While few would dispute that Jack Welch, Lou Gerstner, Narayana Murthy or indeed Wikipedia founder Jimmy Wales each has considerable expertise in organisation design, most would find it hard to pinpoint exactly what such expertise consists of. Is it experience in managing an organisation? A sensitivity to politics and “people matters”? A good understanding of how people behave and think?
Very likely, all of the above. But it’s good to begin with the basics. A graduate student, Eucman Lee, and I set out to find out exactly how experts think differently from non-experts in terms of the most basic decisions about organisation design- namely about how the goals of an organisation are translated into a division of labour, and what provisions are made for the integration of efforts. We also wanted to understand how expertise could be created.
Cognitive psychologists have developed a standard way to understand the nature of expertise in a domain – take a problem and have an expert and a novice each solve it independently in controlled lab conditions while “thinking out aloud”. Basically, the problem solver keeps up a steady stream of commentary reflecting their internal thought processes as best as they can articulate them, while working on the problem. Based on this technique, researchers have learned exactly what experts do differently from non-experts when they play chess, do math, drive taxis, fight fires and formulate business strategies.
So that’s what we did too. We got 16 Master’s level students (not MBA’s) with an average of two years of experience with no exposure to organisation design (“novices”) and 16 experts with an average of 14 years of experience who had significant exposure to solving organisation design problems in their careers, and had them individually solve the same design problems. They each had to propose two designs, one for an entrepreneurial start-up and one for a re-organisation in an established company, based on short case descriptions of real life scenarios. They “thought aloud” for our benefit, which we duly recorded and coded.
What did we find? Of course, the experts more often get it “right”, as would be judged by other experts of organisation design, but that’s not the surprising bit. More interesting is how they think differently: We found that experts rely to a greater degree on analogy to previously encountered problems. Experience and exposure to similar problems pays off. They are also more visual in how they represent the problem. We aren’t talking about just box and arrow charts (though those are important too), but also process and network charts. They spend more time than novices considering the goals and the boundaries of the organisation they are designing- they want to understand the scope of the strategy as well as what activities must be done in-house, before they propose design solutions.
Most importantly, experts are more balanced in the attention they devote to how to divide up work and how to put it back together. Novices tend to be more skewed in their thinking: they focus a lot more energy on the dividing up of work, but a lot less on the mechanisms for integrating and coordinating work. This tendency to focus more on partition than on re-integration when designing organisations is known as “partition focus”. Experts appear to be less prone to this problem. Finally, we also found that the differences between experts and novices are largest when they solve re-organisation problems, rather than greenfield design problems. This because there are fewer constraints on design in the latter case.
The implications? Expertise will pay off particularly when you are confronting reorganisations rather than greenfield design problems. Expertise requires exposure to a large number of similar problems. Therefore, it seems very likely that experienced specialists in organisation design are indeed likely to know things non-specialists won’t. Experience is an important basis for expertise in organization design.
But expertise can also be improved simply by avoiding certain common errors novices make such as concentrating too much on how to break down and assign responsibilities (the boxes in organisation charts), but not enough on how to integrate work (within and across these boxes); or paying too little attention to the goals and boundaries of the organisation being designed. Put simply, both experience and training can help build expertise.
For my next column, I want to tackle a question that I often find myself being asked by senior business leaders: “how much of my organisation should I design?” Most large organisations have multiple layers of management, and how realistic is it for the leaders of the organisation to be able to design all these layers? Take the short (2 + 4 background questions) and anonymous survey “How deep do you design” to help us find out. I will summarise and discuss the results in my next column. Click here to take part in the survey</.p>