Customer service metrics Featured

10:20am EDT March 30, 2006
British statesman Benjamin Disraeli once said, “There are three kinds of lies: lies, damned lies and statistics.”

InfoCision’s Steve Boyazis respectfully disagrees. Boyazis believes that you can glean plenty of good customer-care information from statistics — providing you know what you’re looking for, and providing you can get apples-to-apples statistics.

“But you have to dig deeper into the numbers,” Boyazis says. “Just looking at whatever numbers your call center provides doesn’t lead to satisfied customers.”

Boyazis is senior vice president for business development at InfoCision, which provides teleservices to corporate America. In an exclusive interview with Smart Business, here’s more of what he said.

What are the most important statistics when judging the effectiveness of call center operations?
The two or three basics that people think of when they think of customer care in call centers are: (1) How fast is my call answered? (2) How long are people on hold? and (3) How many callers abandon the call before being serviced?

For instance, are 80 percent of your calls answered in 20 seconds or less? Are less than 5 percent of your calls abandoned? The longer people are on hold, the more likely they are to hang up. If they call to order something and they’re put on hold, they get frustrated or change their minds. If they call to complain, they just simmer and get angrier.

Of course, it depends on the business or industry you’re talking about, but generally 20 to 30 seconds is acceptable as a wait time. When you start getting up near five minutes, that’s when people start getting frustrated.

What are some of the nuances behind those numbers?
You have to find out what’s driving the numbers, and you also have to determine what each customer is worth.

For instance, we recently ran a test with a client interested in ‘save’ calls. Those are callers who are calling to downgrade or stop a service. The company felt its current vendor was doing a pretty good job. It had a 3 percent abandonment rate, a wait time of 2.5 minutes, and it dispositioned 37 percent of the calls as ‘savable.’

We were able to improve the wait time to 20 seconds, which led to the abandon rate decreasing to 1 percent. And — because our agents better understood the customers, their needs, and what levers we could pull on behalf of the service provider to ensure customer satisfaction — we identified almost 43 percent of the calls as savable.

What does that mean? That’s where the ‘nuances’ come in.

Let’s say the opportunity is 100,000 phone calls. With 2 percent fewer abandons, that means you have 2,000 more opportunities to talk to customers. With 6 percent more dispositioned ‘savables,’ that means 6,000 additional opportunities. If both companies saved 1/3 of their potentials, we’d save 2,600 more clients than the competitor, which simply defines their key customer service metrics by wait times and abandons.

But it even goes a little deeper. To really decide how aggressively you want to save customers, you need to understand their lifetime value and build an ROI (return on investment). For instance, if the value of the customer is $500 and you just paid $30 for a customer acquisition, that’s a fantastic return. If the value of that same customer is just $20, then the effort isn’t worth it.

Are those kinds of numbers fairly typical?
The customer service numbers are industry standard. But you really have to start by understanding them at the front end: why the customers are calling, what they need, what you can do to fix them. And you’ve got to identify as many of those people as humanly possible that are savable. Then, by applying the lifetime value of that customer, it’s easy to put together an ROI on every customer who’s a candidate.

What about customer service minimums?
There are several ways to handle calls. One is scripted with few options, for frequent, well documented programs. Another is the more complex relationship-building call, where you have to understand the customer and what he or she needs to be saved and what’s going to work for that particular customer.

What are other factors determining level of service?
Volume is one. Different call centers are set up to handle different volumes. In order to be able to train people to a certain level, you have to have enough critical mass.

Another is language. Customer care provided by folks with different language skill sets has an affect on the ROI. If you go offshore, the cost is much lower; if you choose a domestic provider, you get a better customer experience with agents who speak the same language as the caller.

STEVE BOYAZIS is senior vice president for business development for InfoCision, Inc. Reach him at (330) 668-1400 or