Accelerate
by Nicole Forsgren, Jez Humble & Gene Kim
The book that turned 'good engineering org' from opinion into measurable claim.
- Status
- Read · August 2020
- Bought
- August 21, 2020
- For
- Engineering leaders defending delivery practices with data · Platform and SRE teams choosing what to measure · Anyone tired of arguing about DevOps from anecdote
Where it earned its place
The one-paragraph verdict
What sets this apart from the usual DevOps shelf is that it is research, not testimony: Forsgren ran the State of DevOps survey through real statistical methods and came out with the four DORA metrics — deploy frequency, lead time, change-fail rate, time to restore. That gives you something rare in this field, a claim you can actually defend in a leadership meeting. The cost is that the book is front-loaded: Part 1 is the gold, Part 2 is a methods appendix that reads like one, and the whole thing repeats its own thesis more than it needs to. You can get 90% of the value from the first hundred pages. But that 90% reframed how I argue about engineering quality.
Who should read it
Read it if you lead or influence delivery practice and keep losing the argument by anecdote — it hands you the vocabulary and the evidence. Platform and SRE teams choosing their first dashboard should start here. Skip it if you want a hands-on CI/CD implementation guide; this is the why measure, not the how to build.
Where it earned its place
The DORA framing is exactly why I treat pipeline work as product work, not overhead — the argument I make in engineering investment in the roadmap. And the four metrics are the reason BlueRobin’s GitHub Actions CI/CD optimises for lead time and low change-fail rate rather than raw build speed.
Skip it if…
You already live and breathe the DORA metrics, or you want code on the page. This book sells you the measurement; it does not hand you the implementation. [Accelerate on Goodreads]