Shahin

Time-decay Attribution

Intermediate
Updated Jul 11, 2026 Shahin editorial 2 min read
PronunciationEN/taɪm dɪˈkeɪ ˌætrɪˈbjuːʃən/

Time-decay attribution splits a conversion's credit among touchpoints but gives more weight to touches closer to the moment of purchase.

Time-decay attribution comes from the assumption that the closer a touch is to the purchase the more effective its role in closing, so credit grows exponentially from the past toward the present. It sits between two extremes: unlike linear it does not weight all equally, and unlike last-click it does not give everything to one point. For short-to-medium purchase cycles where recency really matters — a promotional campaign with a deadline, say — it is a reasonable choice. But its bias is systematic: top-of-funnel channels, the early discoverers, are always under-weighted, so budgeting by this model risks starving brand awareness. The decay speed (the half-life) is itself a tunable parameter, and like any multi-touch model, running it requires stitching the user's sessions.