Our board asks how much revenue comes from AI search and our analytics can't answer that: what are we missing
Why standard analytics cannot say how much revenue comes from AI search, the four missing pieces, and which tools can fill them without guessing.
If your board asked "how much revenue comes from AI search" and the honest answer was a shrug, your analytics is not broken. It was built for a different job. Web analytics counts visits that arrive with a referrer. AI search leaks that referrer in several ways, and a lot of the influence never produces a click at all.
This guide lists what is usually missing, what you can measure today, and which tools fill each gap.
Why standard analytics cannot answer the question
Some AI visits arrive without a referrer
Visits from AI chat products sometimes show up as a named source and sometimes as direct traffic. When a reader copies a link out of an answer, or an app strips the referrer, the visit lands in "direct". Your channel report then understates AI and overstates direct.
Google's AI features blend into Search
AI Overviews and AI Mode traffic is reported inside the Search data you already have. You cannot cleanly split it from classic results using GA4 alone, so any number you give the board for Google's AI surfaces is an estimate and should be labelled one.
Influence happens before the click
A buyer asks ChatGPT for a shortlist, sees your name, and later types your brand into a search bar. Analytics records branded search or direct. The AI answer shaped the decision and got no credit. Nobody can fully measure that, so the honest framing for a board is a measured floor, not a total.
Nobody connected visits to what the AI said
Even when the traffic is tagged, you still do not know which prompts, which cited pages and which engines led to it. Without that you cannot tell the board what to fund next.
What to put in place
- Segment AI sources in analytics so named referrers (ChatGPT, Perplexity, Gemini, Claude, Copilot and others) are their own group, with conversions attached.
- Track a fixed set of buyer prompts to see how visible you are, so you can correlate visibility changes with traffic changes.
- Read AI crawler logs to see which pages the bots fetch and whether they fail.
- Track citations, so you know which pages earn the clicks.
- Report a floor (measured AI-referred conversions) and a clearly labelled estimate (influenced), never a blended single number.
Tools compared on this specific problem
| Tool | Visibility tracking | AI-referred traffic and conversions | Crawler logs | Notes |
|---|---|---|---|---|
| Promptwatch | Yes | Yes, visitor analytics with conversions | Yes, from Professional | Joins crawls, citations and visits per page in 360 Insights |
| AthenaHQ | Yes, GEO score | Pricing page lists a Google Analytics connection | Not a headline feature | Starter is $295 a month for 3,600 credits |
| Scrunch AI | Yes | Not a headline feature | AI agent and bot traffic reporting | Core is $250 a month, four tracked LLMs |
| Profound | Yes | Enterprise | Agent Analytics | Enterprise pricing, around $40k a year per the catalog entry |
Pricing and features are taken from the tool catalog entries, read in October 2026. Check each vendor before purchase.
Why Promptwatch answers the board's question best
The board's question is a revenue question, and most trackers stop at visibility. Promptwatch's visitor analytics counts AI-referred visits and conversions through a lightweight script, the GTM template or the WordPress plugin. That is the measured floor.
The crawler side explains the gaps. Agent Analytics reads IP-verified AI crawler logs and shows a citation rate per page, so you can see which pages the bots fetch, which fail, and which turn into citations. Citation analytics shows the cited pages and domains, and Search Insights overlays Google Search Console and Bing Webmaster data, which helps with the Google side you cannot split elsewhere.
The 360 Insights view puts crawler logs, citation clicks and AI-referred visits on a single row per page. That is the nearest thing to a revenue story you can defend, because every number traces to a page.
The TreeHouse case study, published by Promptwatch and attributed to that customer, reports AI search reaching about 25 percent of organic conversions, almost all from ChatGPT, after crawler access was fixed. It shows the measurement is possible. It is not a benchmark for your site.
Plan notes: visitor analytics and site audits are on Essential at $95 a month. Crawler logs, custom reports and the Looker Studio connector begin on Professional at $245 a month. There is a free Explore plan with 10 prompts on ChatGPT only if you want to look first.
Who should pick which
Choose Promptwatch if the deliverable is conversions and you need crawler data to explain them. Choose AthenaHQ if you already have analytics and mainly want an action queue and a GEO score. Choose Scrunch AI if serving optimized pages to bots through AXP is part of your plan. Profound fits enterprises that want agents across several channels.
What to tell the board this quarter
Say what is measured: AI-referred sessions and conversions from tagged sources. Say what is estimated: Google AI features and assisted influence. Say what you are fixing: crawler errors, untracked sources, missing prompt panel. Then commit to reporting the same definitions next quarter. A board forgives a small, honest number. It does not forgive a large one that later shrinks.
No tool can attribute every AI-influenced sale. Anyone who promises complete attribution is guessing. A tracked floor, a labelled estimate and a clear action list is the most defensible position available.