Improve Your Performance Measures With Logic Models

Mid Century Modern by Honey Cup

Mid Century Modern by Honey Cup

After an insightful conversation with performance measures guru, Gary Better, I’m sold on logic models as a method to reorienting your measures from crappy, boring, and onerous to awesome—if such a change appeals to you. Of course, as any good guru will do, Gary would tell you that logic models are part of a more holistic approach to setting up a performance management program. Indeed. However, the logic model is a simple, straightforward way figure out what makes sense to add given your program’s maturity and available data.

Once your inventory all of your measures and bucket them according to your program's objectives, you’re ready to tackle the logic model.

Why use logic models? For me, the number one reason is that they get you (and by extension your management team) focused on what is strategically important (outcomes) rather than simply what is easily measured (data, process, and output). Every program and organization needs a blend of each of these four categories.  

The problem is that we tend to load up on measures at the front end of this process at the expense of looking closely at the actual outcomes we’re achieving.  Though we all do it, it’s totally lame and there is a better way.

Here’s one version of a performance measure logic model that works for all types of programs.  

Performance Measurement Logic Model

The idea is that you’d walk through each step backwards (outcome, output, process, then data) in the simple flow chart and ask yourself, your management team—and maybe if you’re feeling crazy—a customer or two to following questions…

  1. What outcomes are we trying to achieve?  List a couple and by when, if possible. An example of an outcome might be something like, “A safer and healthier work place.”

  2. What program output are ongoing or completed? An example could be “Number of safety issues addressed through completed projects.”

  3. What processes are being tested?  A process example is, “Health and safety inspections completed with work orders generated in the system.”

  4. What supporting data is needed? To close out this example, the data might include number of buildings with asbestos or estimated remediation cost.

Note:  It helps to have your desired outcomes handy.  These should be in your strategic plan—though they might be a little vague.  That’s ok.  Spend (a little) time tweaking, if needed.

While not simple necessarily, this is a straightforward approach to rebooting your measures approach and getting everyone focused on the desired outcomes-- a big win for our programs.

Robin Camarote

I'm the co-founder of Federal MicroConsulting and strategic planning consultant based on Falls Church, VA. I am intent on helping leaders get more done with fewer headaches by outlining clear, creative strategies and solutions that build momentum and buy-in at all organizational levels. In addition to consulting, I write regularly for Inc.com, GovExec.com, and Bloomberg Government on leadership and how to increase your positive impact at work. She is the author of a best-selling book on organizational behavior entitled, Flock, Getting Leaders to Follow and Own It: Drive Your Career to a Place of Happiness and Success. I live with my husband and three children in Falls Church, Virginia.