An attribution model is the rule set that decides which touchpoints get credit for a conversion — last click, first click, data-driven, or position-based — and every model tells a different story about the same revenue.
Definition
Attribution answers 'which ad gets credit' by rule: last-click gives everything to the final touch, data-driven models distribute credit algorithmically across the observed path, and platform-side attribution counts within each platform's own windows (click-through and view-through). The structural problem: each platform attributes in its own walled garden, so summing platform-reported conversions double-counts — three platforms can each claim the same sale.
Why it matters
Attribution isn't a measurement detail; it's the routing table for budget. Change the model and channel 'performance' reshuffles overnight without a single dollar performing differently. Teams that know which model produced each number argue about strategy; teams that don't argue about dashboards.
In agentic execution
Adlyse's cross-platform layer reads performance in one consistent frame rather than three self-graded ones — so when agents propose reallocations, the comparison behind them is apples to apples, and a human approves what changes.
Related terms
See these concepts running on your own accounts.
Start with a free audit and see what an agentic workforce would change first.
Start a free audit