If you have never been exposed to Eli Goldratt's work, this book probably should not be the first of his books one reads. It might be a good idea to read THE GOAL first and become somewhat familiar with Goldratt's terminology, but it is not essential. This book will hold the most interest for those who either work with, or make decisions based on, cost data. If you are looking for a "how-to" book, this is not for you. I recommend Dettmer's "Goldratt's Theory of Constraints". However, if you need a good primer on what Goldratt's philosophy is, and how it applies to business data and decision making, this is an excellent resource.
The Haystack Syndrome is composed of three parts, each roughly equal in size. Part One "Formalizing the Decision Process" is the most interesting and has the widest appeal since it is an excellent discussion of how cost data can be used and misused, depending on one's perception of what type of business environment one is in. This is where the real meat of the book lies. Part Two "The Architecture of an Information System" I found less interesting, but it does contain some useful information. However, many find it too dry and technical. Part Three "Scheduling" can be skipped without fear of missing anything you did not get in the first two parts. It is a nuts and bolts description of how Goldratt's ideal scheduling system should work. Since at the time of its publication Goldratt's company sold scheduling software, I think it was a thinly disguised sales pitch. Throughout the book, Goldratt does an excellent job of defining his terms, which is good because he quite often uses familiar terms but defines them in different, but logical ways. Pay attention to his definitions!
The book begins with the recognition that most of us have lots of data but never enough information. Goldratt distinguishes information from data and says that different levels and requirements for information exist within any organization. He believes that too many of our information systems are nothing more than data systems since much of what they produce is not actionable.
Even more important than a new information system, though, is the understanding of how to use that system. This requires a new management philosophy. Part One is primarily a discussion and illustration of that new philosophy. Most companies and individuals, he says, operate in a world defined by the traditional "cost world" philosophy which has served industry well for decades. In the "cost world", decisions are usually made based primarily on how they will affect costs. However, Goldratt examines the assumptions underlying the cost world philosophy and demonstrates how these assumptions are largely invalid now. In fact, decisions based on the old philosophy may be counter-productive. The new philosophy Goldratt advocates is one which emphasizes the maximization of throughput which he defines as the "rate at which the company generates money through sales." Adding the two words "through sales" is important because in the traditional cost world, building inventory is often treated the same as sales. In fact, cost effects are delegated to third priority in the throughput world.
The biggest difference between the cost world and the throughput world is in their perceptions of the systems in which we operate. Underlying the cost world perception, for example, is the belief that most organizations operate in a system of many independent, variable functions - a change in any one of which influences the bottom line. Due to this perception, much effort is expended to identify what the influence of nearly every given change will be. The throughput world view, on the other hand, holds to the perception that we operate in a system where many functions have to carry out many tasks in a synchronized manner until the product is delivered. This implies, therefore, that only a small portion of the functions are truly independent and it is only these functions that determine the results of the entire system. Therefore, any effort expended to optimize the performance of the non-independent functions is futile. These few, truly independent functions are called "constraints" (hence the Theory of Constraints). Several problems are presented and it is highly recommended that you take the time to work them. Most people have no idea how immersed in the cost world they are until they see the solutions to the problems.
Goldratt relates in Chapter 5 the sad story of a $9 billion (sales) company, well-known but unnamed, who applied cost world view thinking to its operations in order to determine "how much each part costs us to make". They then proceeded to buy every part that, according to their cost data, could be bought cheaper. This process was repeated over a period of several quarters until, within a couple years, they went into Chapter 11. To the end, they really had no clue because their system was telling them they were doing the right thing. Unfortunately, they asked questions without examining the assumptions underlying their questions. They got exactly what they asked for, good and hard. (I met Dr. Goldratt in Washington, DC in 1997 and had the privilege to spend an hour with him discussing his ideas. I asked him who the unnamed company was. It was International Harvester).
I first read this book in 1991, but I still find myself referring to it now and then. The style of writing in Part One is very readable, but much like Goldratt's speaking style (without the accent): he's an expert practitioner of the Socratic method so he uses lots of questions and thought experiments, along with many capitalized words and exclamation marks for emphasis. If you have little or no previous exposure to the concepts in the book, you may find it helpful to take notes. Goldratt's ideas have made inroads in the last ten years, and he is continuously refining them, but this book is a good place to start if you want to try to understand his "new way of thinking."