Understand adoption
Review assistant usage and engagement across teams. Find where people use the tools and where adoption needs a closer look.
AI Coding Assistant Impact
See what changes when your teams use AI
Compare AI adoption with review time, lead time, quality, and rework. Follow the engineering outcomes behind increased assistant usage.
AI Impact
Adoption alongside delivery and qualityAI Impact Overview
How AI-assisted development relates to engineering outcomes
Coding Efficiency67.7%↑ 7.3%
Higher AI adoption coincided with improved coding efficiency.
Coding EfficiencyExplore the trend
- Security
- A
- Maintainability
- A
- Reliability
- C
- Technical debt
- 86d
Select a metric to compare its trend and read the insight alongside it.
Explore impact by coding assistant
From signals to decisions
Evaluate adoption and outcomes in the same review
Start with usage, then inspect the delivery and quality signals that matter to your team.
Compare delivery outcomes
Read AI contribution alongside PR time to merge, lead time, and quality. Follow the trend over time before deciding what changed.
Look for rework and risk
Inspect code churn, review pressure, and defects as adoption grows. Use the evidence to focus enablement and workflow improvements.
What to look at
Ask what improved, and what needs attention
Use the trend to focus the investigation. Compare the same team, scope, and period before drawing conclusions.
- AI adoption and contribution
- Pull request time to merge and lead time
- Code churn and coding efficiency
- Defects and code quality findings
- Is AI helping changes move through review?
- Compare AI adoption with PR time to merge. Follow a slowdown into review activity and the changes waiting for attention.
- Is more AI-written code leading to more rework?
- Read AI code share alongside churn and coding efficiency. Review task type and recent changes to understand the pattern.
- Are delivery gains accompanied by stable quality?
- Inspect defects and quality findings alongside lead time and production delivery. Review the same team and period consistently.
From signals to decisions
Bring code origin, cost, and team context together
Connect the signals needed to understand AI-assisted development from the change itself to the production outcome.
Code origin
Blamely records AI and human contribution context in Git. Use attribution to understand the work behind an adoption trend.
Usage and cost
Compare consumption, credit forecasts, and spending with adoption and delivery. Include the cost of the tools in your evaluation.
Team context
Review work distribution, coding efficiency, and active symptoms with the team. Decide where to investigate next.
Common questions
AI impact measurement questions
Need to discuss your team or technical setup? Contact us.
What does AI Impact measure?
AI Impact connects coding assistant adoption and usage with engineering signals such as pull request flow, lead time, code quality, and rework. Use the combined view to investigate changes in engineering outcomes.
Which assistants can we evaluate?
Oobeya supports AI-assisted development measurement across tools such as GitHub Copilot, Cursor, Claude, and Codex (including Enterprise). AI Cost monitoring supports Claude, Codex, and GitHub Copilot. Available adoption, attribution, and financial measures depend on the connected tools and capture methods.
Can we compare AI usage with PR time to merge?
Yes. Read AI adoption alongside pull request flow and time to merge, then inspect the underlying review activity and work type for context.
How does Blamely add attribution context?
Blamely captures the origin of supported AI-assisted edits and stores attribution in Git notes. That context helps teams investigate how AI-written changes relate to review, rework, and delivery in Oobeya.
Does a correlation prove that AI caused an improvement?
No. Changes in delivery can also reflect task mix, team changes, process improvements, and other factors. Compare consistent periods and investigate the work behind the trend.
How should we evaluate AI investment?
Review adoption and cost alongside delivery, quality, and rework. Use the evidence to guide rollout and enablement decisions rather than treating usage or code volume as a complete ROI measure.
See Oobeya in practice
Review AI impact with your engineering priorities
Walk through adoption, attribution, delivery, quality, and cost with your toolchain and team structure in mind.