What is attribution modelling?
Attribution modelling is a marketing analytics method for deciding how much credit each touchpoint a customer interacted with, an ad, a search click, an email, deserves for an eventual conversion.
Models like first-touch, last-touch, and multi-touch all exist to answer the same question in different ways: which of the several things a customer saw before buying actually mattered. They were all built on an assumption baked into the analytics tools themselves, that every touchpoint worth crediting left a trackable click or visit behind it.
An AI-generated answer that fully satisfies a customer's question, with no click and no visit, is invisible to every one of these models by construction. It cannot be the first touch, the last touch, or any touch in a multi-touch chain if no analytics event was ever recorded. As more research happens inside AI conversations, a growing share of what actually influenced a customer sits permanently outside what any attribution model can measure, not because it did not matter but because there was nothing to log.
Related
- Dark trafficDark traffic is visits that arrive at a site with no identifiable referrer, so analytics tools file them under direct even though the visitor actually came from somewhere else.
- Conversion rateConversion rate is the share of visitors to a site who complete a desired action, such as a purchase or a signup, out of everyone who arrived.
- Zero-click searchA zero-click search is one where the user gets what they needed from the results page or an assistant's answer and never visits a website.
Want to know where you actually stand on this? Run a free visibility check or try the free tools.