Jerry Sokol writes for Biz Meets Tech and he recently wrote about something that’s always top of mind at Interaction Metrics—good customer experience metrics.

Numbers are only useful insofar as they help you improve—simply tracking outcome metrics or just collecting data is never the answer. The right customer experience metrics don’t just tell you that customers are unhappy—they show you where, why, and what specific customer types are the most displeased.

If your numbers can you tell you the specifics of the problem, then next steps become abundantly clear. Jerry explains the dilemma businesses face when it comes to customer experience metrics, and poses good questions about how to improve customer service.

Jerry Sokol: Business thrives on numbers. Because numbers remove ambiguity and allow for clinical decisions. Now, of course numbers can be misused (“lies, damned lies, and statistics”), but for now, let’s sidestep that and talk about numbers being used for good.

Finance does this best. Finance metrics all boil down to “how much did this cost and how much did I make?” Very simple. Since money is involved, you had better believe that the backup is there to not only support those numbers, but help point to how to improve them.

As we get away from finance, things are not so clear. We want metrics that unambiguously tell us what’s going on with our operation, much as profit and loss do for finance. In fact, we want metrics that can, in some way, eventually correlate to profit and loss. And I would submit that most of them aren’t very good.

This is particularly true in customer care. Don’t get me wrong – the standard call center metrics absolutely have their place, and I’ll talk about that later. But they correlate very poorly to customer satisfaction, real cost and true agent performance. What’s worse is that they are frequently misused to drive bad and costly practices.

The industry attempts to address this problem have been poor. The mantra is that “we are swimming in data”, and through the magic of better organizational tools, the problem will be solved. We’ve had better charting of the same numbers, data warehousing, Business Intelligence tools, and now Big Data². Yet I still see the struggle for actionable information virtually everywhere I go.

From what I see, we’re going about this backwards. We are delivering metrics without first considering what QUESTIONS we want answered.

Given that the entire raison d’être of a call center is to address a caller’s concern, the #1 question to answer is “Did we solve the customer’s concern?” OK, First Call Resolution almost answers that, and is pretty well understood (if not used) by most call centers.

And the follow up is, “If not, why not?” Now it gets tougher.

Zooming out further, what about the most obvious question? How about “Why did they have to call in the first place?” This information – the tracking of call drivers – is rarely done well if at all.

With those three questions, I’m sure most of you could figure out what kind of information you would have to collect, and even come up with metrics that you could use. But, I’ll help more in future posts.

As a final note, ACTIONABLE is the important word. Your metric can be useful, but if the backing data doesn’t point to how you can improve, well…

Jerry Sokol is an independent consultant that improves customer service by improving how a call center’s processes and technology work together and align to the business.  Read more about him at

Categories: Customer Satisfaction Metrics
E-book cover

Get our short CX Trends Report, published quarterly.

  • This field is for validation purposes and should be left unchanged.

Written by the analysts at Interaction Metrics, we highlight the latest developments in the fast-changing world of CX.