GitHub’s Engineering System Success Playbook

Published via GitHub Executive Insights

At GitHub, we know that better business outcomes aren’t driven just by good-quality code, speed, or developer happiness in isolation. It’s actually when quality, velocity, and developer happiness are working in unison that organizations see their best results. If you're looking for engineering to provide greater value to your business, it’s crucial to strengthen these — let’s call them — foundational zones, and create better conditions for your teams to thrive.

This is the crux of GitHub’s Engineering System Success Playbook (ESSP) — a three-step process that can help you drive meaningful, measurable improvements in your organization, whether you’re looking to adopt a new AI tool like GitHub Copilot or identify and unlock bottlenecks that have been hindering performance.

Inspired by multiple frameworks, including SPACE and DevEx, DX Core 4, and DORA, our playbook offers a balanced and comprehensive approach, helping you assign metrics to each “zone” that you can track over time and iterate as needed.

Here’s a quick breakdown of the process:

• Step 1: Identify the current barriers to success
• Step 2: Evaluate what needs to be done to achieve your goals
• Step 3: Implement your changes, monitor results, and adjust

At the heart of our ESSP is a systems thinking approach that prioritizes long-term, sustainable improvements. While quick wins can be a great way to get an initiative started, they can produce negative downstream effects. For example, accelerating code review turnaround time can speed up development, but without addressing the broader system – like testing infrastructure and documentation practices – you may risk creating bottlenecks downstream and compromising code quality.

This project was created in response to many customer requests for prescriptive guidance on creating meaningful downstream impact from changes in their engineering systems — often with the introduction of GitHub Copilot. We also engaged with DevEx and DevOps metrics vendors to understand both the challenges and successes they’ve experienced while helping customers elevate engineering performance or to justify the investment in GenerativeAI. So these steps were created to balance the inherent complexity of engineering success with practical, achievable steps for teams, including those earlier in their improvement journey.

In our playbook, we outline suggested metrics to monitor as part of your improvement efforts for each zone. Keep in mind that these metrics are downstream, or lagging metrics, and in the majority of cases should be complemented with leading metrics. Both leading and lagging metrics may be measured using telemetry and/or survey data, depending on your context, and the way these metrics are calculated will depend on your teams’ engineering workflows and the systems supporting them.

As you dig into our playbook, we encourage you to keep a few concepts in mind:

• Always bring a team perspective to improvement
• Select and use metrics with care to avoid gamification
• Balance the cost of measurement with the benefits of measurement
• Focus on improvements over time rather than overindexing on benchmarks

Engineering teams have the potential to fuel incredible change and accelerate business outcomes.

With GitHub’s ESSP, you can unlock engineering’s potential through creating a culture of excellence that inspires and supports engineers to do their best work. Learn more in our eBook and one-pager.