Multi-Touch Attribution
Multi-touch attribution distributes conversion credit across multiple ad touchpoints in a buyer's journey, rather than awarding it all to the first or last click.
Common questions
Common questions
- What is multi-touch attribution?
- Multi-touch attribution distributes conversion credit across multiple ad touchpoints in a buyer's journey rather than assigning it all to one. Models include linear (equal credit), time-decay (more credit near conversion), position-based (first and last weighted), and data-driven (ML-based allocation).
- How is multi-touch attribution better than last-click?
- Multi-touch attribution gives partial credit to awareness-stage touchpoints that last-click ignores entirely. This produces a more balanced view of which channels contribute at each stage of the funnel, reducing the systematic under-investment in prospecting that last-click produces.
- What is the main limitation of multi-touch attribution?
- All multi-touch models measure correlation, not causation. A touchpoint appearing on every converting journey receives high credit, whether or not it caused the conversion. Only incrementality testing, which uses a holdout group, can confirm causal contribution.
Multi-touch attribution models attempt to solve the fundamental limitation of single-touch models (first-click and last-click): the reality that most purchases involve multiple ad interactions across multiple channels before a conversion happens. The question multi-touch models try to answer is: how should credit be divided among those interactions?
Common multi-touch models include linear (equal credit to all touchpoints), time-decay (more credit to touchpoints closer in time to the conversion), position-based (40 percent first, 40 percent last, 20 percent middle), and data-driven (machine-learning-based allocation based on which paths historically led to conversion). Google Analytics 4 uses data-driven attribution as its default, replacing the old last-click default.
The limitation of all multi-touch attribution models is that they measure correlation, not causation. A touchpoint that appears on every converting customer's journey will receive high credit in a multi-touch model, but that does not mean the touchpoint caused the conversion. It may simply mean it was difficult to avoid (a branded search ad, for example). Incrementality testing is the only method that isolates causal impact.
Example
A customer journey spans five touchpoints over 14 days: Meta prospecting view, Meta prospecting click, YouTube pre-roll, email retargeting click, branded Google Search click, purchase. A linear multi-touch model splits credit equally across all five. A position-based model gives 40 percent to Meta prospecting (first) and 40 percent to Google Search (last), with 20 percent shared across the middle three.
Always on · Algorithm-led · Human-approved
Walk this operation against your account.
Thirty focused minutes with one of the founders. We map Multi-Touch Attribution against your stack and tell you which entry point fits your stage.