How Our Marketing Attribution Works

Attribution is only as good as the journey data underneath it, so measurement is validated before any model is applied. A single user identifier persists across sessions and, where the client has a logged-in state, across devices; without that, cross-device journeys are reported as separate people and every model built on them is wrong in the same direction. Touchpoints are collected with consistent channel definitions applied at collection rather than reconstructed later from referrer strings. Several models are reported side by side, commonly first touch, last touch, linear and time decay, because the difference between them is itself the finding: a channel that looks strong only under last touch is closing demand rather than creating it. Data-driven attribution is used only where conversion volume supports it. Results are reconciled against the source system of record, since a model that does not sum to actual revenue is measuring something other than revenue. Modelled and observed conversions are reported separately and never added together as if they were the same kind of number.