To see if AI platforms like ChatGPT, Claude, and Perplexity are actually sending home buyers to your website, you do not need to guess. You just need to filter for specific referral sources inside Google Analytics 4 (GA4).
Home buyers are increasingly using generative search to ask highly specific questions, such as, "Which neighborhoods in Austin have the best schools and four-bedroom homes under $800k?" If an AI engine cites your market report or neighborhood guide as the answer, those clicks represent motivated, highly qualified leads.
The problem is that analytics platforms often bury this traffic. Without the right GA4 filters, AI referrals get lumped into generic "direct" or "referral" buckets, making it impossible to know if your discoverability efforts are paying off. Before you spend hours updating your WordPress content or installing a new schema plugin to feed better structured data to these platforms, you need a baseline. You need to know if AI systems are already finding, trusting, and recommending your pages to buyers.
Here is exactly how to isolate AI traffic in GA4 so you can measure what is actually working.
How do you track AI referrals in GA4 for Realtors GEO?
To track AI referrals in Google Analytics 4 (GA4), you must build a custom exploration that isolates traffic from specific AI domain names like chatgpt.com and perplexity.ai. Without this setup, visitors sent by AI tools get dumped into your generic traffic buckets, leaving you completely blind to whether your AI visibility efforts are actually attracting home buyers.
When a prospective buyer asks ChatGPT for "top real estate agents in Denver" and clicks your profile, that is a highly qualified lead. You need to know where it came from. By default, GA4 treats many of these clicks as standard referral traffic. To find them manually, open your GA4 Traffic Acquisition report and search the "Session source" column for domains like chatgpt.com, perplexity.ai, and claude.ai.
To stop searching manually every week, build a permanent view. Go to the Explore tab and start a blank report. Create a custom segment that includes any session where the source matches those specific AI domains. You can use the official Google Analytics support documentation if you need help navigating the Explore tool. Save this as your "AI Search Referrals" segment. Apply it to see exactly how many AI-referred visitors viewed your active property listings or booked a consultation.
You will also need to account for hidden AI traffic. Often, AI platforms strip the tracking data entirely before sending the user to your site. This makes the visit look like generic "Direct" traffic, as if the buyer typed your exact URL into their browser. To separate an AI traffic tool from true direct traffic, monitor your Direct traffic bucket for sudden spikes to specific, deep pages. If a deeply buried neighborhood guide about local property taxes suddenly gets fifty direct visits in a week, an AI engine is likely citing it. Find those pages and ensure your phone number and contact forms are immediately visible at the top to capture the lead.
Why do standard GA4 setups miss Realtors GEO traffic?
Standard Google Analytics 4 setups miss AI traffic because many platforms strip away tracking data before sending a home buyer to your website. If you rely entirely on default reports, you will severely underestimate how many leads actually found your real estate practice through ChatGPT or Perplexity, leaving you blind to what is driving your pipeline.
The biggest issue stems from mobile AI applications. When a buyer uses the ChatGPT iOS app to search for "top buyer agents in Miami" and taps your website link, the app often removes the referral data to protect user privacy. In GA4, this visit gets dumped into your "Direct" traffic bucket. A highly qualified lead looks exactly like someone who typed your web address from memory. To catch these hidden visitors, filter your Direct traffic report for sudden spikes on deep, specific pages. If a neighborhood guide suddenly jumps from zero to thirty direct visits in a week, an AI engine likely cited it. Add a clear contact form directly to the top of that specific page to capture the traffic.
You also lose tracking data through the "dark social" problem of shared AI chats. If a buyer asks Claude to compare property taxes in two counties and shares that chat log with their spouse, any subsequent clicks to your site arrive with zero origin data. To build a safety net, you can use UTM parameters - simple text tags added to the end of a web address that force GA4 to record where the click originated. While you cannot force ChatGPT to append UTMs naturally, you can manually add them to the links on your Google Business Profile and local directory listings to ensure those foundational citations track cleanly.
Finally, traditional SEO metrics fail completely here. Google Search Central only tracks clicks from traditional Google searches, not private AI chats. To accurately measure AI discovery, stop relying strictly on software. Ask every new consultation, "Did an AI tool like ChatGPT suggest us?" and record the answer. Connecting real conversations to your pipeline is the most reliable way to prove your AI visibility efforts are generating signed contracts.
How does the right schema plugin help AI platforms find your listings?
The right schema plugin translates your website text into a structured database that AI engines can read instantly. Without this translation layer, ChatGPT has to guess whether you are an active real estate agent in Denver or just a blogger writing about the housing market - meaning you remain invisible when buyers ask for local recommendations.
Schema is essentially a digital name tag for your business, and JSON-LD is the specific code format used to write that name tag behind the scenes. When an AI tool scans your site, a RealEstateAgent schema tag explicitly feeds it your name, your service area, and your contact details. This establishes your entity - a verified, distinct identity in the AI's underlying knowledge base. To set this up today, install a foundational plugin like Yoast SEO or All in One SEO and configure your local business settings specifically to the real estate category. This simple step ensures AI tools confidently categorize your practice as a local professional rather than a generic website.
AI platforms also need exact details to answer specific buyer queries, like finding a "three-bedroom home in Austin under 500k." If your property listings are just standard <p> text paragraphs, AI crawlers might misunderstand or skip the details entirely. By structuring property features with JSON-LD, you hand the AI a clean, factual list it can cite directly to home buyers. You can write this code yourself and insert it using a safe snippet manager like WPCode, but that becomes tedious if you manage dozens of properties. Check your current listing software to ensure it automatically outputs structured property data into the <head> section of your pages so AI systems can extract those facts immediately.
To rank well in complex AI answers, you also need nested schema - a way to connect multiple facts together, like linking a specific property directly to your agent profile. When choosing tools, look for options that automate this connection. LovedByAI automatically scans your pages and injects nested JSON-LD to link your firm, your agents, and your properties without manual coding. If you choose the free manual route, use the official Schema.org documentation to map out your firm's relationships, generate the code, and paste it into your WordPress templates. Pick a workflow that ensures every new listing automatically broadcasts clear, structured facts to the AI engines driving your next buyer.
What should you fix first if AI engines are not recommending your firm?
If ChatGPT is recommending other local Realtors over you, fix your foundational local data and start answering highly specific buyer questions directly on your website. AI engines do not guess; they cross-reference your site against the rest of the internet to verify you are a legitimate, active real estate agent. If they find conflicting information, they skip you and cite a competitor instead. To see exactly what the AI sees, open an incognito window and search for your brokerage's name and phone number. Make a list of every outdated profile or broken link that appears on the first two pages of results.
AI systems rely heavily on major real estate directories like Zillow, Realtor.com, and your Google Business Profile to confirm your service areas and contact details. If your office address on Yelp does not match the footer of your WordPress site, the AI loses confidence in your business identity and removes you from its answer pool. Go through your list of outdated profiles and manually update your address, phone number, and brokerage name to match your website exactly. If you have dozens of profiles, you can use a generic citation management service to push the correct data out automatically, but fixing the top five directories by hand is free and immediately improves your AI discoverability.
Once your basic identity is verified, you need to feed the AI the exact answers home buyers are looking for. Buyers frequently ask AI complex questions like, "What are the average property taxes for a four-bedroom house in Scottsdale?" If your neighborhood guide answers this, wrap that text in FAQPage schema - a specific type of JSON-LD code that formats your questions and answers into a neat, machine-readable list. Without this, AI crawlers just see a wall of text inside standard <div> or <p> wrappers and often miss the details entirely. You can use LovedByAI to automatically generate and inject this markup for you, or manually write the code using the guidelines on Schema.org and paste it into your WordPress page header. Add three to five specific, locally focused questions to your top neighborhood pages to start capturing these high-intent buyer inquiries.
How to Build an AI Referral Report in GA4
You spend time optimizing your real estate site to appear in AI answers, but standard analytics tools bury the results. GA4 often lumps AI traffic into general referral buckets, hiding whether ChatGPT or Perplexity is actually sending buyers to your property listings.
To measure your AI visibility, you need a custom exploration. Here is how to build a dashboard that tracks your known AI referrals.
Step 1: Open your GA4 property and navigate to the Explore tab. Click to create a new Blank exploration.
Step 2: In the Variables column, click the plus icon next to Dimensions. Import Session source/medium and Landing page + query string. Then, click the plus next to Metrics and import Sessions and Key events (this tracks your consultation requests or form fills). Double-click all of these to add them to your report.
Step 3: Scroll down to the Filters section in the Tab Settings column. Create a custom filter where the dimension Session source/medium uses the condition "matches regex" (regular expression). Paste this exact snippet:
openai|perplexity|claude|anthropic
Step 4: Review the output. With the Landing page added as a secondary dimension, you can see exactly which neighborhood guides or property listings the AI platforms are citing.
Step 5: Name and save the exploration as your primary AI visibility dashboard. You can use this to monitor traffic changes after updating your structured data or after you check your site for missing schema.
WordPress Implementation Note:
For this to work on a WordPress site, your GA4 measurement ID must fire cleanly in the <head> section of your pages. If you use a caching plugin, ensure your tracking scripts are not being delayed or deferred incorrectly, as this can drop fragile referral strings.
What to watch for: This report only catches traffic where the AI tool passes referral data. Mobile AI apps (like the ChatGPT iOS app) often pass zero referral information, so that traffic will still show up as "Direct" in your GA4 analytics platform. Treat this exploration as a baseline indicator of your AI search performance, not the absolute total.
Conclusion
Knowing which AI platforms are actually sending home buyers to your real estate website removes the guesswork from your marketing strategy. By setting up these simple tracking parameters in Google Analytics 4, you move past vanity metrics and start seeing real business outcomes. If you notice that a specific platform like ChatGPT is driving qualified leads to your property listings, you can focus your optimization efforts right there.
Take the time to configure your tracking today so you have baseline data for tomorrow. The real estate market moves fast, and understanding how modern buyers find you gives your agency a distinct advantage. Once you know where your traffic originates, you can start shaping how those AI systems understand your local expertise.
For a complete guide to AI SEO strategies for Realtors GEO, check out our Realtors GEO AI SEO page.
