How To Measure Anything - by Douglas W. Hubbard

This is a book about measurement.  In it, Mr. Hubbard explains why measurement of anything that can be observed is possible.  Most equate the concept of measurement with observing and recording precise and purely objective metrics.  If we instead think of it in terms of anything that reduces uncertainty about a quantity, measurement of some type can be devised for even those things traditionally considered impossible to measure.

This is a business math book.  While not a page-turner, it is filled with techniques, cases, and examples for using measurement to inform decision-making.  To lay the groundwork, the author describes three samples of extremely ingenious measurement approaches.  First, way back in 200 BC, Eratosthenes measured the circumference of Earth to within an accuracy of 1%.  Enrico Fermi used simple confetti to help estimate the yield of the first atom bomb.  At the age of eleven, Emily Rosa devised an experiment to conclusively debunk claims of "therapeutic touch" and became the youngest author ever to be published in the Journal of the American Medical Association (JAMA).

This book is not just a collection of historical measurement anecdotes, though.  The book gets into a fair amount of statistics and builds towards Hubbard's Applied Information Economics (AIE) model which brings many of the concepts together.