Book review: "Good strategy bad strategy"

Dick Rumelt was my prof in Strategy when I did my MBA at INSEAD. I was quite impressed by him, but always wondered why he did not write a book. Now he has written that book. It is a good, but only for a niche audience.

Most books on strategy or management are written for novices. Harvard Business Review Press is probably the best example of formulaic publishing. They take an author with some good ideas and turn the ideas into a rather dumbed-down book. I do not know their formula, but it goes something like this: use around five companies as examples, avoid numbers in the text, use figures sparingly, use a fairly big font size, use wide margins and 1.2 line spacing, never use any footnotes, use decent quality binding and paper, publish a teaser of the book in Harvard Business Review. I have often wondered who actually reads a book that could have been condensed into 25-30 pages or the same number of slides.

Had Rumelt sent a draft to HBR Press, it must have been rejected. The book is rather poorly structured with kind of disconnected chapters, two chapters deal with a case study from the 1980s, and the book does not have a clear take-away. However, it is full of interesting ideas. You decide which kind of book you want to read. Some ideas are big and some are small. Many of them are only in embryonic form. Here are a few:

• Attractor states. Take any industry and do a thought experiment what would be the globally rational structure. I think Rumelt has something like maximisation of long-term consumer surplus in mind. The resulting industry structure is an attractor state. There is no certainly that we will reach that state, but we might very well end up somewhere close.
• Virtual advisory panel. It is good to test your ideas with smart, but critical people. If you tell your mother about a personal problem, you probably know what she will say before she opens her mouth. Use the same principle for your business decisions. Create a virtual panel maybe consisting of a university professor you liked, your first boss that actually hired you, somebody you admire in the industry, somebody with specific domain knowledge. If you do not know the person on your virtual panel, read some biographies or articles. I found this a fascinating idea. I am not sure it will work, but I loved reading about it.
• Very predictable forecast errors. Rumelt does not mention Soros's reflexivity theory but that is what he is describing when companies take investment decisions based on stock market performance. HTC share price increased around 400% in 2010. I was probably not rational to set the investment budget accordingly. Another predictable error for durable goods is how faster sales today will lead to lower sales tomorrow. Blackberry's new operating system generated sales in early 2013, but once loyal customers had upgraded to a new handset, there was very little additional demand.
• Strategy is not decision making. It is more like design. Crafting a tightly coupled value chain for instance (e.g. Ryanair).

This is certainly not a book written for novices. Not that the book is really difficult, but it assumes the reader has knowledge of the field already. It also assumes that the reader is really curious about the topic. There are no quick fixes or directly implementable models in this book. It is a book that requires work to be useful. We need more books by intelligent people that have spent decades thinking about strategic management. Five stars.