlights this limitation, along with the fact that data from Search Labs experimental AI experiences is not recorded in Search Console at all.

The practical move is to compare your AI Overviews impressions against your overall query performance. If you see high impressions for "drop-in yoga class [your neighborhood]" in your standard Performance report but zero in the AI Overviews filter, it means Google is ranking your page but the AI layer is not surfacing it in the answer. That is the gap you need to close, and it is usually closed by getting mentioned on the third-party pages that AI engines cite, not by optimizing your own page further. This is where the crossover happens between standard SEO and AEO for local businesses, because the fix is often about getting reviewed or mentioned on the platforms AI reads, not about changing your title tags.

Catch AI Assistant Referrals in Google Analytics 4

Search Console tells you what happens on Google. GA4 tells you what happens on your website, and that is where you see traffic from ChatGPT, Perplexity, and Gemini. When someone asks ChatGPT for a yoga studio recommendation and it generates an answer with a link to your intro offer page, that visitor arrives at your site as a referral from chatgpt.com. GA4 captures that automatically, but the default reporting view buries it under "Direct" or a generic referral bucket, so you have to build a custom segment to see it clearly.

Open GA4 and go to Admin, then Data Streams, and confirm your web stream is collecting data. Then navigate to Reports, Acquisition, and Traffic Acquisition. The default view shows you session source and medium, which groups traffic by broad categories. To isolate AI assistants, you need to create a custom segment. Click the segment icon (the overlapping circles) at the top of the report, create a new segment, and under Conditions, set the Session source to match a regex pattern: chatgpt\.com|perplexity\.ai|gemini\.google\.com|you\.com|claude\.ai|copilot\.microsoft\.com. Google's GA4 documentation on segments and filters covers the mechanics of this, and the regex pattern is the same approach you used in Search Console, just applied to referral sources instead of search queries.

When you apply that segment, you see every session that arrived from an AI assistant over the selected date range. The number will be small for most studios, and that is honest. What matters is the trend line and the landing pages. If you see five sessions from chatgpt.com that all landed on your intro offer page, that tells you ChatGPT is recommending your studio specifically for new student offers. If you see sessions from perplexity.ai landing on your class schedule, Perplexity is pulling your schedule into its answers.

The landing page data is the actionable part. It tells you what AI assistants think your studio is known for, which is not always what you think you are known for. I have seen studios that position themselves around advanced teacher training discover that AI assistants are recommending their beginner classes, because the beginner class page had more third-party mentions and reviews. That mismatch is a strategic insight you only get by checking the referral data, because your own positioning does not always match how the web describes you.

Connect AI Queries to Actual Class Bookings

Finding the queries and the referrals is only half the work. The question that matters is whether those visitors actually booked a class, signed up for an intro offer, or joined a waitlist. A visit from ChatGPT that bounces without booking is a vanity metric. A visit from Perplexity that converts into a paid intro offer is a signal that AI search is already driving revenue for your studio, and that is where you focus your effort.

In GA4, navigate to Conversions under the Reports section. If you have conversion events set up for booking confirmations, intro offer purchases, or contact form submissions, you can cross-reference those events against your AI referral segment. Go back to the Traffic Acquisition report with your AI assistant segment applied, and look at the conversions column. If your booking confirmation event fires on sessions from chatgpt.com, you have a direct line from AI recommendation to paid booking.

The cross-reference works the same way in Search Console. Take the conversational queries you found using your regex filter in Step 1, and check which landing pages they hit. In the Performance report, switch from the Queries tab to the Pages tab while your regex filter is still active. You will see which pages are earning impressions for those conversational queries. If your intro offer page is getting impressions for "best yoga intro offer near me" but has a low click-through rate, the AI layer is intercepting the click. If your class schedule page is getting clicks from that query, visitors are finding what they want and moving toward a booking.

The goal is to build a simple map: query, landing page, conversion event. When you have that map, you know exactly which AI-driven search paths end in a booked class and which ones are leaking. The leaking paths are where you focus, because a leak means someone asked an AI for a recommendation, found your studio, visited your page, and left without booking. That is usually a page quality issue, a pricing clarity issue, or a booking friction issue, all of which you can fix without touching your SEO.

What to Do When You Find AI Traffic (Or Don't)

If you followed the steps above and found AI referral traffic in GA4, even a handful of sessions, the next move is to optimize the pages those visitors landed on. Make sure your intro offer is prominent, your schedule is current, and your booking flow works on mobile without redirects or popups. If ChatGPT is sending people to your intro offer page, that page needs to load fast, answer the question "how much does it cost," and make it possible to book in under three taps. AI referral traffic is high-intent traffic, because the user already got a recommendation from a source they trust. They are coming to your site to confirm and book, not to browse.

If you found conversational queries in Search Console but saw zero AI referral traffic in GA4, it means AI assistants are not yet recommending your studio. People are asking the questions, Google is showing your pages in standard results, but ChatGPT and Perplexity are naming other studios. That gap is not going to close by itself, and it is not a Google ranking problem. It is an answer engine visibility problem. The fix is getting your studio mentioned on the pages AI assistants cite when they generate recommendations, which means getting your Google Business Profile recommended by AI, getting reviewed on platforms AI reads, and participating in the local forum threads and directory listings that AI engines pull from.

If you found nothing at all, no conversational queries in Search Console and no AI referrals in GA4, your studio is in the earliest stage of AI invisibility. This is common for studios that have relied entirely on word-of-mouth and Google Maps. The work here is to start creating the content and third-party presence that AI engines need to find you. Write a clear FAQ page answering the five questions new students ask most. Make sure your pricing is visible as plain text, not an image. Get listed on local directories and class aggregators. Those are the basics that make you discoverable to AI assistants in the first place, and without them, no amount of Search Console filtering will show you AI traffic because there is none to find.

Don't Wait for Google to Show You the Full Picture

Search Console and GA4 are lagging indicators. They tell you what already happened, not what is about to happen. The conversational queries you find in Search Console show you what people asked last month, and the AI referrals in GA4 show you who visited last week. By the time you see a clear trend in either tool, the shift has already happened, and the studios that moved first have already captured the AI recommendation slot for your category.

Google's new AI-powered configuration feature in Search Console, announced recently, can translate natural-language questions into filters for your Performance report, which makes it easier to ask "show me queries related to yoga intro offers" and get the right view. But even with that tool, you are still looking backward at impressions and clicks rather than forward at whether AI assistants will recommend you when the next person asks.

The studios that win this shift are the ones who check their AI visibility proactively, not the ones who wait for Google to surface it in a report. You can run the regex filters and GA4 segments today to see where you stand, and if you want to know whether AI assistants actually recommend your studio when people ask the questions that matter, the GEO Rank Tracker runs those questions against live engines and shows you exactly what ChatGPT, Perplexity, and Gemini say. Pair that with tracking your AI rankings over time, and you have both the traffic data and the recommendation data in one view. The studios two blocks down are already doing this. The question is whether you check before they fill the class you could have filled.