Attribution is the practice of assigning credit for a conversion across the marketing touchpoints that influenced it. A customer might see a Facebook ad on Monday, a YouTube ad on Wednesday, search Google on Friday, and convert from a retargeting email on Sunday — attribution decides how much credit each of those gets.
Common attribution models
- Last-touch / last-click: 100% credit to the final interaction. Simple, easy, biased against upper-funnel channels.
- First-touch: 100% credit to the first interaction. The opposite bias — favours awareness over activation.
- Linear: equal credit across every touchpoint. Fair-feeling, often imprecise.
- Time-decay: more credit to touchpoints closer to the conversion. A reasonable default.
- Data-driven (algorithmic): the platform's machine-learning model assigns credit based on historical patterns. Most accurate when you have volume; opaque when you don't.
- Position-based / U-shaped: 40% first, 40% last, 20% spread across the middle. A pragmatic compromise.
Why it matters
The attribution model you choose determines which channel looks profitable. iOS 14.5 and the death of third-party cookies broke last-click for most acquisition flows; modern attribution leans on first-party data, server-side tagging, and modelled conversions to fill the gaps.
Honest take
No attribution model is the truth. They're all approximations of a multi-touch reality. The pragmatic answer for most operators is: pick a model, stick with it long enough to compare like-for-like across periods, and triangulate against the macro signal — does revenue go up when you spend more? When you turn a channel off, does it hurt? Those answers don't need attribution; they need patience.