6/5/12

The Problem with Perfection

If you're not familiar with the law of diminishing returns, it states that at a certain point adding more effort will not produce significantly more gains. The challenge is knowing when you've reached that point. For many managers this is an important question: How far do I keep going on a project before I declare that it's "good enough" — and that further effort will not significantly change the outcome?

Several years ago I worked with a project team charged with increasing sales to its large corporate customers. At the first meeting the team brainstormed ways to drive up sales, but before moving ahead decided to collect data about current sales and survey sales managers and customers. Since it wasn't clear which ideas might work, this seemed like a logical next step — until the data analysis work dragged on for months as the team tried to reach the perfect answer.

I've seen this pattern in many organizations where, instead of moving into action, managers insist on doing more analysis. In some cases this is part of a company-wide "paralysis by analysis" culture, while in others it is a personal tendency of the manager or team involved. Either way this oft-repeated pattern results not only in wasted effort, but significant delays in moving forward.

From my experience, there are two often-unconscious reasons for this unproductive quest for perfection. The first is the fear of failing. In many organizations, coming up with a recommendation that doesn't ultimately succeed can be career limiting. So to avoid this fate, managers put in extra effort to get the "right" answer, and back it up with as much data and justification as possible. Then, if it doesn't work, nobody can say that they didn't do their homework.

The second driver of unproductive perfection is the anxiety about taking action. Studying problems and coming up with recommendations is safe territory; while changing processes, procedures, incentives, systems, or anything else is much higher risk. Action forces managers and teams out of their comfort zones, driving them to sell ideas, deal with resistance, orchestrate work plans, and potentially disrupt work processes for colleagues and even customers. So one way to avoid dealing with these messy issues is to keep the study going as long as possible, thus delaying any action.

Because of these psychological dynamics, breaking free of unproductive perfection is not easy. But if you are a project sponsor, leader, or team member, and want to move into action more quickly, here's an approach you can try: Instead of viewing "action" as something that follows research, think about how action can occur parallel to research. In other words, rather than coming up with perfect recommendations and then flipping the switch months later, start by testing some of your initial ideas on a small scale immediately — while collecting more data. Then you can feed the lessons from these experiments into the research process, while continuing to implement and scale additional ideas.

For example, in the sales case described above, the team shifted its patterns by selecting three corporate customers where they could quickly test some of their ideas, in a low-risk way, in collaboration with the sales teams. With one customer, the sales leader experimented with selling products and services together, rather than having services as an after-sell. A second sales leader added a paid advisory service to his offering. The third worked on building relationships higher up in the C-suite. The lessons from these experiments were then incorporated into the team's recommendations, which were then tested with several more customers and so on. Within a year, most of the corporate sales teams were working differently and increasing their overall sales.

Clearly the ideas that first emerge through this iterative approach are not going to be perfect, but by sharpening them through field-testing rather than theoretical analysis they will eventually become good enough to deliver results. Working in this way also reduces the risk of recommending the "wrong" ideas and the anxiety about managing change, since small-scale tests provide rapid feedback and engage others in the organization right from the beginning.

Perfection certainly makes sense when designing an airplane or an office building. But if the search for perfection is leading you to diminishing returns and an avoidance of action, it might be worth taking a different path.

Does your organization have a problem with perfection?

http://blogs.hbr.org/ashkenas/2011/08/the-problem-with-perfection.html

No comments: