Skills

  • Math & modeling

  • Detection

  • Analysis

  • Output

  • Delivery

Skill · multi-touch-attribution

Multi-touch attribution

Calibrated

Data-Driven (removal-effect), Last-touch, Linear, Time-decay, First-touch.

how it works

Computes credit per touchpoint across five attribution models. Data-Driven uses removal-effect on a Markov walk of paths; the other four are deterministic rules. Refit nightly against closed-won deals.

used by

  • Daily Brief
  • Anomaly Hunter
  • Revenue Optimizer
Demo data