Good to Great, or Just Good?
Perhaps it shows a bit of an attitude problem that I take joy when certain business assumptions are disproven. I find it especially comical when business leaders blindly follow the practices of a particular author because it’s the popular thing to do, or continue to misuse the same idioms over and over again to explain their faith in a particular methodology.
The book Good to Great being a bestseller, and having been so widely circulated doesn’t mean a whole lot to me, but I’ve seen business leaders that treat the thing like it’s a business bible. So I revel somewhat in reading the article Good to Great, or Just Good? This article doesn’t disprove the findings of Good to Great, but it does create a severe credibility problem for the author.
The article centers around two excellent arguments relating to the research methods used by GTG author Jim Collins. Collins is no dummy but after reviewing the article, you can see how the research methods seriously stretch credibility.
The first error Collins’ team made centered around their use of data mining. Data mining is the process of searching collected data for patterns and formulating explanations for those patterns. These two steps were actually followed by Collin’s team. After this has been done, extensive testing must be performed to test veracity of the patterns formed from the data. This is where Collins’ team blew it. They created a series of patterns from the data that they gathered but they didn’t do the extensive testing necessary to prove that those patterns had any real empirical value.
Collins’ whole point was to try to gather evidence without “starting with a set of assumptions”. We call those assumptions hypotheses in the scientific world. Have a predefined set of ideas to prove or disprove would have actually worked in Collins’ favor. The second fatal flaw was assuming causation where none existed. Collins started with what he felt were great companies and found 5 commonalities. He reasoned that if a company possessed those 5 traits they would be great. If he had started with the traits and then found companies that possessed them his argument would have had more meaning.
To put it bluntly, the research methods used by Collins’ team were completely backwards. While it is a neat idea to go in without any pre-conceived notions, it’s obvious you can find commonalities in anything without a set of guiding principle at the outset. Having a stated set of hypotheses to prove or disprove is a tried and true research method. I fail to see a point in trying to reinvent that process if you’re doing research. Although the book sold like crazy, it has the business credibility of “Mad Magazine” in my mind. Spotty research disguised as empirical data creates a book good for little more than bathroom reading. The article on the other hand, was great.
Evidence Based Management
The second reading was a great article with the above name. It compared evidence based medicine with management styles. A lot of discussion was given to using evidence in decision-making rather than using some of the older management methods for doing so.
One of the main reasons given for the lack of use of evidence in decision-making is the feeling of managers that their experience trumps information collected in a study. In some situations this probably makes good sense, but evidence based management has a huge following because it’s about doing what makes the most sense for the company, not the manager.
Probably the best example of the problem faced by companies that don’t practice evidence based medicine is their lack of ability to fight through bad ideas and the “that’s how we do things here” syndrome prevalent in many organizations. The best example cited in the article, was the use of forced rankings in performance reviews. While it works well for a company like GE, it can be and is a huge problem for other companies. The evidence suggests that this form of performance appraisal is hugely unpopular and creates distrust between employer and employee. There are a host of other reasons not to manage in this manner, but many business leaders will cite the success of GE. The problem is, there is no contravening evidence of what GE’s performance would look like without this practice. There are hundreds of examples of other companies that have suffered from this practice but GE touts it as a critical success factor.
Overall, the idea of the article was to use evidence to make decisions. I can’t imagine how anyone could argue logically against the idea. But it appears many companies and many managers are still managing by the “gut feeling” or “that’s how I’ve always done it” approach. While these practices may not always cause problems, the won’t ever be evidence based. Evidence based management, just like evidence based medicine is necessary to keep us honest and moving forward. As the article states, if doctors practiced medicine the way many companies practice management, there would be more unnecessarily sick or dead patients and many more doctors in jail.