GitKraken Insights

Find out what AI is actually delivering.

Last updated: September 2026

GitKraken Insights gives engineering leaders a single view of how AI tools, code delivery, and team capacity work together. The AI Adoption section measures how much your team is actually using AI, how autonomously, and what AI is actually delivering — in time and dollars.

Start where you sit
Executive

Prove the investment

Adoption, productivity uplift, and the CapEx/OpEx split.

Engineering leader

Find the bottleneck

DORA, cycle time, and review load across your teams.

Team lead

Coach the team

Per-developer adoption, without it becoming a performance review.

Admin

Connect the data

Git providers, AI tools, trackers, and the settings that move scores.

The metric map

21 metrics · 5 families
Adoption & Agentic5 metrics
Agent Adoption ScoreAgent Autonomy ScoreAI TierMaturity FactorCursor Boost
Output & Throughput4 metrics
Output ScoreThroughputDirect CommitsEffort Score
Flow & Cycle Time4 metrics
Cycle TimeReview CyclesFirst-Pass RateWIP
DORA & Quality4 metrics
Deployment FrequencyLead Time for ChangesChange Failure Rate (CFR)Mean Time to Recovery (MTTR)
AI Impact & Cost4 metrics
Productivity UpliftAI-Assisted PercentageCapEx / OpEx SplitSpend by Tier

Dashed marks a setting or modifier that changes other scores rather than being a score of its own.

What Insights answers for your team

Business Impact & ROI

Change in productivity, additional developer hours, and capacity returned — in dollars. Per team, per tool, per developer. The numbers your CFO needs to justify ongoing AI spend.

Business Impact dashboard showing productivity change, additional developer hours, cost impact, and AI-assisted changes

Adoption & Autonomy

Two scores per developer. Adoption measures daily use, depth, and output. Autonomy measures how much of the work is handed off to agents. Cohort views surface who is leading, who is learning, and who needs help before the rollout stalls.

Top developers table ranked by AI adoption and autonomy scores with daily activity trend

Repo Readiness

Every repo scored on what agents need to work well: AI instruction files, CI configuration, test coverage, and build reliability. Action items when something is missing. An agent is only as good as the repo it works in.

Repository readiness cards showing AI instruction coverage scores and action items

Tool & Model Performance

Claude Code, Cursor, Codex — plus every model powering them. Ranked per developer, per team, per task by cost, cycle time, PR size, and bugs shipped. The scoreboard you open when someone asks which tools to cut.

Comparison cards for Claude Code, Copilot, Codex, and Windsurf with adoption and cycle time metrics

Developer-Level View

A private view for each developer showing how they work with AI: adoption, autonomy, cost, and outcomes — benchmarked to your org, not used to name and shame. Plus personalized coaching on which models and skills to reach for next.

Individual developer profile showing AI adoption score, top skills by value, and coaching recommendations

Plan: GitKraken Insights
Platform: Browser only via gitkraken.dev
Role: Lead, Admin, or Owner

Where AI Adoption shows up in the product

  • /ai-adoption/developers — primary surface. Each row has Adoption, Agentic, Tier, and a heatmap.
  • /ai-adoption/teams — team averages and tier mix bars.
  • /ai-adoption/ai-tools-comparison — cohort comparisons (e.g. team A vs. team B, or Claude vs. Codex users).
  • /ai-adoption/executive — hero KPI (“AI Adoption %”) and trend lines.
  • /ai-adoption/ai-impact — autonomy deep dive and Business Impact / ROI.
  • /ai-adoption/capex — CapEx / OpEx split as the primary surface.
  • Adjacent surfaces in the same nav family: /ai-adoption/board-metrics, /ai-adoption/data-connections, /ai-adoption/data-explorer, /ai-adoption/settings/*.