As my wife is fond of commenting, "change is seldom for the better." A new chef at the highly rated San Francisco restaurant Ducca decided to make his mark by adding chili to some pasta dishes and grapes to others -- with bad results. Mozilla's latest browser version seems to crash more than their last. There is a company that sells software so that Microsoft Office 2007 works like 2003, for those that preferred the former version. A former senior official at an important federal government agency told me that each time a new head took over, that person undid what the previous person was doing and began something different -- regardless of what was working.
The tendency for new leaders to want to put their "stamp" on the organization and its products is a natural result of the desire to self-enhance -- to want to feel good about ourselves by showing what we can do. Continuing what has been successful in the past doesn't feed our egos nearly as much as making a change that will be identified with us.
The problem is that all too frequently, change for change's sake is harmful or worse for organizational performance. You need to know the data: although there is a lot of emphasis on the benefits of change and innovation in much of the popular press, the evidence shows that change is most often bad for all concerned. First of all, most new ideas -- like most new products -- aren't very good, which is why there's such a high failure rate for innovations in virtually every industry. Second, change, even change to a better way of doing things, is inevitably disruptive to existing routines and demanding of new competencies and skills. While a company is making the transition, things can go wrong and costs increase, literally threatening the survival of the organization.
A study of 150 entrepreneurial start-ups in the Silicon Valley by organizational behavior professors Jim Baron and Mike Hannan illustrates this phenomenon nicely. Baron and Hannan found that companies founded with a commitment model for managing their people were more likely to reach an initial public offering or be acquired at a good price and were much less likely to fail than companies managing their people using other approaches. But -- and this is what's important -- companies that shifted to this more effective way of managing people were actually more likely to fail than companies than had begun with a different, less effective way of managing but stayed with it. That's because the benefits of changing to a better way of doing things did not outweigh the disruptive consequences of the transition. There is a great deal of evidence in this intellectual tradition -- the population ecology of organizations -- showing the same thing: change, in leaders, in strategy, or in organizing models, almost always leads to a higher risk of failure.
Change for its own sake also causes cynicism and resistance on the part of the rank and file. Since employees know that management approaches come and go as leaders transition in and out, they don't take the new initiatives very seriously. At a large bank Bob Sutton and I studied, branch managers and other executives knew that new initiatives would pass as leaders moved to new positions -- so they never bothered to implement anything. Why should they, when the next person to take over would just undo it and try something different? Consequently, good ideas could not get traction and the work force was largely disengaged, watching the comings and goings and the initiatives of those in charge with bemusement.
Smart leaders understand these dynamics. They focus on changing only what needs to be changed because it isn't working -- the recipes that aren't up to snuff or the product features that bother customers -- and they keep what works, even if it's a legacy from the past. Second, they understand the costs and risks of change and losing focus, so they don't overburden the company by trying to do too many new things at once. Every business has a few core elements that make it successful, and the shrewd leader focuses on the minimum amount of change needed to improve those things, not making a bunch of other disruptions in activities that matter less.
The evidence from numerous studies shows that the aphorism "change or die" is incorrect. It's more likely to be "change and die." The best companies know this and act accordingly. As is often the case, keeping leaders' egos in check is crucial for avoiding the "change stuff to make my mark" problem.