How to Measure Anything: Finding the Value of Intangibles in Business [Hardcover]

Hubbard writes lucidly about thinking, statistics and decision making. His measurement examples are mostly about businesses, but the methods and issues discussed are very general:

-how do you break down a problem into smaller ones (that are more amenable to analysis)
-how critical thinking can pinpoint key elements of a complex issue
-how even small amount of data is valuable if nothing was known before
-it makes sense to try to measure the immeasurable rather than ignore it
-you need to measure the variable that is least known, even if it is the most difficult one to measure
-there is often no need to measure things because extra decimal of accuracy is mostly useless
-measurement cost (direct and indirect) must be considered
-what are you going to do with the newly found data ? There must some reason to do the measurement,
some decision must made, some money invested, some project dropped. If not, why bother ?

One of the key strengths of this book is that it always reminds about measurement errors and confidence intervals:
data without understanding its reliability is of little use.

Hubbard's book is one of the best examples of a management book i have recently read. While the title focuses on measurement, the focus of the book is not measurement for its own sake but as an instrument to reducing uncertainty in a cost effective manner.

Hubbard quickly does away with the falacies that most things are not measurable or that measurement needs to be expensive and introduces very simple techniques for simple measurmeents, which on their own need not cost a lot but provide the user with a significant reduction in uncertainty about decisions.

On top of introducing practical examples of how measurement can be conducted, and some basic statistics behind them (this is not a book on statistics, mind you), Hubbard also devotes attention to valuing the measurements in terms of reducing uncertainty - making decisions on whether certain measures are cost effective or not much more straightforward.

After the first section on how many more things are actually easily measurable than one would assume, Hubbard delves deeper into the how's and when's of measurement in the subsequent sections, which slowly get more technical - yet never stray beyond the understanding of an even mildly moderate reader. And as the author claims, often most value can be derived from some simple calibration exercises for our intuition, and by conceptually understanding how to go about measurement, even if not every reader will be able to replicate the maths behind it.

This book is something every management consultant should read - it presents some pretty powerful arguments one can use in the sales context and it can be used relatively easily in justifying the expense for uncertainty reduction. At the same time a client, educated with the knowledge presented by Hubbard should be relatively well equipped for decidiing which of those proposals actually have a scope for raising the quality of decisions and which are relatively useless (not worth the money asked for them).

Finally, the writing style is fresh and the examples are both well chosen, and often sufficiently intriguing for the reader, so as to keep the book a good read, even if one is well acquainted with the methods discussed.

"I wrote this book to correct a myth that permeates many organizations today: that certain things can't be measured." Douglas Hubbard goes on to note that he has made a career out of measuring the sorts of things many thought were immeasurable. Intangibles, for example, "that appear to be completely intractable to be measured...in a way that is economically justified." Hubbard notes that there are several common misconceptions about intangibles. He offers what he characterizes as a "universal approach," Applied Information Economics (AIE), to measure an intangible, providing with that explanation some "interesting methods for particular problems."

He duly recognizes that only what is most important (tangible or intangible) should be measured; also, that what is currently most important may not retain that importance; and, that information needs change, sometimes significantly and unexpectedly. That said, basic questions must constantly be asked and answered:

1. What are our most important information needs? Why?

2. How best to obtain and then verify that information?

3. What will we then do with that information?

4. How can we then measure (accurately, consistently, and sufficiently) the impact of actions taken based on that information?

To his credit, Hubbard makes every effort to provide information, explanations, and recommendations that are (in his words) as "simple as can be"; nonetheless, some of the material may prove daunting, at least it did to me. I appreciate the inclusion of dozens of real-world examples that illustrate key points. Hubbard also makes effective use of other reader-friendly devices, such as checklists inserted throughout his narrative. In his own words, here is how he organizes his material:

In Section One (Chapters 1-3), he "makes the case that everything is measurable and offers some examples that should inspire readers to attempt measurements even when it seems impossible."

In Section Two (Chapters 4-7), he "begins to get into more specific substance about how to measure things - specifically uncertainty, risk - and the value of information."

In Section Three (Chapters 8-10), he "deals with how to reduce uncertainty by various methods of observation including random sampling and controlled experiments."

And then in Section Four (Chapters 11-14), Hubbard offers "an eclectic collection of interesting measurement solutions and case examples."

Many readers will appreciate having the Appendix (Pages 269-278) which provides both the questions and answers for various calibration tests, including "Calibration Survey for Binary: B" that also includes percentages to indicate degree of confidence that the respondent is correct.

Earlier, I suggested that this is by no means an "easy read." It isn't. Nor will this book respond directly to every executive's immediate needs and objectives. However, it will generously reward those who need assistance with finding and measuring the intangibles in business if they absorb and digest the material with appropriate care. To those about to begin reading this book, Douglas Hubbard' offers this recommendation: Write down those things they believe are immeasurable or, at least they are not sure to how to measure. "After reading this book, my goal is that you are able to identify methods for measuring each and every one of them." I presume to add another recommendation: Highlights key passages and titles of checklists. By doing so, you will be able to facilitate, indeed expedite frequent review of key concepts and insights later.