Agile Estimating and Planning: A Comprehensive Guide to Agile Success

Mike Cohn nails the hard truth about software estimation: it's not a science. It's an art. And most teams get it wrong because they treat uncertainty as a problem to eliminate rather than a reality to manage.

This book gave me a framework I still use. Story points, planning poker, velocity tracking — Cohn explains each technique with enough depth to actually implement them. The emphasis on collective estimation resonated. When the whole team estimates together, you get better accuracy and shared ownership. I've seen this play out in every team I've led.

The release planning section is where the book earns its keep. Cohn lays out a clear approach: set priorities, define an MVP, gather feedback early, adjust. This sounds obvious. It's not. Most teams I've worked with struggle to separate "nice to have" from "must ship." Cohn gives you the tools to have that conversation.

His stance on flexibility is the right one. Plans should change. If your plan from three months ago still looks exactly the same, you're not learning anything from the work. Cohn makes this point convincingly.

Where the book falls short: it assumes a fairly standard agile setup. If you're dealing with legacy systems, cross-team dependencies, or organizational resistance to agile, you'll need more than what's here. The book also doesn't address estimation in high-uncertainty domains like R&D or infrastructure migration, where even relative sizing breaks down.

Some of the content feels dated now. The agile landscape has evolved — continuous delivery, NoEstimates movement, shape-up methodology — and Cohn's framework doesn't account for these shifts.

That said, the fundamentals hold. If your team struggles with estimation or planning, start here. It's practical, opinionated, and grounded in real experience. I'd hand this to any new engineering manager on day one.