What Your Git History Is Trying to Tell You

2026-04-22 · Source: Chris Shayan – Medium · Field: Technology & Digital — Software Development & Engineering, Data Science & Analytics, Artificial Intelligence & Machine Learning · Depth: Intermediate, long

Summary

An engineering leader, after nearly two decades of managing teams, realized a critical blind spot: a lack of visibility into the software development pipeline itself, beyond just output metrics like time to market or MTTR. By integrating LinearB with Git repositories and Jira boards, the author gained data-driven insights into DORA metrics. The analysis revealed a 7-day, 3-hour cycle time, a 15.25 deployments/day frequency, a near-zero change failure rate, and a 17-day, 5-hour Mean Time to Restore (MTTR). A deeper dive into cycle time showed "Pickup" (time between PR opening and review start) as a major bottleneck, consuming 2 days and 8 hours. Planning accuracy was a low 25%, indicating significant discrepancies between planned and completed work. The article also explores an early experiment with AI-powered code review to address the pickup problem.

Key takeaway

For CTOs and Directors of Engineering seeking to improve development efficiency, you must instrument your engineering pipeline with DORA metrics. Focus on understanding the breakdown of metrics like cycle time to pinpoint specific bottlenecks, such as excessive PR pickup time, rather than just aggregate numbers. Use this data to drive structural changes and foster transparency with your teams, enabling them to self-organize solutions and improve overall system health without falling prey to single-metric optimization.

Key insights

Data-driven visibility into engineering pipelines reveals hidden bottlenecks and enables targeted process improvements.

Principles

• DORA metrics provide a holistic view of engineering performance.
• Goodhart's Law warns against optimizing single metrics in isolation.
• Transparency of data empowers engineers to self-organize improvements.

Method

Connect Git and Jira to a tool like LinearB to measure DORA metrics, analyze cycle time breakdowns, and track planning accuracy to identify specific bottlenecks and areas for process improvement.

In practice

• Implement review SLAs to reduce PR pickup time.
• Encourage smaller pull requests to accelerate review cycles.
• Use historical data for estimation calibration sessions.

Topics

• DORA Metrics
• Engineering Performance
• Cycle Time Analysis
• Code Review Optimization
• Planning Accuracy

Best for: CTO, Director of AI/ML, VP of Engineering/Data

Related on AIssential

• The 8 software engineering metrics AI broke — AI-coding tools decouple engineering effort from output, rendering traditional activity-based metrics unreliable and making Goodhart's Law unavoidable.
• The cardinality wall: The hidden data bottleneck for LEO constellations — The exponential growth of LEO satellite telemetry creates a "cardinality wall" that breaks traditional ground system architectures, demanding new distributed approaches.
• Get a Good Return on Your AI Investments — AI amplifies existing software development processes, demanding robust platforms and verification to realize true productivity gains.
• I Spent Years in Data. It Took the Energy Industry to Humble Me.[Part -II] — Energy industry data systems are complex, real-time, and critical for operations, regulation, and revenue, demanding specialized data engineering.
• The Final Bottleneck — Accelerated code generation creates new bottlenecks in review and integration, leading to unsustainable backlogs.
• BI Serving Pointers; Maximizing for Performance and TCO — Databricks' BI serving stack centralizes metric definitions and automates physical data optimizations for consistent, high-performance BI.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Chris Shayan – Medium.