Wednesday, November 14, 2012

Using measurements - how to get started


Sometimes the hardest part of doing something new is getting started. Here are a couple of ways to find your way into the process of using outcome measures. You don't have to try them in the order I've listed them - sometimes it helps to approach a problem from several angles.

  • Think hard about what you do, and what you want to know about it. What is the end result of your services? If you provide education or tutoring, what sort of improvement do you expect to see? What does that improvement mean in the long run for the students you are serving?
  • Why are you looking at numbers in general, and this process in particular? For example, if an important function is backlogged, you can use numbers to tell that the backlog has been cleared. But then you can take a deeper look, using what you've learned to identify the root cause of the problem. The goal is to prevent it from recurring once you have cleaned it up.
  • What data systems do you have? How can you harness them to provide numbers? Bring in your IT and QI staff. But don't use a number just because you can measure it. 
  • Make sure your measure tells you what you think it does - don't measure something just because it's convenient. One example - it's very easy to measure the number of people who start a program. Maybe comparing that number to the number who complete the program will tell you something you need to know. But if it doesn't don't use it!
  •  Don't rely on a single measurement - you are likely to miss nuance. At the same time, don't try to measure too many things. And, as always with numbers, remember the context.

Here's a good description of an effective use of numbers, from McKinsey Quarterly's Report, "The Global Gender Agenda."
McKinsey’s more general work on transforming the performance of companies shows that those with a clear understanding of their starting point are more than twice as likely to succeed as those that are less well prepared. In a gender diversity context, this understanding means knowing the gender balance at every level of the organization; comprehending the numbers by level, function, business unit, and region; and then monitoring metrics such as pay levels, attrition rates, reasons women drop out, and the ratio between women promoted and women eligible for promotion.
Why go to this expense? Establishing the facts is the first step toward awareness, understanding, and dedication to improvement. Using a diagnostic tool, one company simulated how much hiring, promoting, and retaining of women it would require to increase the number of senior women managers. That approach helped it set an achievable and, just as important, sustainable target that would not compromise a highly meritocratic corporate culture. With an overall target—that 25 percent of managing directors and directors should be women by 2018—and a clear understanding that the bar for promotion could not be lowered, managers now look harder for high-potential women and start working with them earlier to develop that potential.
You can see my earlier post about that report here.

This is another in a series of occasional posts about developing and using outcome measures. You can see a previous post, with links to related posts, here.

No comments:

Popular Posts