Most of the other reviews are absolutely right: this book seriously lacks any quantitative explanation. No need to look for kind words; this is a serious oversight. And yes, this book does read like a long sales resentation.
While the authors adequatley describe broadly how economists and financial executives solve contingency claims problems (generally using binomial methods, simulation, or partial differential equations), they don't teach any of these methods in any useful way. At best, after reading this book, you will be able to recognize whether or not your organization has any "real options".
Beyond the quantitative short-comings of this book, however, there are some flawed fundamentals about their whole approach: this book treats real options as a new finance panacea for the 1990's, and suggests that the world of finance in 20 years will be a very different place because of these revolutionary ideas. Contingency claims problems are limited to a very specific set of economic phenomena with specific criteria. If the criteria are not present, contingency claims models fall apart. Consider the amount of abuse something as well-known as the black-scholes option pricing equation is subject to when it is applied to "real options valuation": the black-scholes equation is a function of two variables, primarily: time and stock price variance. When you take this equation and try to apply it to, say, the valuation of an option to market patented drug, how do you define variance and time? Time in an option contract is fixed in the contract. Variance is empirically observable from stock prices. Plus, how do we know that the value of drug patents resembles stock prices (log-normal process)? What if it is more like the behavior of a commodity (mean-reverting process)? And where are we going to get the data from anyway? In that case, the black-scholes equation needs to be abandoned and an alternative partial differential equation needs to be developed. But who is going to do that? At what cost? Obviously, at a certain point the benefits derived from exactly modelling your options is eclipsed by the cost and effort involved in doing so. The scariest part, however, happens when you realize that the greater the variance (risk) and the longer the timeframe chosen, the greater the final value of a project or investment. Now the project manager who wants to sell ice to the eskimos has the quntitative methods available to justify such a high risk project. (Just think, the project manager could sell this project to top management as a long-term investment anticipating the melting of the polar ice caps, when the price of ice in Greenland is expected to go through the roof).
This book tries to reach too far, suggesting that phenomena which never should be valued as contingency claims can be valued as such. Real options (or contingency claims) are best treated as a very specialized set of quantitative techniques used to model very specific phenomena which a company may or may not be subject to see "Investment under Uncertainty" by Dixit and Pindyck for an inventory of those phenomena). Push the envelope too far and the paper tears as it does here.