💡

InsightHunt

Hunt the Insights

N

Nicole Forsgren

Partner at Microsoft Research

Microsoft Research (formerly GitHub, Google/DORA)

⚡ Execution (1)👥 Team & Culture (1)🎯 Product Strategy (1)

Key Takeaways

  • 1.Speed and stability move together; shipping smaller batches more frequently reduces the 'blast radius' of errors.
  • 2.Developer satisfaction is a leading indicator of productivity; when satisfaction drops, system reliability often follows.
  • 3.Never measure developer productivity by lines of code or sheer activity metrics; it incentivizes the wrong behaviors.
  • 4.Data from people (surveys) is often more accurate than data from systems (telemetry) because systems miss context (e.g., work done outside the repo).
  • 5.AI is shifting developer work from writing code to reviewing code, which changes how we must measure productivity (focusing more on flow and cognition).
  • 6.To improve strategy, start with 'Words' (goals) before 'Data' (metrics) to ensure you are measuring what actually matters.

Methodologies(3)

⚡ Execution

A widely adopted framework that uses four key metrics to classify engineering performance. It proves that high performers do not trade off speed for stability; they achieve both by reducing batch sizes and automating processes.

Core Principles

  • 1.Deployment Frequency: How often code is deployed to production (Elite: On-demand).
  • 2.Lead Time for Changes: Time from code commit to code running in production (Elite: < 1 day).
  • 3.Mean Time to Restore (MTTR): Time to restore service after an incident (Elite: < 1 hour).
  • +1 more...

"Speed and stability move together... If you push changes less frequently, you will have more unstable systems."

#(four#keys)#execution
View Deep Dive →
The SPACE Framework

by Nicole Forsgren

👥 Team & Culture

A multidimensional framework for measuring productivity. To use it correctly, teams should select metrics from at least three of the five dimensions to ensure a balanced view and avoid gaming the system.

Core Principles

  • 1.Satisfaction & Wellbeing: How developers feel about the tools and culture (e.g., Burnout, eNPS).
  • 2.Performance: Outcomes of the process (e.g., Reliability, Velocity).
  • 3.Activity: Count of actions (e.g., # of Commits, CI runs) - strictly coupled with other metrics.
  • +3 more...

"If you're talking about culture, this is totally different than if you're talking about friction in toolchains."

#space#team#culture
View Deep Dive →
🎯 Product Strategy

A visual tool to validate metrics. It separates the conceptual hypothesis (Words) from the measurement implementation (Data) to identify bad proxies or spurious correlations.

Core Principles

  • 1.Top Row (Words): Define the relationship conceptually (e.g., 'Customer Satisfaction' -> 'Return Business').
  • 2.Bottom Row (Data): Define the metrics acting as proxies (e.g., 'NPS' -> 'Retention Rate').
  • 3.Validation Step 1: Do the 'Words' actually make sense strategically?
  • +2 more...

"Always start with words. You do not start with data."

#four-box#(words#data)
View Deep Dive →