Multi-layered attribution

High-fidelity API data
Direct integration with tools like Cursor. The most accurate signal available.
Git co-authorship
Attribution from commit metadata where API data isn't available.
Time-series bucket algorithm
Fallback inference. Still directionally accurate, clearly labeled as estimated.
What AI Usage tracks

Foreground vs. Background AI
Usage categorized into foreground AI (interactive IDE assistants) and background AI (autonomous agents). Distinguish active pairing from automated assistance.
AI output percentage
See the exact portion of your codebase generated or modified by AI. Track how that percentage changes over time by team, repo, or individual.
Efficiency correlation
Understand how AI usage tracks against engineering output, bug ratios, and cycle times. Know whether higher adoption is producing better outcomes — or just more volume.
Cost per active user
See which teams are getting the most value per dollar spent. Connect subscription costs to actual usage and output data.
Know what's being used, by whom, how much it costs, and whether it's working.
AI Usage gives you the full picture.

