Using a single metric (don’t even consider using a single metric as a predictive tool unless you have multiple observations for analysis) to represent the performance of entire team or organization akin to slicing a ball in half.  The team can move in an infinite number of directions from any point.  The lack of constraint is both a benefit and a curse.   This paradox is an artifact of human nature which foretells the tendency to maximize performance based on the perceived goals and constraints.  A single metric can lead to a scenario that leads to maximization of measure that is perceived to be most important perhaps to the expense of other critical measures.  An example I observed (more than once) was a contract that specifies payment on productivity (output per unit of input) without mechanisms to temper human nature.  In most cases time-to-market and quality where measured but were not involved in payment.  In each case, productivity was maximized at the expense of quality or time-to-market.  These were unintended consequences of poorly constructed contracts; in my opinion neither side of the contractual equation consequently wanted to compromise quality or time-to-market.

 

While developing the a single, ultimate metric might is an admirable goal, the process of constructing this type of metric and ensuring the data from each metric is synchronized will require substantial thought and effort.  Metrics programs that are still in their development period typically cannot afford the time or effort required for developing a single metric (or the loss of organizational capital if they fail). Regardless of where the metrics program is in its development process, I would suggest an approach that develops an index of individual metrics from an existing balanced scorecard is more expeditious.  Creating an index from a set of well know measures and metrics will leverage the basic knowledge and understanding during the development and implementation of the individual measures and metrics.