Managing the Design Factory [Hardcover]

I just can't say enough about this book; some other specific books on Agile software development are helpful to give you ideas of specific things to do, but this book is absolutely crucial to learn and use in your daily decision-making process.

For example, if we were to view the investment in design work as a depreciating asset, like work-in-process inventory in the factory, we would be able to make better decisions about time, manpower, and project delay tradeoffs.

Key concepts include: valuing design work based on its financial impact on the organization; learning as much as possible as early as possible in the development cycle; managing queues in the development process; creating specifications which are flexible for as long as possible, so that evolving customer requirements can be accommodated.

He clearly shows that we can optimize development work on only one of the following parameters: Product cost, product performance, speed of development, development expense. The approach for each one is different, and it is important to be clear which one is primary.

There is a wealth of useful and practical advice in this book. For example, here are some comments on testing:

"Too often testing is viewed as a necessary evil in the development process. It only exists because we make mistakes. If we made fewer mistakes, we would not need to do all this testing. We should spend our money on 'designing in quality' instead of finding defects by testing. The result of such an attitude may be a test department that is under-resourced and under-managed. Unfortunately, by viewing testing as a problem, rather than an asset, we miss the opportunity to capitalize on the extraordinary improvements that can take place in product testing.
"Let us start by putting testing in perspective. The elapsed schedule time for product testing is typically 30 to 60 percent of overall development cycle length. This is not another minor activity, it is a major design activity. ... text results have inherently high information content. In fact, testing is usually the stage of design process that generates the greatest amount of information.... ...Most companies misunderstand the role of testing ... because they fail to distinguish between design testing and manufacturing testing. ... Manufacturing testing is done to identify defects in the manufacturing process. ... Design testing is done to generate information about the design. A good outcome is high information generation early in the design process. ... We want a failure rate close to 50 percent...." [pp 230-232]

I highly recommend this book to senior managers in product development, and their Marketing and Finance counterparts.

Reviewed by John Levy,
...

Unlike many management books, it's not 20 pages of information stretched out to 200 pages in order to make a book. Also, unlike most product development books, this book is of great value not just to product managers and designers, but would be a great read for financial managers and marketing managers. A manufacturing manager reading this book will smile with satisfaction at seeing common modern manufacturing principles well applied to the design realm.

The only weak points I can think of are: 1) That it may be useful for the author to break out case studies rather than keeping them in the same typeface intermingled with the rest of the text. 2) No real advice is given on how to overcome real-world resistance to these ideas. Some sage advice on how to introduce these concepts and tools into organizations with existing biases and cultures could be a real benefit to practitioners. These are minor objections though.

Whether you're in a software start-up or part of a Fortune 500 company design team doing existing product improvement, this book contains useful information that will enhance your understanding of what you're doing right and what you could do better - and WHY!