This morning I received the book The Halo Effect by Phil Rosenzweig. It is recommended by Nassim Nicolas Taleb. On the cover it reads "One of the most important management books of all time". Well, you can't blame book publishers for lack of trying to get a book sold. The under title of the book is intriguing. It is called .
..and the eight Other Business Delusions That Deceive Managers. Of course, we don't like to admit how little we know. The social psychologist Eliot Aronson observed that people are not rational beings so much as
rationalizing beings. We want explanations. We want the world around us to make sense. (p. 12).
Managers don't usually care to wade through discussions about data validity and methodology and statistical model and probabilities. We prefer explanations that are definitive and offer clear implications for action. We like stories.( p. 15). Maybe that's why management is (still) not an academic discipline like geology or medicine. There is nothing wrong with stories, but some stories (like the constant flood of business books) are dresses up as science. They take the form of science and claim the authority of science, but lack the rigor and logic of science. They're better described as
pseudo science.(p. 16). The famous
Cargo Cult Science of Richard Feynman is used as illustration. Evidence based decision making is for the manager who does not go for the quick fix. Many manager complain they have so little time to take informed decisions. But there is of course a trade off. Good decisions taken at a slower pace or bad decisions at a rapid pace. This point is also made by professor Rob Briner in his video
using evidence based management in your day job.
The halo effect
The halo effect is a way for the mind to create and maintain a coherent and consistent picture, to reduce cognitive dissonance ( p. 50-51). Why is it so hard to understand why some companies succeed and others fail? In fact, our thinking about business is shaped by a number of delusions, the first of which is the Halo effect. So many things we-managers, journalists, professors and consultants- commonly think
contribute to company performance are often attributions
based on performance. And even if we try to gather data in large-scale examples, like the Fortune survey or the Great Place to work study, we often do little more than multiply the Halo effect. (p. 64). Great examples are given with cases studies on Cisco and ABB. When the S&P ratings are up, journalists, professors and consultants tumble over each other to applaud and offer praise for the vision and steady leadership of the CEO. However, when a year later the stock market collapses, the same leaders are blamed for ruthlessness, ego mania and lack of customer focus. Other delusions include the mix up of correlation and causality, the delusion of single explanation (anecdotal evidence), connecting the winning dots (selective sampling, i.e. only picking winners), lasting success (cookbook recipes), absolute performance (blueprints for success?), wrong end of the stick and finally organizational physics
The famous studies In
Search of Excellence (Peters and Waterman),
Good to Great (Collins and Porras) and
Build to last (Collins and Porras)are critically appraised.It is off course easy to point out flaws, but what lessons can be learned from these mistakes? For one thing, there is no holy grail for high performance organizations (yet). Finding evidence to support management decision making is not at all easy. These examples just illustrates the gap between management and science. Furthermore, putting evidence in context, avoiding delusions and understanding the evidence (if available) is a team effort. Even then, it is still difficult and it is no guarantee for success.
What do I think of this book?
The book is easy to read and the examples (Lego, Cisco, ABB, Microsoft) are well chosen. I agree with the message of the author that managers should be skeptical, if not critical of management fads, business books and self proclaimed guru's who supposedly know the magic formula for lasting success or a high performing organization. On the other hand, it is pretty demanding to expect from a manager to look for "the evidence" if it is hard to find let alone being able to appraise it. As a researcher I have my work cut out for me.
Progress by the current school of thought of evidence based/informed decision making is slow, but steady. That's the good news. One thing I've learned so far, is that evidence does not speak for itself. Contrary to evidence based medicine, random control tests (the golden standard in research) in management situations are difficult if not impossible. Achieving the highest level of evidence in general has to be ruled out. Evidence has to be appraised, which makes it subjective. It is all about the context, interpretation and the quality of the data. Bob Sutton was also musing on this in his posting on intuition vs. data driven decision making: some rough ideas. Our brains are primed on previous experiences. (That's why we don't need all the syllables in a sentence before we interpret what is being conveyed). We humans are pattern seeking animals. This has all kinds of evolutionary benefits (hence intuition, gut feeling, etc.), but also some major drawbacks. To put it mildly, we suck at judging facts. That's the bad news. This raises all kinds of philosophical issues on knowledge, epistemology and ontology, but I don't want to discuss that right now. To illustrate how bad we are at appraising information, you can read books like Fooled by randomess and Black Swan by Nassim Nicolas Taleb or the Halo Effect by Phil Rosenzweig. But, it is easier to watch this clip of a speech by Beau Lotto called Optical illusions show which he recently gave at TED.
