The consequences of this model were fairly predictable. Pushy individuals did well, often at the expense of equally smart but less assertive colleagues. Long hours were expected; family life was given little attention. Several of the top executives ended up divorced. Business units in high-growth areas attracted talent, but the more established businesses, even if they were profitable, struggled to keep their good people.
Taken together, these problems might not be fatal. Indeed, there are many successful companies with similar management models. Oracle, for example, is single-minded in its efforts to hire highly motivated people and weed out those who can’t cope with a high-pressure environment. As an executive from Oracle commented during the research, the organization is like “the engine of a Ferrari, which revs at very high RPMs, but can burn out at any minute.”
But such “support light” organizations run several risks. One issue is sustainability. The model relies on continuing growth — and a buoyant stock market — to keep everyone motivated. When the market turns down, there is a real risk that the Ferrari engine will burn out, or explode. Second, this model favors the highflyer at the expense of the steady performer who is content to do the same job year after year. Ultimately, a successful company needs both. Without an effective support structure, Enron lost many of these steady performers.
At the organization level, lack of support typically results in business units doing their own thing and often reinventing the wheel or duplicating effort. This sacrifices productivity and adds needlessly to a company’s costs. For example, a few years ago, executives in Ericsson’s central research and development organization discovered no less than five separate development teams in different countries all working on their version of a “screen phone” — a telephone with a small TV screen for Internet access. Steps were quickly taken to bring these teams together and to encourage a more coordinated effort. But the problem also highlighted just how difficult it is to develop the necessary levels of communication and coordination in an R&D organization as large and complex as Ericsson’s.
Support systems are an essential means for large organizations to help individuals and business units perform to their highest potential. But at the same time, such systems can become oppressive if they are too numerous or are forced on individuals from above. The free-market model Enron developed is excellent in many respects, but in a large organization, such a model typically needs to be balanced with certain top-down controls. Here, appropriate balance between the extremes can be found by following some basic principles:
- Put in place enough support systems to help individuals and make sure they know where to go for help. Individuals probably should take responsibility for managing their own careers, but the company can facilitate their efforts through an internal labor-market system that is structured to optimize the placement of people in jobs based on the person’s talents and the needs of the business. It can provide different career tracks for different types of people, and it can make training and development programs available, rather than mandate them.
- Support systems should encourage business units to collaborate on their own. The underlying logic here is that well-intentioned business units will likely collaborate with their peers if they see value in doing so. Corporate management’s role is to put in place systems or forums to facilitate rather than enforce collaboration.
BP’s well-regarded “peer review” program is a case in point. Business units are clustered into peer groups, and the executives responsible for them are told that their rewards will be based in part on the performance of those peer groups. This then encourages the groups to share best practices and collaborate where possible, but it does not mandate any particular behaviors. In a similar vein, Ericsson has instituted a variety of informal cross-unit forums for sharing research plans to avoid the duplication of effort that has occurred in the past.