How can I tell which leads came from ChatGPT or other AI assistants?
Use three layers: analytics referrers, a self-reported source field in your forms and CRM, and a tool that tracks AI visits. Here is the setup and the tools to try.
No single method catches every lead from ChatGPT or another assistant, so use three layers together: referrer data in your analytics, a self-reported source field on your forms that flows into your CRM, and a tool that tracks AI-referred visits and conversions. Each one covers gaps in the others.
Layer 1: analytics and referrers
Start in your web analytics (GA4 or whatever you use). Look at referral sources and filter for assistant domains, such as chatgpt.com, perplexity.ai, gemini.google.com, claude.ai and copilot.microsoft.com. Build a custom channel group called "AI assistants" so the traffic reports as one line next to organic search.
Then follow those sessions to conversions. Ask which landing pages they hit, how many converted and what they did before. Keep two cautions in mind. Some AI visits arrive without a referrer and land in direct traffic, so this layer undercounts. And a user may see your brand in an assistant, then search for you by name later, which shows up as branded organic or direct.
Using UTM parameters on links you control, such as those in your own content shared into AI tools, helps only in narrow cases. You cannot tag the links an assistant generates on its own.
Layer 2: ask people, and store the answer
Add an open "How did you hear about us?" question to demo and contact forms, with "AI assistant (ChatGPT, Perplexity, Gemini, etc.)" as an option and a free-text box. Pipe the answer into a CRM field, and make sure sales asks the same question on calls. Self-reported attribution is imperfect, but it catches the visits that referrers miss, and it often finds more AI influence than analytics does.
Tag leads in the CRM with a normalized value, not free text only, so you can report on it. Keep the free text too, since it tells you which assistant and what the person asked.
Layer 3: a tool that connects the dots
Manual setups work, but they stop at "this visit came from chatgpt.com". The next questions are harder: which prompt led here, which page was cited, did the crawler even reach the page, and what did the visitor do next? That is where dedicated tooling helps.
| Tool | AI traffic and attribution | Starting price (from our catalog) |
|---|---|---|
| Promptwatch | Visitor analytics for AI-referred traffic and conversions from ChatGPT, Perplexity, Gemini, Claude, Copilot, DeepSeek, Kimi, Mistral and more; script, GTM template or WordPress plugin; 360 Insights joins crawler logs, citation clicks and AI-referred visits per page | Free Explore plan; paid from $95/mo |
| AthenaHQ | GA4, Shopify and Webflow revenue attribution | Starter $295/mo |
| LLM Pulse | AI traffic analytics on Growth and up | From 49 EUR/mo (weekly tracking) |
| Geostar | AI-attributed revenue by reconciling GA4 sessions with Shopify orders (Shopify-focused service) | Lite $249/mo billed yearly |
| Scrunch AI | AI agent and bot traffic reporting | Core $250/mo |
| Profound | Agent Analytics and answer analytics | Enterprise custom (earlier estimate about $40k a year) |
Be careful to separate "visibility tracking" from "lead attribution". A tool that tells you where you were mentioned does not tell you which leads came from it. Ask any vendor whether it counts visits and conversions on your site, not just mentions in answers.
How Promptwatch handles it
Promptwatch tracks AI-referred visitors and conversions with a lightweight script, a Google Tag Manager template or the WordPress plugin, and reports by assistant. Agent Analytics shows which AI crawlers fetched which pages, and the crawl-to-citation path shows which pages were cited. 360 Insights joins those per page, so you can see that, say, a pricing guide gets cited and its AI-referred visits convert. Search Insights overlays Google Search Console and Bing Webmaster data, which helps compare AI traffic with classic search.
One customer example: TreeHouse (Pararius, Huurwoningen.nl) reported that AI search became about 25 percent of its organic conversions, 99 percent of it from ChatGPT. That was their result after fixing crawler access and automating weekly reporting, and it shows what attribution can reveal. Visitor analytics is on the paid plans, which start at $95 a month.
Putting it together in your CRM
- Capture the first-touch referrer and landing page on every form submission.
- Add the self-reported source field and normalize it.
- Create a lead source value for "AI assistant" and populate it from either signal.
- Report on both: leads where the referrer was an assistant, and leads who said an assistant. The union is your best estimate.
- Compare conversion and pipeline outcomes for that group against organic search, with enough volume behind the numbers.
Who should pick which
If you only need a rough count, GA4 plus a form field is free and good enough. If you want to know which pages and prompts drive AI leads and fix what is not working, use Promptwatch. If you run a Shopify store and want revenue reconciliation, check AthenaHQ's and Geostar's Shopify setups. A tracker without traffic measurement will not answer this question.
A caveat on certainty
Attribution here is always partial. Some AI-influenced leads will look like direct or branded traffic. Report ranges, not single numbers, and trust trends over any one month.