Enterprise AI Measurement Guide
Agent Effectiveness
Team Breakdown
How do we see each engineer's AI coding agent proficiency and cost efficiency side by side, so we can identify who to learn from and who needs development support?
What it shows
The Team Breakdown table takes the Agent Effectiveness metrics to the individual contributor level, showing each engineer's app session count, the Practices and Skills they have demonstrated, MCP Servers and Plugins they use, their total sessions in the period, and their individual Cost per Outcome. This is the per-engineer scorecard that makes both proficiency development and cost management actionable at the person level.
Why it matters
Org-level averages hide the distribution that matters most for intervention. A team with an average Cost per Outcome of $18 might include engineers at $6 and engineers at $70, and the $70 engineer either needs a different approach (a training opportunity) or a spending limit (a governance response). The Team Breakdown table makes that distribution visible without requiring managers to build their own tracking systems.
The Larridin angle
The combination of Practices, Skills, and Cost per Outcome at the per-engineer level is the data foundation for finding a power user, extracting what they do differently, and distributing that across the team. Engineering leaders consistently identify that as their highest-value proficiency investment. Without per-person behavioral data alongside per-person cost data, that use case is a hypothesis rather than an evidence-based program.