When a colleague of the nobel award winning physicist Wolfgang Pauli showed him a research paper by one of his students, asking for his opinion, Pauli is said to have given his legendary remark:
Not only is he not right. He’s not even wrong.
What Pauli was saying is that the student’s hypotheses wasn’t falsifiable. There was no way to set up an experiment that would produce an observation that could conflict with the hypotheses’ predictions, thus proving it to be false. The hypotheses could therefore not be meaningfully tested, which meant that it didn’t adhere to the scientific method.
Luckily for Pauli, he was no longer of this world when the Harvard Business Review published an article titled “Tesla’s not as disruptive as you might think“. The article reports on research being done on Tesla by Clayton Christenssen and his colleagues on whether the car maker fits into Christensen’s theory of Disruptive Innovation, one of the most popular ways of thinking about the effects of new technology on business. The article quotes the research excitedly:
If Tesla is following a disruptive innovation strategy, theory predicts that it will continue to see no strong competitive response … However, because it’s a sustaining innovation, theory predicts that competitors will emerge [when they expand their market] …
Wow, a business theory that can be used for prediction, how exciting! So exciting in fact that the book originally introduced it, the Innovator’s Dilemma, was followed by an even more ambitious book, Seeing What’s Next. In it the author and his associates
present a groundbreaking framework for predicting outcomes in the evolution of any industry … Through in-depth case studies of industries from aviation to health care, the authors illustrate the predictive power of innovation theory in action.
There are two problems with all this. The first one is that they’re not even wrong. The theory of disruptive innovation cannot be falsified. Even when someone points out examples from the books where the theory eventually failed, as the New Yorker did quite colourfully a while back, this is always explained away by a company having been wrongly identified as being disruptive or by some minor niggle in the theory which has now been corrected.
The second problem is that business studies don’t lend themselves well to the scientific method to begin with. That’s what makes the “case study” method so good. It embraces the fact that business studies are more like a collection of good advice you get from your mother before flying the nest than a framework of theories that predict the future behaviour of natural phenomena. In business, there’s no way to make enough observations of a particular business in a particular situation in order for any results to be statistically significant. It’s therefore impossible to conduct any meaningful experiments on business theories.
The theory of disruptive innovation isn’t all bad. There are some case studies in there that show how companies that have dominated their industry are eventually destroyed by competitors they didn’t see coming. But the it is not a scientific theory with the power to predict. And its authors should stop pretending otherwise.