I think anyone who solves serious problems by analyzing data will want to own a copy of this book. Being able to organize data into the right visual image can often make no less a difference than that between seeing the answer to the problem vs. getting lost in the complexity and variation in the data.
This is a uniquely comprehensive encyclopedia of graphical techniques with just enough detail on each technique to help you choose the right one for each situation.
There are no long, detailed explanations of principles. What you get are a few illustrations of each type of graph, with a general description of the strengths of that particular technique and several variations to show how it could be applied to different situations which share some central similarity.
One review criticized the alphabetic listing of the techniques, which is a reasonable critique in general. However I think the weakness is mitigated significantly by the way the graphs are grouped together into broad categories once you get to those. The alphabetically listed individual headings are mainly for cross-reference. It seems clear to me that the book wasn't intended to be read from front to back alphabetically, but that the reader would have a rough idea what sort of graph they needed, would start with the heading for that category, and then when neccessary, would refer to the cross-referenced section alphabetically.
In any case, I found it useful to place sticker-tabs on the pages for the main categories of graph that I care most about, and use those tabs as my starting place for choosing the right graphic. There are about ten broad categories of graphs I usually care most about, such as bar, area, column, line, and point graphs, control charts, statistical distribution charts, and time/activity charts. In addition there are about another dozen or so big categories of topics about graphs in general, such as choosing the right aspect ratio, the right font, and the right scale.
Don't get the wrong idea here, none of these topics is covered in great detail, this book is wonderful *index* to visual techniques for showing data for operational purposes but it is not a detailed how-to or an academic treatise on the individual techniques. Also, the book is not intended for creating flashy presentation or marketing graphics, nor does it cover argument maps, truth maps, or any other single sort of conceptual maps in any great detail (although it does touch on the topic in general).
A welcome bonus is that the bibliography is particularly well selected, and not just a list of popular books on graphs. Some of his references are difficult to get and I suspect that some of these sources may even out of print, but some of them like Tukey's work and William Cleveland's texts are well worth searching for.
This is an indispensible encyclopedia of operational information graphics for helping you to help data tell its own story in its clearest and most revealing light, whether you are trying to manage the quality of a process or track down the source of a problem. The examples are extremely well chosen and representative, and the explanations are concise and helpful in a way that lets you use this as a quick reference and not just as a textbook.