The phrase, "what gets measured gets done," has become a rallying cry for trainers and evaluators. We use this to justify our work and convince CEOs that they should invest in performance measurement. However, as I've argued in previous posts, the saying is not always true and, in fact, is misleading.

One implication of the phrase is that if you measure something (customer service, productivity, sales, revenue, etc.), people will pay attention to what is being measured and do what they can to improve Survey responses those outcomes. Many examples refute this logic. GM measures quality of every part and every car yet still has recalled 29 million vehicles so far this year. The Veterans Health Administration measures patient waiting time yet still is under congressional scrutiny for wait times that were much too long. Lehman Brothers, once one of the largest investment banks in the U.S., constantly measured the performance of the securities it owned and managed, yet still had to declare bankruptcy in 2008.

In each of those cases, it appears that key stakeholders had the data but did not use the data to make their decisions. It’s as if they were trying to fulfill a compliance requirement without a commitment to improvement. Or they didn’t want to know because that would mean they would have to change something. Measurement alone is not sufficient; it’s the application of those results to decision-making that gets things done.

A variation on "what gets measured gets done," is, “what you measure is what you get.” To me, this saying has a slightly different meaning. This is more about the importance of choosing the right measure for the situation so that you are reinforcing the intended behavior and not something that you don’t want. I once consulted with a state Blue Cross Blue Shield office that proclaimed their commitment to customer service but evaluated customer service reps on the basis of how many calls they handled each hour. Number of calls handled went up; customer service went down.

Jane Bozarth, in her recent column for Learning Solutions Magazine, writes this about choosing the right measure:

Begin with the end in mind: who is the target audience, what do they need to do, how do we measure whether they are the ones accessing the program, and how do we measure their performance?

So: When looking for measures, try to find things that are meaningful, that give you real information to help real people do their jobs and to help organizations perform more efficiently. Beware of easy measures and vanity metrics.

Good advice! The tendency so often is to look for the lost key under the streetlamp because that’s where the light is. Measures are chosen because “we’ve always done it that way” or because “that’s what we know how to measure” or “that’s what everyone else does.” As Bozarth suggests, decide on what behavior you want and then decide on the best way to measure that behavior. In that way, you’re more likely to get the data and results that you need.

However, here too, the phrase has limits. What you measure is not always what you get. Many organizational factors can intervene. Maybe you are measuring the right things in the best way, but managers don’t value those outcomes, or the findings are not communicated to the stakeholders, or intervening events and unintended consequences are not factored into the results. Again, it’s not measurement per se, but what is done with those measures that makes the difference.