“Businesspeople” – be they managers, executives, entrepreneurs or investors – are among the most voracious creators and consumers of “theory” on the planet. This is because businesspeople thrive or perish by the quality of their decisions, so any theory that improves those decisions is precious indeed. Yet there’s a problem. While certain disciplines routinely teach the rigors of theory building, it’s all too often absent in business education and practice. This will not do.
While there are certainly exceptions, for the most part businesspeople have little ability to objectively evaluate a theory’s quality beyond resonance with their “gut” intuition or limited personal experience. The result is rampant malpractice in management theory, with profound consequences for businesses and the communities that depend on them.
Imagine a world where, rather than approaching medicine through disciplined research, testing and statistical validation across large samples of patients and caregivers, medicine was practiced according to each doctor’s individual gut intuition or experience. As unthinkable as that may be in modern medicine it’s largely the norm in business, where the impact is every bit as dire. While businesspeople aren’t generally thought of as the custodians of community wellness, the unfortunate fact is… they are.
No army of doctors or swarm of humanitarians has consistently matched the lasting ability of business to transform a population’s wellbeing (in good ways and bad). Anyone who doubts this need only compare Detroit during and after its automotive heyday, or the average American family a year before and after 2008. The science of business is, without exaggeration, among the most critical, far reaching and grave of humanity’s domains. For better or worse, this is a reality.
Going back to business decision making, there’s no shortage of literature and commentary about theory building and the Scientific Method going all the way back to Aristotle, Galileo, Kepler, Kant and Popper. While this article is hardly a substitute for deeper understanding, it offers an introductory shortcut for those who may at least want a cursory look at theory building. Consider this a quick and dirty introduction, not a treatise.
Six Steps of Theory Building
Any framework, model, heuristic, process or other tool for improving decisions (I’ll use the word ‘theory’) can be seen as stationed along a continuum. Borrowing from Christensen and Carlile[i], the picture below illustrates this continuum in six steps:
Let’s imagine an example where you have to decide whether or not to merge two companies you’ve just acquired. Step one is observation; in deciding what to do you might gather lots of data about mergers and study them. Step two is to organize the data into categories based on attributes. Mergers that worked may be sorted from ones that didn’t. Large mergers may be compared with small mergers. Mergers in your industry may be compared with mergers in other industries, and so forth. Step three is to see if there are any statistically significant correlations between your categorization schemes and the observed results. You may find small mergers to be statistically more likely to succeed than large mergers, or perhaps the opposite.
Steps one through three are descriptive, where the goal is to accurately describe your data and understand what’s happened in hindsight. This is not to be confused with what’s prescriptive, where the goal is to predict the future. Steps four through six are where the prescriptive begins to emerge.
In other words, while finding a correlation can take lots of work, you’re not done yet. Before you run off, make your decision and start penning a bestseller, there are at least three more steps. Step four is to create hypotheses of causation based on your correlations. A correlation between two things doesn’t mean one necessarily caused the other, or even that the two things are connected in a helpful way. For example, there may be a correlation between eating nachos and brain cancer (I made that up), but does that mean nachos cause brain cancer, that brain cancer causes you to eat more nachos, or is the correlation a mere coincidence? Is the correlation “spurious”? You need to find out what causes what.
To help tease out causation, step 5 is to categorize your correlated hypotheses based on circumstances. If nachos cause brain cancer, you’d expect to find more brain cancer in circumstances where people eat more nachos and less brain cancer in circumstances where nobody eats nachos. Step 6 retests such circumstance-based causal hypotheses for statistical significance across old and new data sets. If your analysis still holds, you might have a useful theory. If it doesn’t, you need to improve it or toss it aside.
Nobody said theory building was easy, but that’s why good theories can be game changing and confer competitive advantages. Beyond providing an outline for theory building, another key lesson here is that businesspeople have a bad habit of skipping steps. The most common and flagrant violation is to stop at step 2 and prematurely treat a theory as “prescriptive.” For example businesspeople like to look at data, organize it based on some attributes, and then proclaim that those attributes can be used to make better decisions in the future. The barely descriptive gets mistaken for prescriptive.
Anyone who doubts this proclivity need only read Good to Great by Jim Collins[ii]. Good to Great is an all-time business bestseller that (in summary) examined 1,435 firms, decided 11 of them were “great,” and found that those 11 firms had humble managers. From this Good to Great asserted a prescriptive theory that in order to be great, firms must have humble managers. Collins stopped at step 2, declared victory and sold millions of copies. Yet this isn’t a victimless crime. Beyond being a grossly incomplete basis for decision making, anyone who’s been fired, undervalued or passed over for not being “humble enough” by the standards of Good to Great has felt the needless damage flimsy theories can have on businesses, individuals, their families and the communities that depend on them.
The goals of this article are to (1) provoke businesspeople to become more conscious about how they use theory, and to (2) advocate a duty of disciplined quality control (both as theory creators and consumers).
This is not to say that all ideas have to reach step 6 before being given a second thought. Even the best theories begin at step one, and that’s nothing to be ashamed of. We must simply be honest about how far our theories have progressed, how far they haven’t, and about what remains to be done. In medicine it matters a great deal whether a drug has been robustly researched, or quickly cooked up using a grab-bag of guesswork compounds on a dirty spoon. Both may be promising at some point in the future, but only one should be entrusted for patient care today. Bad business theories destroy lives. As such, we businesspeople must take personal responsibility for the theories we employ and strike out against rampant malpractice.
[i] Christensen & Carlile, The Cycles of Theory Building in Management Research, (2004); see also Christensen & Raynor, Why hard-nosed executives should care about management theory, Harvard Business Review On Point (September 1, 2003).
[ii] Collins, Good to Great: Why Some Companies Make the Leap… and Others Don’t, HarperCollins Publishers Inc., (2001).