Our Sites

Performance metrics drive behavior

The best metrics help a shop improve, but the worst create wasteful conflict

performance metric map

Providing a big-picture view, a metric map overlays targeted metrics over the org chart.

Ever notice employees who seem to act irrationally, and yet they think their behavior to be entirely rational? How can this be? It might have to do with your performance metrics.

Performance metrics identify what is important and can be the source for meaningful feedback. They speak loudly to your workforce, leadership, owners, and other stakeholders. And they drive behavior. The question is, Do they drive the right behavior?

How Does a Metric Drive Behavior?

Deep down we’re all competitive, though some of us are more competitive than others. When people understand what an attainable performance metric is intending to measure, they will work to achieve the desired result. That’s positive behavior. If the metric is beyond people’s capability to influence, or is just outright impossible to attain, then you can expect negative behavior. Why should I try? I’ll just keep my head down until this passes.

Think about metrics in terms of baseline metrics and target metrics. Baseline metrics tell you about your current state or starting point; target metrics tell you where you want to be. Both are critical if you want to drive people toward the desired outcomes.

Imagine Richard, a welder in the final weld operation. Have you provided Richard with meaningful performance metrics? If he spends his day just doing whatever he can, does not get feedback, and assumes that whatever he has done is OK, then you probably have not provided Richard with the guidance he deserves to have.

On the other hand, if Richard knows the pace he needs to be producing at (takt time or cycle time), understands what his quality specifications are (quality at the source and standard work instructions), and knows how the material flows (visual controls and 5S), then he should be adequately prepared to do a good day’s work that is in line with what the company needs him to do.

What happens if people aren’t aligned around organizational performance metrics? Chances are you have chaos as employees work cross-purposes, engage in unproductive competition, or “suboptimize the whole” by optimizing their small piece of the overall process.

One way to create alignment is with a metric map. The map lays the targeted metrics on top of the organization chart. It shows alignment among levels, identifies which performance metrics are important, and describes how the various components of the organization fit together. How a metric map looks depends on the business and how it’s organized. Regardless, the logic of the metric map should work for all businesses.

How Do I Define a Performance Metric?

Once you have identified what you want to measure, give that performance metric a name. Make it meaningful and relevant.

Think of Richard in the final weld area. An important measure might be “defects per shift.” First, you need to define what a defect is for each job, based on an established welding code, company-specific standards, or a combination of both.

Now that you’ve defined the metric, consider how it will drive Richard’s behavior. Clearly, he’ll want to avoid defects and get welding right the first time. Richard should know this process is being measured; his performance is directly tied to the metric (his work affects the outcome); and supervisors can help and give constructive, positive feedback.

Next, assess the inputs to the metric. What are the sources of inputs? What are the precise components, and how exactly are they calculated? How is the information gathered, and how frequently? Old data is less meaningful. Is the data gathering repeatable? If data can’t be gathered the same way twice, the resulting metrics might be meaningless. What about the units of measurement? These questions and more will help you to determine if the performance metric is really feasible.

How you gather the data can be critical. The best data gathering occurs automatically. Software and other information systems make automatic data collection and analysis more practical in more areas of the shop—think machine uptime tracking and arc-on-time monitoring. But Richard’s “weld defects per shift” data would likely be gathered manually, perhaps using a combination of visual inspection and nondestructive and destructive testing.

Now that you have the data, what are you going to do with it? How do you convey and communicate it, and to whom? Think about how people will interpret the data. How can you use that data to manage the business and drive desired behaviors? Does the performance metric have the desired effect on behavior?

As you can see, all this involves much more than just saying, “Hey, let’s measure such and such!” If you follow a disciplined approach, the metrics you gather are more likely to be highly effective.

What Makes a Metric Effective?

Good performance metrics should help you deliver results to the business. Poor performance metrics do not help employees understand what is expected.

Poor metrics often put different individuals or functions at odds with each other, creating unnecessary tension and distraction. If the metrics focus only on the outputs of a particular person or function (a welder or welding department, for example), and they are not balanced with upstream processes, then the information you get from those metrics won’t be very useful. Upstream processes could have contributed to the problem, and those processes have already happened, so it’s too late for corrective action.

Say Richard receives a batch of work with suboptimal weld preparation. It’s not terrible. Everything is within tolerance for the job, barely. But Richard’s a skilled welder, so he makes it work.

Still, Richard produces more defective parts than usual—garbage in, garbage out. In this case, measuring those defects probably wouldn’t give you meaningful information, because the problem was in weld preparation, where tolerance standards might need to be re-examined. By the time the work reaches welding, it’s too late to correct the problem.

A performance metric that drives the wrong or unintended behavior is counterproductive. The classic high-product-mix example deals with workcell efficiency, like parts per hour. The more parts per hour the people in a workcell produce, the better their performance metrics look. But then they sequence jobs in a way that optimizes their productivity and yet creates processing headaches downstream. They look great, but their behavior leads to unintended consequences.

Defining Good Metrics

If your performance metrics provide clarity and motivate people, you are heading in the right direction. Do the metrics set expectations consistent with both your company’s needs and your customers’ needs? Effective metrics create alignment among different individuals, functions, and processes within the organization.

Ultimately, effective performance metrics help get work done. For instance, imagine now that Richard works in an environment with effective metrics. He still works with the weld-defect performance metric, but that metric works in concert with measurements upstream and downstream. Richard rarely receives workpieces with poor weld prep. Why? The weld prep department has its own performance metrics that make Richard’s job easier. In this environment, Richard is a more effective and willing contributor. That is good for Richard and it is good for you.

If your metrics help Richard be more productive and give him objective feedback, he then will view the metrics as constructive. On the other hand, if Richard views metrics as a way to “catch me doing something wrong”—that is, as something punitive—then, obviously, your metrics won’t be well-received.

Performance metrics should help all Richards be effective contributors, regardless of where they work or what their job titles are. All employees need and seek guidance on what it takes to do a good job, and they need feedback. Performance metrics are an important part of the equation for a successful employee.

Jeff Sipes is principal of Back2Basics LLC. If you have improvement ideas you’d like to read about, email him at jwsipes@back2basics-lean.com or Senior Editor Tim Heston at timh@thefabricator.com.

  
About the Author
Back2Basics  LLC

Jeff Sipes

Principal

9250 Eagle Meadow Dr.

Indianapolis, IN 46234

(317) 439-7960