What is marketing attribution?

Q: What is marketing attribution?

Marketing attribution is the method of crediting a conversion to the different touchpoints a customer had with your brand. Common models: last click, first click, linear, data-driven. Attribution has become more complex with iOS 14.5 and the end of third-party cookies.

Detailed explanation

Marketing attribution answers a simple but crucial question: which channel or campaign actually caused a customer's conversion? In a typical purchase journey, a customer interacts with multiple touchpoints (Meta ad, Google result, email, direct, retargeting) before ordering. Attribution determines how to distribute credit for the sale among these touchpoints to correctly measure each channel's ROI.

The most common attribution models are: last click (100% to the last touchpoint), first click (100% to the first), linear (equal split), time decay (more weight to later touchpoints), position based (more to first and last) and data-driven (algorithm distributing based on historical data). Each model has its biases: last click underestimates discovery and overestimates retargeting, first click does the opposite.

Since iOS 14.5 and App Tracking Transparency, attribution has become significantly more complex. Loss of tracking signals reduces accuracy of platforms like Meta and Google, which often show underestimated conversions. The progressive end of third-party cookies worsens the problem. E-commerce operators must now combine multiple approaches: server-side tracking (CAPI), first-party models, and incrementality measurement to obtain a reliable view.

The modern solution combines robust server-side tracking (to send conversions to platforms with proper data), multi-touch attribution (to understand the full journey), and blended measurement (global ROAS and POAS based on real store figures). This triple approach enables steering acquisition while accounting for current technical limits and making better budget decisions.

Concrete example

A customer sees a Meta ad (click), returns 3 days later via Google Search (click), receives an email (open) and converts direct by clicking their history. In last click, 100% credit goes to direct. In first click, 100% to Meta. In linear, 25% each. In data-driven, the algorithm might attribute 40% to Meta, 30% to Google, 20% to email and 10% to direct based on historical patterns.