AI citation tracking that connects
every citation to revenue

AI citation tracking is the process of monitoring whether large language models — ChatGPT, Claude, Gemini, Perplexity, and their peers — name your domain or brand inside the answers they generate. A citation is the AI-search equivalent of a SERP position: it is the visible inclusion of your site as a source. Citation tracking logs the event each time it happens, attributes it to a specific prompt and engine, and trends it over time. The honest version of citation tracking ends one step further than the count: it joins the citation to the click that follows and then to the Stripe payment that the click eventually produces. Without that join the citation is a presence metric, not a performance metric.

Citation tracking is the longitudinal log: prompt P was scanned on engine E at time T and your domain was either cited or not, alongside the full list of sources the engine cited. Citation analysis is the cross-sectional interpretation: across your scan history, your domain shares citation real estate with the competitors you track, you over-index on ChatGPT, you under-index on Gemini, and 22% of all your citations are concentrated on three prompts. Tracking gives you the event stream; analysis gives you the patterns and the budget implications. Attrifast does both in the same dashboard, but the value of analysis is fully realized only when revenue is layered on top — which is the part most analysis-only tools omit.

Because not every citation produces a click, and not every click produces a paying customer. Claude in particular often references your brand in a summary without a clickable link at all, which is a brand asset but not a revenue line. ChatGPT and Perplexity link out far more reliably, but the click rate from an AI answer is materially different from the click rate of a top-three Google organic result. Then the visitors that do click vary wildly in conversion behavior based on the prompt intent (buying vs research) and the engine. In our own measurement across instrumented sites, the dispersion between "got cited" and "got paid" is wider than most teams expect. The fix is to report both numbers side by side and let revenue decide which citations to chase.

Mechanically: (1) The citation scan logs which engine cited your domain on which prompt at which time. (2) The Attrifast tracking script on your site captures the referrer server-side on every visit — chatgpt.com, perplexity.ai, gemini.google.com, etc. — and stores a first-party session id. (3) When that session later completes a Stripe checkout, the payment_intent.succeeded webhook fires and Attrifast joins the payment to the originating session and back through the session to the engine. (4) Reporting aggregates the joined revenue per AI engine, shown alongside the per-prompt cited/uncited matrix and the top cited sources list. The chain depends on capturing the referrer before GA4 strips it into Direct, which is why server-side first-party capture matters — see dark AI traffic in GA4 for the deeper walkthrough.

ChatGPT (with web search), Claude, and Google Gemini — with more engines, including Perplexity, rolling out. Each engine has different citation density and link-out behavior, so each is scored separately. ChatGPT with web search returns structured results, which makes it the engine where citation parsing is cleanest. Claude is the hardest to attribute because it most often summarizes without a link. Gemini sits in the middle.

Profound's Answer Engine Insights is the deepest enterprise citation tracker in the market, with strong source-share analysis and a public Index leaderboard. Geoptie ships free per-LLM rank trackers — useful as a top-of-funnel signal but not joined to your site analytics. Attrifast is not trying to out-build Profound on the citation side. We pair a competent citation tracker with the part both products leave open: the Stripe webhook join that tells you which AI engines' citations produced paid trials. If you need the deepest citation analytics, Profound. If you need to defend a marketing budget at SMB scale, Attrifast.

Yes. Every scan parses the full cited-source list each engine returns, not just your own domain. You declare your competitor set during onboarding — up to eight competitor domains on Pro, three on Starter (we suggest a starting set based on your category) — and the dashboard then shows competitor citation share alongside yours across your tracked prompts. We cannot report competitor revenue — only your own Stripe webhook joins back — but the citation-share leaderboard is fully visible in the dashboard.

Attrifast starts at $9.99/mo on Starter — 10 prompts monitored, rescanned every 30 days — and the Pro plan is $49/mo for up to 30 prompts monitored, rescanned every 10 days across ChatGPT, Claude, Gemini, and Perplexity. Both include a 7-day free trial, full first-party tracking on your site, and Stripe webhook attribution. For comparison, enterprise citation tracking platforms are typically sold via demo with annual contracts in the four-to-five-figure-per-month range; the open-priced AI-visibility tools sit in the $50-$200/mo entry band. We are deliberately the cheapest tool in the category that also does the revenue join — that is the wedge.