Last-click attribution is lying to you — and it's costing you money. When you give 100% conversion credit to the final touchpoint before purchase, you systematically underfund the channels that build awareness and consideration. You also consistently overfund bottom-of-funnel channels that close deals but don't create demand.
Attribution Models Compared
There are six common attribution models, and each tells a different story. Last-click and first-click are the simplest but most misleading. Linear distributes credit evenly. Time-decay gives more credit to recent touchpoints. Position-based (40-20-40) rewards first and last touch. Data-driven uses machine learning to assign credit based on actual conversion patterns.
- Last-click: Simple, but over-credits paid search and direct
- First-click: Good for awareness channels, ignores nurture
- Linear: Fair but dilutes high-impact touchpoints
- Time-decay: Good for short sales cycles
- Data-driven: Gold standard — requires 3,000+ conversions/month
Incrementality Testing: The Truth Serum
Attribution models tell you correlation. Incrementality tests tell you causation. A geo holdout test (running ads in some markets, not others) or a ghost ad study (showing fake ads to a control group) reveals the true lift your campaigns drive beyond what would have happened organically.
“If you're not running incrementality tests, you don't actually know which of your campaigns are working. You just have a story that fits the data.”
— Omar Farouk, Analytics Director”