Data-driven attribution

Data-driven attribution allocates conversion credit using a model informed by patterns in measured interactions rather than only a fixed rule such as last click. GA4 describes a model using property data for each key event, which can assign fractional credit across several interactions.[1] Those fractions describe allocated reporting credit, not fractions of physically placed orders.

A store can use this approach to investigate channels appearing earlier in the journey. Interpret results within the available measurement coverage and changing channel mix. Data-driven credit is not equivalent to causally proven additional sales: lift measurement compares a treatment group with a group withheld from the advertising.[2] Attribution and experiments therefore address different, complementary business questions.