MTA

Multi-touch attribution promises what single-touch models can't: credit for every step of the journey. Instead of crowning the last click, MTA tracks an individual user across touchpoints — the YouTube view, the Google search, the Meta retargeting click — and distributes credit across all of them. Done well, it's the most complete picture of how marketing actually drives conversions. The problem is that MTA depends entirely on identity. To credit a path, you have to recognize the same user across channels and devices — and that recognition relied on third-party cookies and stable device identifiers. Apple's App Tracking Transparency (ATT), third-party cookie deprecation, and broader privacy restrictions have shredded that signal. Large shares of journeys are now unobservable, and the paths MTA can reconstruct are incomplete and biased toward trackable users. That's why many teams raised MTA as the answer a few years ago and then quietly dropped it. The model is sound; the data it needs is disappearing. Most performance teams have shifted toward platform-reported attribution plus incrementality testing and media-mix modeling, treating MTA as directional at best.