Agent Effectiveness
How well engineers use AI coding agents: spend, outcomes, session depth, fluency, and cost per person.
- Agent Effectiveness Score What seven dimensions make up the Agent Effectiveness Score breakdown, and what does the score show about AI coding agent use?
- Outcome Success What percentage of AI coding agent sessions produce a merged pull request in the selected period?
- Cost per Outcome What does the Cost per Outcome figure show for AI coding agent activity, and how does it relate to Outcome Success?
- Session Depth What is the average number of prompts or turns per AI coding agent session in the selected period?
- Spend How does the amount we're billed for AI coding agents compare to what our telemetry shows we actually consumed, and what does the gap tell us about our agent cost governance?
- App Adoption How do we compare the cost-efficiency of every AI coding agent we're running, across adoption, sessions, spend, and outcomes, in a single view?
- Model Spend Breakdown Which foundation models are our AI coding agents calling, how much is each one costing, and are we routing to expensive frontier models for tasks that cheaper models handle equally well?
- AI Fluency Score Summary What specific engineering behaviors and infrastructure elements make up our AI Fluency Score, and which ones should we invest in developing to improve 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?