Agile Software Requirements: Preface
In the past decade, the movement to lighter-weight and increasingly agile methods has been the most significant change to affect the software enterprise since the advent of the waterfall model in the 1970s. Originated by a variety of thought and practice leaders and proven in real-world, successful experiments, the methods have proven themselves to deliver outstanding benefits on the big four measures of productivity, quality, morale and time to market.
In the past five years, the methods spread virally. Within the larger enterprise, the initiatives usually started out with individual teams adopting some or all of the practices espoused by the various methods, primarily XP, Scrum, Lean, Kanban (later), and various combinations and variants.
However, as the methods spread to the enterprise level, a number of extensions to the basic agile methods were necessary to address the larger process, organizational, application scope, and governance challenges of the larger enterprise.
Not the least of these is the challenge of agile requirements, which is the necessity to scale the basic, lightweight practices of team agile—product backlogs, user stories, and the like—to the needs of the enterprise’s Program and Portfolio levels. For example, agile development practices introduced, adopted, and extended the XP-originated “user story” as the primary currency for expressing application requirements. The just-in-time application of the user story provided a much leaner approach and helped eliminate many waterfall-like practices, such as imposing overly detailed and constraining requirements specifications on development teams.
However, as powerful as this innovative concept is, the user story by itself does not provide an adequate, nor sufficiently lean, construct for reasoning about investment, system-level requirements, and acceptance testing across the larger software enterprises project Team, Program, and Portfolio organizational levels. That is the purpose of this book.
This book describes an agile requirements artifact model, corresponding practices, suggested roles, and an organizational model that provides a quintessentially lean and agile requirements subset for the agile project teams that write and test the code but also scales to the full needs of the largest software enterprise.
Why Write This Book?
In 2000, after about 25 years of managing software development as an entrepreneur and executive, along with my coauthor Don Widrig, I published my first book: Managing Software Requirements: A Unified Approach. In 2003, we updated the book with a second edition: Managing Software Requirements: A Use Case Approach. These are considered to be definitive texts on managing application requirements—a lot of copies were sold, and the books have been translated into five languages. More importantly, many individuals, teams, and companies told me that these works helped them achieve better software outcomes. That was always the goal.
In the following years, I turned my attention to agile development methods. I continue to be more and more impressed with the power of these innovative methods, the quality and productivity results they delivered, and the way in which they reenergized and empowered software teams. Though the methods were developed and proven in small team environments, the challenges of building software at scale is a more fascinating puzzle—part science, part art, part engineering, part organizational psychology. As a result, I became engaged in helping a number of larger enterprises in adopting and adapting these methods in projects affecting hundreds—and then thousands—of software practitioners. Fortunately, with some extensions, the methods did scale to the challenge. Based on these experiences, I published Scaling Software Agility: Best Practices for Large Enterprises in 2007, a book designed to help larger enterprises achieve the benefits of agile development.
Scaling Software Agility took a broad view of software methods and didn’t focus much on software requirements. Even though the management of requirements continued to be a struggle for many agile teams, there were bigger organization and cultural challenges, as well as a number of emerging agile technical practices, that needed to be addressed.
In the past couple of years, the movement to lean thinking in software development captured my interest, in part because I have some background in lean manufacturing from earlier days. Generally, lean provides a comprehensive, deeply principled, rigorous, and mathematical framework for reasoning about product development economics and the increasingly important subset, software development.
So, my thinking, along with that of many others, evolved further. Many of us started to see agile development, especially agile at scale, as a “software instance of lean.” In addition, lean scales beyond the software development labs and provides tools to address changes in other departments such as deployment, IT, distribution, and program and portfolio management. Simply, lean provides a broader framework for organizational change, and it helps us address these larger challenges. I’m a big fan of lean thinking.
At its core, lean focuses on the value stream and provides philosophies, principles, and tools to continually decrease time to market, enhance value delivery, and eliminate waste and delays. As enterprises head down the lean path, it is again beneficial to focus on optimizing the understanding and implementation of software requirements, because they are the unique carriers—or at least the best proxy—for that value stream.
Lean thinking brings us full circle. Once again, it is useful to focus on requirements management practices in our agile—and increasingly lean—software development paradigm. That’s why I wrote this book.
My hope is that the book will help the individual software practitioner, project team, program, and enterprise adopt and adapt agile and lean practices, deliver better solutions to their users and stakeholders, and thereby achieve the personal and business benefits that success engenders. After all, you can never be to too rich or too lean.