QUICK SUMMARY
Performance measurement is about deciding why an agency or organization wants to measure performance (since different purposes require different measures) and then choosing measures that fit that purpose.
VIDEO OVERVIEW
STRATEGY DETAILS
STRATEGY DETAILSQ1. What is performance measurement?
Performance measurement sounds like a self-evident concept, yet doing it well requires thoughtfulness and savvy. That's because performance measurement requires, first and foremost, deciding why your organization wants to measure performance, since different purposes require different measures. The next step is then choosing measures that fit that purpose.
Q2. What are the main types of performance measures?
Some of the main types of performance measures are shown in the figure below, presented in the form of a logic model. They include:
- Input measures: Measures related to the resources used by the organization (or program or office), such as money, staff time and anything else needed to do the work.
- Activity or process measures: Measures related to the everyday work of the organization. Put another way, these measures show how inputs get turned into outputs, which are discussed next.
- Output measures:Measures related to what the organization produces. For example, for a health department that could be the number of houses checked for lead paint. Or for a training program it could be the number of individuals trained.
- Outcome measures: Measures related to what the organization aims to achieve. These are sometimes divided into short-term, medium-term and long-term measures. To continue the job training example, for instance, a near-term outcome measures might be the number of training graduates (let's assume they were unemployed) who become employed within six months of the training. A long-term outcome measures might the the percentage of graduates who are still employed (in any job) three years later.
- Another category that's not shown in the figure is efficiency measures, which are created by dividing outputs by inputs. For example, the number of job training staff (an input) divided by the number of training graduates (an output) is an example of an efficiency measure.
Q3. Can you walk me through an example?
Let's take the case of a city health department and, in particular, a division that focuses on preventing lead poisoning. The main inputs include the staff and funding to do their work. Activities include doing outreach around lead paint abatement, training staff, setting up appointments to test individuals for lead. Outputs include residences checked for lead paint and, another output, children tested for lead. The primary intended outcome is, say, a declining lead-poisoning rate in the city.
So what's the key difference between outputs and outcomes? It is "locus of control," meaning that outputs are largely within the control of an organization, while outcomes depend on multiple factors. But don't get hung up on worrying if your categorization is right or not. (Is this an activity or an output? Is is an output or an outcome?) First of all, these types of categorizations take practice. Second, do what makes sense to you and your team and you'll be in good shape.
Q4. We've identified inputs, activities, outputs and outcomes; how do we create associated measures?
Let's take an example. The staff for an agency is an input. An associated measure might be the "The number of staff by quarter." What's the difference? A performance metric has three components: 1) a quantity, 2) the thing being measured, and 3) a time period. In this case, "the number of" is the quantity; "staff" is the thing being measured; and "by quarter" is the time period.Here's another example. Say that one output for a city transportation agency is be potholes filled. The associated performance metric might be: The number of potholes fixed per month. In this case, "the number of" is again the quantity. (Another metric might start with, "The percentage of...") The "potholes fixed" is the thing being measured. And "per month" is the time period.
Q5. What measures should your organization track?
Maybe the most useful piece of advice on performance management comes from Robert Behn of the Harvard Kennedy School who advises, "Always start with purpose." By that he means: Always start by being clear aboutwhy you're measuring performance. In his Public Administration Review article, "Why Measure Performance? Different Purposes Require Different Measures," Behn provides examples of some of the possible reasons for measuring performance and some of the different types of measures that would be useful:
- To motivate staff: That requires real-time outputs compared with target levels.
- To evaluate organizational performance: That requires examining changes in outputs compared to changes in outcomes.
- To promote your organization to stakeholders and the public: That requires easily understandable aspects of your organization's performance that citizens really care about.
- To celebrate (with your staff) your organization's success: That requires periodic and significant performance targets that, when achieved, provide people with a real sense of personal and collective accomplishment.
- To improve organizational performance: That requires measuring inside-the-black-box relationships that connect changes in operations to changes in outputs and outcomes.
Q6. Why is being aware of unintended consequences so important with performance measurement?
This piece of advice goes beyond just performance measurement and steps into performance management (i.e., how the measures will be used), but it's important to always be aware of possible unintended consequences when measuring performance and using those measures to reward performance, even if it's simply celebration. For example, creating high pressure on specific measures (and targets) can lead to distortions. To take an example, if DMV staff are under strong pressure to process people faster, customer service or accuracy can drop. Examples like these underscore the importance of 1) Being wary of creating too much pressure on any given measure and 2) Having a comprehensive set of measures that leaders track (to watch out for distortions), even if those leaders emphasize just a few key goals to their staff.
ADDITIONAL RESOURCES
ADDITIONAL RESOURCES- Article: "Why Measure Performance? Different Purposes Require Different Measures," by Robert Behn inPublic Administration Review.
- Book: Performance Measurement: Getting Results by Harry Hatry
- Gov Innovator podcast episodes:
- Successful performance management starts with purpose: Robert Behn, Harvard Kennedy School
- How to design performance measures to better measure impact: Peter Schochet, Mathematica Policy Research
- Avoiding performance perversity: Donald Moynihan, (then) La Follette School of Public Affairs, University of Wisconsin-Madison
CUSTOMIZED ASSISTANCE
CUSTOMIZED ASSISTANCEPlease contact us if your organization needs help with performance measurement, including staff trainings or workshops on the topic.
PDF VERSION OF THIS PAGE (click here)
