In the 1960s, as mainframe computers were entering the business world—and inspiring much awe—sociologist William Bruce Cameron observed that it would be nice if everything could be represented in numbers and “run through an IBM machine.” But he warned, “not everything that can be counted counts, and not everything that counts can be counted.”

It’s a quotation sometimes attributed to Albert Einstein, and though the source may not be clear, the message certainly is. And it’s fair warning. You have to be mindful about the metrics you establish, how much you read into them, and how you use them to guide decisions. Here are some common “metric pitfalls” to avoid:

Having blind spots: I recall meeting with executives of a large firm that provides HR and payroll services. They had won awards for the quality of their contact center service. But they were seeing some troubling downward trends in customer satisfaction. Why? What was wrong? As we discussed the problem, I pulled up their social channels on a laptop projecting to a screen at the front of the room—and heard a collective gasp.

Alongside well-intentioned and upbeat messages from the company was a string of posts from a frustrated customer who needed help with a simple password reset, but had received no response. His posts were in all-capital letters; it was clear he was upset. The company didn’t have a service presence (let alone metrics!) in some of the channels their customers expected. It was time to broaden the reach of what they were measuring.

Establishing too many metrics: Another pitfall is having far too many metrics, along with the message that they somehow must all be tracked and managed. There is a tendency for that to happen: someone sees an improvement opportunity or operational gap, believes that a report could provide visibility, and adds it to the mix. However, there’s often a lack of similar efforts to cut down and eliminate reports that aren’t helpful. Or to differentiate between supporting metrics and those that are most important. Clutter has a significant downside. If there are too many metrics, your team’s focus will be diluted.

Painting an overly optimistic picture: Reporting the contact center in the best possible light can undermine success. There are a lot of ways to produce reports so that the center looks as effective as possible to senior management. However, if you mask serious resource deficiencies or problems with processes, the center is less likely to get the resources and support it needs. That, in turn, will undermine your ability to perform. There’s also the related issue of psychology. When upper-level managers see room for improvement, they tend to feel more assured they are getting the whole story (plus, being transparent and presenting the whole picture is the right thing to do).

Creating conflicting messages: I too often see service metrics that conflict with each other, leaving the team to guess what’s most important. For example, an insurance company recently began focusing on first-contact resolution. This is a wise move, and customers appreciate it; it contributes to true effectiveness. But their cost per interaction began going up, largely for a reason that is not immediately obvious: they have proportionally fewer interactions, because they are preventing repeat and unnecessary work. So fixed costs are being allocated across fewer contacts. Their total costs are going down, and they’ve had to re-educate managers to not look at average cost per contact in a vacuum. It doesn’t tell you everything you need to know.

Encouraging the wrong behavior: An all-too-common pitfall I see is when metrics encourage the wrong behavior. Perhaps you’ve had a service technician say, “Hey, you will receive a survey within the next week, and if you don’t provide a top score, it will reflect on me.” That kind of conversation—begging for scores, as some call it—is awkward for employees and customers. And it defeats much of the value in what you can learn about the customer’s experience.

Trying to match “benchmarks”: As you work toward achieving important objectives, it’s essential that your team recognizes what the metrics are really saying. Don’t get trapped by benchmarks. For example:

• Maybe your first-contact resolution rate is lower than what similar organizations say they are achieving; but you might be measuring it more rigorously and/or working to handle easy contacts without the assistance of agents.
• Your center’s average handling time (AHT) might be higher than that of other centers. However, if you’re using that time to prevent repeat contacts, refine processes and capture information that can improve products and marketing, your efficiency (and associated costs) may be far better than they are for those other centers. Your service level or response time objectives may be more modest than others in your industry. But you might actually be doing better than they are, if you hit them increment after increment, day after day.

It can be a minefield out there. But it doesn’t have to be. Not if you are aware of these pitfalls, and do the work necessary to leverage the potential of your metrics to drive innovation and improvement. Jeff Rumburg, co-founder of research firm MetricNet, estimates that 100% of organizations have KPIs, 80% can define their KPIs, 30% understand KPI cause and effect, 10% use metrics to continuously improve, and around 5% leverage KPIs for “world-class” performance. I concur with these estimates—there’s a lot of opportunity in many contact centers.

Excerpt from Contact Center Management on Fast Forward by Brad Cleveland.