Whenever we create performance measures, we do so with the intention of monitoring results so that we can learn from what’s happened, and hopefully improve in the future. But measurement is more than numbers. Any time you monitor something it will impact behavior, and it’s important to consider those elements when creating your metrics. Here’s an example from an information technology (IT) team we worked with recently.
This group was in the process of developing measures throughout their organization, taking that all-important step of using the power of linked performance metrics to generate alignment from top to bottom. At one point during the workshop, their team lead asked a question that had been bothering him.
IT Team Lead: “We’ve been told that minimizing expenses is crucial to the organization, and so we’ve created a measure of reducing vendor costs. What we’re going to do is negotiate with software and hardware vendors, and consultants to try and drive down our overall IT costs.”
Us: “And if you do that, what will happen?”
IT Team Lead: “Well, we’re concerned that if we insist on lower costs from our vendors that could lead them to cut some corners, and ultimately result in poorer service to our customers here in the company … and that’s the last thing we want to happen.”
It was clear from the look on his face this was a dedicated professional who wanted to do the right thing for the organization, but was concerned that his measures could actually harm his goals by creating some unintended consequences. So, we recommended a counterbalanced measure to mitigate the risk.
The IT lead knew that reducing costs was important to the bottom line but didn’t want those lower costs translating into poorer service for his customers. Therefore, he chose a measure of “customer satisfaction with IT services” to counterbalance vendor costs. Over time he’ll monitor the two, looking for correlations that may require his intervention. If, for example, vendor costs do decrease but he also sees a decline in customer satisfaction, he can hypothesize the two are correlated and use this information to possibly reconsider targets for vendor cost reduction. Maybe the initial target was too aggressive, leading to a degradation of the services provided to his customers.
Consider these ideas as you create your own measures.
- Review what you have now
Carefully review each measure you currently track and critically examine what behaviors they may drive within (and outside) your organization. - Add counterbalanced measures where applicable
For any measures you feel hold the potential of driving the wrong behavior, work to create a counterbalanced measure you can monitor to mitigate any negative effects. - Take action
In the example above, the IT manager decided that monitoring customer satisfaction could counterbalance lower vendor costs. Once you see a negative reaction occurring, in this case if costs are going down (that’s good) but so is satisfaction (that’s bad), you need to intervene immediately. How severe is the dip in satisfaction, and what is more important — lowering your costs or keeping customers happy? That’s an important strategic discussion to have, and it can only be held once you have the data to support it.
An old saying reminds us, “You get what you measure.” Therefore, it’s critical to think carefully about the potential ramifications of the measures you’re using to run your business today, ensuring you’re balancing any possible negative outcomes with measures that provide an early warning system for analysis and remediation.
