Designing Organization Structures - A Question of Principles or Metrics?

Ever been presented with two equally attractive but mutually incompatible choices? Of course you have. It's all part of the human condition. In the demotic lingo of our time, there is no doubt that it sucks to be in this situation. It particularly sucks that it doesn’t restrict itself to one manageable compartment of our lives. We see it all around us. Take organization design (it took a while, but now I'm on topic) for example.

Every book on the topic and every management consultant's process involves an early stage calling for the establishment of some design ‘principles’ or ‘criteria'. To quote Jay Galbraith, a leading figure on the subject, these should be ‘concrete statements about how the organization will behave.’ Having been involved in organization designs many times over the years, I can tell you that there are typically two frustratingly non-complementary outcomes from this exercise:

1. A sense of exhilaration at the level of ambition and expectation being placed on the result of the soon-to-be-designed structure ("Cool, this is going to sort out everything, once and for all."). The problem with this is that the expectations are quickly dashed by the second outcome….
2. A sense of hopelessness, given that the list of design principles is long and, you guessed it, has many incompatible expectations ("Cripes, are you kidding me? How do we achieve all of this?"). This is problematic. The statements are, all too often, so generic (i.e. anything but concrete) that, coupled with the length of the list, post-design practically any action could be rationalized as having met some element of the criteria - without necessarily doing anything useful.

A friend who is a senior executive in what might be loosely called the technology industry called me recently to chew over some design issues. He described his list of principles. In essence, they were: position for growth, cut costs, be flexible and able to respond to changing circumstances. Yes, and I want to be able to work a short week without any pecuniary disadvantage, eat and drink all I want, and be fit and healthy all at the same time. Sorry, but if it's that important, then choices have to be made so that you are not stuck with aforesaid equally attractive but incompatible goals. Is cutting costs the priority or gearing up for growth? Can you really afford to build in organizational redundancy to achieve the flexibility you are seeking?

I hear you mumbling something along the lines of 'We can all pick holes. Have you got a better idea?' Well, let me try one on you. Maybe we need to start design from a different perspective, borrowed from all those shareholder value enthusiasts. Ask yourself, which metrics have to show improvement, and by how much, in order to justify the disruption of re-organizing. Then take those metrics and figure out what changes would be needed in the organization structure to have an impact on them. You might find that it's not a structural change that's needed. By focusing on the metrics you may well find that the most appropriate response should be a small specific change, rather than risking a big upheaval through restructuring. It might be just a question of making talent choices: who needs to sit where and what type of people need to be hired. It might tell you that deploying a new set of goals is sufficient, or different mechanisms are required for integrating activities across the company.

But even if you do decide that a restructuring is required, you’ll have a much clearer conception of what needs to change. For example, if the metrics show that you need a higher rate of product development then you will work on the innovation centers of the company, if you need more sales, you might need to work on both product development and the sales force organizations. The more specific you get, the more likely any restructuring will be both effective and efficient.

In my friend's case, a merger was driving the need for redesign. Obviously, you cannot avoid a re-organization in this situation, but the metrics based approach still applies. You still need to figure out the specific changes that will lead to performance improvements and justify the merger.

Numbers almost inevitably help you to be more specific and be set in reality, not theory. The more specific you are, the more realistic the list will be (less exhilaration maybe, but less of a downer as well). The more realistic, the more likely you are to achieve the outcomes for which you were hoping. Less effort wasted on generalizations, plus measurably improved outcomes are things that might even bring a smile to the faces of our shareholder value friends.