Five Core Metrics: The Intelligence Behind Successful Software Management

Rating:
Summary: A "reader friendly" instructional how-to guide
Review: Collaboratively written by Lawrence H. Putnam (President of the software management consulting firm Quantitative Software Management) and Ware Myers (a professional independent consultant and contributing editor of "Computer" and "IEEE Software"), Five Core Metrics: The Intelligence Behind Successful Software Management is a "reader friendly" instructional how-to guide to utilizing the reliable development processes and techniques that help software managers efficiently allocate limited resources and carefully track progress, ensuring optimum quality software with a minimum of wasted effort. Five core metrics of Time, Effort, Size, Reliability, and Process Productivity are introduced as a means to measure and adjust ongoing processes to constantly changing real-world conditions. An exceptional business guide in its field, Five Core Metrics is highly recommended reading for anyone charged with the responsibility of using and creating software projects using or incorporating metric measurements.

Rating:
Summary: Evolution and refinement of earlier work
Review: The authors have impeccable credentials in the software estimating discipline, with Putnam's experience dating back to his breakthrough approach using Rayleigh curves to model staffing developed in the early 1970s, and Myers as his coauthor and collaborator for three earlier books from which this one is roughly based and represents a distillation and refinement of earlier ideas.

Material in this book is not done justice if you go solely by the table of contents. It contains deep thought and a wealth of information that support the five core metrics proposed. After introductory material in the first chapter, this book picks up pace by going into what the authors consider to be the right metrics and why. They follow this discussion with a chapter that shows how they align to a development lifecycle (using the RUP's inception, elaboration, construction and transition phases as a framework). This is followed by two chapters that address the five metric areas, time, effort, quality, workload and productivity, and sizing. Chapters 7 and 8 address productivity and reliability as they relate to the metrics.

I liked the material in the final chapters the most because it takes the concepts in the first eight chapters and applies them to problem spaces such as project control, requirements management, trade-off analysis, and how to use estimates to formulate accurate bids. This material is practical and reflects the real world. Among my favorite chapters are 15 (Replan Projects in Trouble), 17 (Evaluate Bids on the Facts), and 21 (Metrics Backstop Negotiation). However, each chapter in between was also on the mark and credible.

If you are immersed in an unmanageable morass of metrics and want to manage to a smaller set of key indicators in projects or maintenance this book is an essential resource. If you are using Ad Hoc metrics or none at all, this material is an ideal starting point.