Multi-touch Attribution

Multi-touch attribution splits the credit for a conversion across all the touchpoints a user had along the path to purchase, instead of assigning it to one channel.

Multi-touch attribution is the answer to the weakness of single-click models: instead of one winner, every channel that played a role in the journey gets a share, so both the early discoverer and the final closer are seen. But it carries two costs. First, it needs data connected across the journey — several of one user's sessions across days and devices must be stitched together, and that cross-identification is fragile on today's web because of cookie and privacy limits. Second, “how to split” is itself a decision, not a fact: linear weights every touch equally, time-decay gives more weight to touches nearer the purchase, U-shaped weights the first and last touch most. None is absolutely “right”; the model choice must fit the business logic, otherwise it merely moves the number around without adding insight.