This book is absolutely the "classic" in IO. The author amazingly provided very clear and coherent exposition to this huge, but diverse field. But, this book is not for everyone as some reviews below implied. First, consultants or managers may not find this book useful for the real-life application, because this is a (rather serious) "theory" book. It has no detailed cases, no practical managerial implications, and so on. For that purpose, there are many other good books like Schere, or Carlton & Perloff, etc. The virtue of this book lies in the author's capability to make accessible otherwise very complicated and abstruse models, which would be mainly the concern of grad students in econ, not consultants. Therefore, econ students who are interested in theory development in IO would quickly grasp the essence of important IO models with a simple algebra. I don't understand what part of the book some people think is useful as a reference for the consultants or even everyone.
Second, simple algebra in the book does not mean it is accessible to "everyone." At the introductory grad level, game thoery and information economics use only basic algebra, but it is still not easy at all. Of course, this book introduces almost all important topics usually taught in the first-year micro sequence at graduate school, like repeated game (Ch 6), basic mechanism design technique (non-linear pricing in Ch 3), principal-agent problem (Ch 1), and so on. It would be misleading if one says these topics are accessible to everyone because of "minimal use of math." Even chapter 1, the theory of firm, is a very deep discussion, introducing the concept of "incomplete contract."
In my opinion, this book will be ideal for econ grad students, who would want to quickly review relevant chapters before moving into more recent IO literature (of course, now with full rigor).