--- title: "Track ChatGPT traffic in GA4 on your real estate website" description: "Find ChatGPT visits in GA4, keep the landing page, and compare the recorded form events with buyer and seller inquiries received by your team." canonical: "https://scalewithsearch.com/articles/track-chatgpt-agent-website-inquiries-ga4" --- ## Site navigation - [Scale With Search](https://scalewithsearch.com/) - Services - Services - [AI systems](https://scalewithsearch.com/ai-systems) - [Site build](https://scalewithsearch.com/work#site-build) - [Whole website fix](https://scalewithsearch.com/work#whole-website-fix) - [Content library with SEO](https://scalewithsearch.com/work#content-library) - [Working session](https://scalewithsearch.com/audit) - Start here - [Send your brief](https://scalewithsearch.com/work#send-your-brief) - [Example scope](https://scalewithsearch.com/work#review-the-example-scope) - [Prices and terms](https://scalewithsearch.com/work#prices-and-terms) - Industries - Real estate - [Real estate agents and teams](https://scalewithsearch.com/for/real-estate) - [Platforms, ownership, and IDX](https://scalewithsearch.com/for/real-estate/real-estate-website-platforms-ownership-idx) - [Search, content, and AI](https://scalewithsearch.com/for/real-estate/real-estate-search-content-ai) - [Inquiries, campaigns, and handoffs](https://scalewithsearch.com/for/real-estate/real-estate-website-inquiries-campaigns-handoffs) - [Buyer, seller, and local pages](https://scalewithsearch.com/for/real-estate/real-estate-buyer-seller-local-resources) - [Rules, privacy, and trust](https://scalewithsearch.com/for/real-estate/real-estate-website-rules-privacy-trust) - Browse by business - [Trades and home services](https://scalewithsearch.com/for/#group-trades) - [Health, wellness, and animal care](https://scalewithsearch.com/for/#group-practices) - [Law, finance, and IT](https://scalewithsearch.com/for/#group-professional) - [Marketing and SEO agencies](https://scalewithsearch.com/for/#group-agencies) - [Shops and manufacturers](https://scalewithsearch.com/for/#group-manufacturing) - [Restaurants, shops, studios, and nonprofits](https://scalewithsearch.com/for/#group-local-services) - Chiropractic practices - [Chiropractic and wellness practices](https://scalewithsearch.com/for/chiropractic-and-wellness-practices) - [Chiropractic and physical therapy clinics](https://scalewithsearch.com/for/chiropractic-and-physical-therapy-clinics) - [All industries](https://scalewithsearch.com/for/) - Resources - Free tools - [AI readiness check](https://scalewithsearch.com/check) - [Free tools](https://scalewithsearch.com/tools/) - [Free Starter Kit](https://scalewithsearch.com/kit/business-memory-starter-kit.zip) - Guides - [AI-native library](https://scalewithsearch.com/ai-native) - [Industry intelligence](https://scalewithsearch.com/intelligence) - [Office job guides](https://scalewithsearch.com/guides/) - [Business memory](https://scalewithsearch.com/articles/business-memory-for-ai-agents-guide) - [Agent governance](https://scalewithsearch.com/articles/ai-agent-governance-guide-small-business) - [Migration and ownership](https://scalewithsearch.com/articles/leaving-vendor-ai-memory-guide) - Articles - [All articles](https://scalewithsearch.com/articles/) - [What changed](https://scalewithsearch.com/changes/) - About - Evidence - [Proof](https://scalewithsearch.com/proof) - Scale With Search - [About](https://scalewithsearch.com/about) - [How it works](https://scalewithsearch.com/how-it-works) # How to track ChatGPT traffic in GA4 on your real estate website To track ChatGPT traffic in GA4, open **Traffic acquisition**, inspect **Session source / medium**, and isolate the ChatGPT source values your property records. Keep the landing page beside those sessions. Then compare your inquiry events with buyer and seller messages received by your team. Google's [Traffic acquisition guidance](https://support.google.com/analytics/answer/12923437?hl=en) explains the report and its session dimensions. Start with the current channel definition. Google introduced an **AI Assistant** default channel on May 13, 2026. Recognized assistant referrers receive the medium `ai-assistant` and campaign `(ai-assistant)`. A custom channel is optional for your reporting needs. [Google's release note](https://support.google.com/analytics/answer/9164320?hl=en#05132026). For an agent, use separate checks for recorded visits, accepted form submissions, and received inquiries. This guide proposes that checking method. It does not report results from your website. ## Find ChatGPT sessions before changing the setup Open **Reports > Acquisition > Traffic acquisition** in the property for your website. Select a date range and write it down. If the report is missing, ask the property editor to restore it. Google documents this report path and the editor's ability to add the report to navigation. [Report instructions](https://support.google.com/analytics/answer/12923437?hl=en). Select **Session default channel group** and inspect **AI Assistant**. Treat that row as the assistant category. Google includes multiple assistants in it and excludes Google's AI Overviews and AI Mode. [Default channel definitions](https://support.google.com/analytics/answer/9756891?hl=en). Switch the table to **Session source / medium**. Search for `chatgpt`, then inspect the returned source names. Check `openai` separately for any additional values. Record the actual rows before deciding which belong in your ChatGPT report. Keep the medium as recorded. Avoid requiring `referral` for every ChatGPT session because Google's current assistant classification uses `ai-assistant`. Preserve other observed source and medium combinations for investigation. [Assistant classification rules](https://support.google.com/analytics/answer/9756891?hl=en). Use session dimensions for this question: which source brought the visit being reviewed? **First user source** answers where GA4 first acquired the user. It can remain the same when that person returns through another source. [Google's scope explanation](https://support.google.com/analytics/answer/11080067?hl=en). ## Create a ChatGPT session segment for repeat checks Use an exploration when you need to revisit the same selection. Google's segment builder supports user, session, and event segments. A session segment selects matching sessions, while a user segment can include other activity by the same users. [Segment builder documentation](https://support.google.com/analytics/answer/9304353?hl=en). In **Explore**, open or create an exploration. Click the plus beside **Segments**, create a custom segment, and select **Session** scope. Name it **Recorded ChatGPT sessions**. Add an inclusion condition using **Session source** and the exact source values verified in your report. Apply the segment. Google's [segment instructions](https://support.google.com/analytics/answer/9304353?hl=en) describe these controls. Keep `chatgpt.com` and `chat.openai.com` as candidate values to inspect, rather than assuming both occur in your property. If both are present, include them with an OR condition. Leave unfamiliar sources unresolved until the website maintainer checks them. Create a separate segment for **Session default channel group exactly matches AI Assistant** if you also want the wider comparison. Keep its label distinct from the ChatGPT segment. Never present all assistant sessions as ChatGPT sessions. Retain the date range and inclusion rules with each export. If you change the source list, preserve the previous definition. Use the same definition when comparing reporting periods. ## Keep the first page beside the source Your next question is which page the visitor reached first: an agent profile, seller page, neighborhood guide, or property page. Google defines the landing page using the first pageview in a session. It documents adding **Session source / medium** to the Landing page report. [Landing page guidance](https://support.google.com/analytics/answer/12931766?hl=en). Open **Reports > Engagement > Landing page**. Add **Session source / medium** as the secondary dimension. Isolate the verified ChatGPT values. Alternatively, use Traffic acquisition with **Session source / medium** and add **Landing page + query string**. Both approaches appear in Google's [report instructions](https://support.google.com/analytics/answer/12931766?hl=en). Save the source, medium, landing page, sessions, and reporting dates together. Add the selected inquiry event after its implementation passes the checks below. Inspect the contact route from each landing page yourself. For a seller guide, follow the invitation to ask about selling. For a buyer page, follow the property or representation inquiry. Write down the form and its receiving destination. Keep the first landing page separate from the page where someone submits the form. Before retaining URLs in analytics, check their contents. Google prohibits sending personally identifiable information, including email addresses and personal phone numbers. Its guidance also covers URL paths, query parameters, titles, and user-entered fields. [Google's privacy requirements](https://support.google.com/analytics/answer/6366371?hl=en). ## Define the event that represents an inquiry Google recommends `generate_lead` when someone submits a form or request for information. Recommended events require implementation; they are not sent automatically. Enhanced measurement separately offers `form_start` and `form_submit`. [Recommended events](https://support.google.com/analytics/answer/9267735?hl=en), [form interaction events](https://support.google.com/analytics/answer/9216061?hl=en). Ask your website maintainer to document the trigger for each seller and buyer form. Require the inquiry event to follow the form handler's successful acceptance. Keep validation failures, button clicks, and repeated confirmation-page loads outside that trigger. These are proposed acceptance requirements for your website. Keep a phone-link click in its own event. Require a call record or recipient confirmation before counting a received phone inquiry. Apply the same distinction to an email-link click and an email that arrives. Once the inquiry event works, ask an authorized property editor to mark it as a key event under **Admin > Data display > Events**. Select the event in **Recent events** and use its star control. This designation affects reporting from that point; it does not rewrite historical data. [Key event instructions](https://support.google.com/analytics/answer/13128484?hl=en). In the acquisition report, select that inquiry event in the key-event selector. Keep other key events out of the inquiry total. Record the event name alongside the exported results. [Google's key-event reporting guidance](https://support.google.com/analytics/answer/12571843?hl=en). ## Require evidence from a controlled test Prepare a test with the person receiving inquiries. Use contact details you control and a recognizable test message. Save the page, submission time, form result, and matching received message privately. Have the maintainer enable debug mode for their test device. **DebugView** displays collected events and their parameters. Google recommends acquisition reports for attribution because DebugView performs limited attribution analysis. [DebugView documentation](https://support.google.com/analytics/answer/7201382?hl=en). Check an accepted submission, a validation failure, and a repeated visit to the confirmation page. Require the event behavior to match the written trigger. Then require the recipient to find the matching inquiry in the intended destination. Keep the two results separate: an event reached GA4, and an inquiry reached the recipient. Do not close the test until both checks pass. If a test uses a manually tagged URL, label it synthetic. It verifies handling of that tag, rather than proving a genuine ChatGPT visit. ## Reconcile received inquiries without guessing identities Prepare a private worksheet with the receiving team. Agree on the reporting dates and timezone. Keep identifiable messages inside your contact system. Use redacted summaries in the website review. The following fields are proposed requirements. They do not assert that your current form or contact system stores them. | Field | Evidence to retain privately | Reporting use | | --- | --- | --- | | Received inquiry | Matching message and receipt time | Confirm that someone can answer it | | Inquiry purpose | Seller, buyer, property, or general question | Distinguish the requested help | | Website context | Form page and first landing page, if captured | Identify the contact route | | Source evidence | Referrer or campaign values captured by your own form, if supported | Compare with the GA4 source definition | | Visitor's account | Optional answer to how they found you | Preserve self-reported ChatGPT discovery | | Review status | Test, duplicate, spam, genuine inquiry, or unresolved | Keep the counting rule visible | Ask the maintainer whether your form can retain source context in its private submission record. Specify the required fields and test them before relying on them. Preserve the visitor's own answer separately from automatically captured source values. Compare aggregate GA4 inquiry events with received inquiries for the same period and form. Investigate differences using submission and delivery records. Do not identify a buyer or seller from a timestamp and landing page alone. Leave an individual match unresolved without evidence supporting it. If your current form retains no source context, report the totals side by side. Avoid claiming a record-level match. Add an optional discovery question to your proposed form brief if that information would help your team. ## Report the limits beside the result Use **Recorded ChatGPT sessions** for the analytics total. Use **Received inquiries with captured ChatGPT source evidence** for your private submission review. Use **Self-reported ChatGPT discovery** for the visitor's own account. Keep unresolved source records visible. Google's automatic assistant classification depends on a recognized referrer. That condition limits what the channel row establishes. A website session report also cannot count someone who reads an answer and never visits your website. [Assistant traffic measurement](https://support.google.com/analytics/answer/9164320?hl=en#05132026). If a period contains no ChatGPT rows, record **no ChatGPT sessions identified in this report**. Avoid declaring that nobody discovered you through ChatGPT. Check the property, dates, source selection, and measurement test before drawing a website conclusion. Keep answer citations and website inquiries in separate records. A source row establishes recorded attribution under your definition. It does not establish the visitor's prompt or the wording of an answer recommending you. Require that evidence separately. ## Bring the failed check into the website scope The ChatGPT report separates recorded referral sessions from uncertain inquiry attribution. Bring the landing-page failure record to [whole website fix: ChatGPT referral evidence](/for/real-estate). A contact test with receipt evidence is included under the whole website fix. [ChatGPT referral evidence, published scope, checked 2026.10.14](https://scalewithsearch.com/work). - [Test the seller inquiry path on your real estate website](/articles/seller-inquiry-path-agent-website) - [Real estate CRM and website: map forms to contact records](/articles/real-estate-crm-and-website) - [Real estate SEO case study: check the evidence before buying](/articles/real-estate-seo-case-study) ## Related: Real estate - [Test the seller inquiry path on your real estate website](/articles/seller-inquiry-path-agent-website) - [Real estate CRM and website: map forms to contact records](/articles/real-estate-crm-and-website) - [Real estate SEO case study: check the evidence before buying](/articles/real-estate-seo-case-study) ---- ```text .|########||. .|########||. .|########||. |##||. .||##|. |##||. .||##|. |##||. .||##|. |#|. .|#|. |#|. .|#|. |#|. .|#|. |#| |#| |#| |#| |#| |#| .#| |#. .#| |#. .#| |#. .#. .#| .#. .#| .#. .#| |#. .#| |#. .#| |#. .#| |# ...... #| |# ...... #| |# ...... #| .# ||#########| #| .# ||#########| #| .# ||#########| #| .#. |######||######|. .#. .#. |######||######|. .#. .#. |######||######|. .#. #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# # .||####|###########################|. # # .||####|###########################|. # # .||####|###########################|. # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || #. .######|#|#########| .# #. .######|#|#########| .# #. .######|#|#########| .# .#. |#####||||#####| .#. .#. |#####||||#####| .#. .#. |#####||||#####| .#. |# ||########|| #. |# ||########|| #. |# ||########|| #. |# ...... #| |# ...... #| |# ...... #| |#. .#| |#. .#| |#. .#| |#. .#. |#. .#. |#. .#. .#| |#. .#| |#. .#| |#. |#| |#| |#| |#| |#| |#| .|#|. .|#|. .|#|. .|#|. .|#|. .|#|. |##||. .||##| |##||. .||##| |##||. .||##| .||########||. .||########||. .||########||. Scale With Search 2026 [scalewithsearch.com](https://scalewithsearch.com) ```