If you’ve heard me give a performance management presentation, you know one of my favorite topics is the pursuit of improved performance in call centers. When I first started using them as my prototypical example several years ago, organizations were in the midst of moving call centers offshore. Less expensive human agents and AI-based electronic agents would allow organizations to reduce their costs and provide higher levels of service – or, at least, so the vendors and consultants claimed.
As I wrote last year in Contact Professional (web site no longer active, PDF), most of these efficiency metrics don’t consider the level of service from the customer’s point of view. As such, the don’t answer fundamental questions such as the following:
- Should service level focus on how long a caller waits before starting to talk to someone (often called availability) or instead how long it takes to get his or her question answered (often called speed of answer)?
- If a call center focuses on availability, should it track the average delay time for all callers or the longest delay for any one caller?
- In other words, do individual customers care that the average delay was 45 seconds or that the delay when they called was 45 seconds?
- Furthermore, do they care that they normally only wait 45 seconds or that they occasionally have to wait 180 seconds?
While average delay is a potentially useful internal metric, it is difficult to tie it to any outcome objective and therefore rarely belongs on a performance management scorecard. On the other hand, since customers tend to remember their most negative experiences, ‘longest delay of any one caller’ is usually a better quantitative indicator of customer satisfaction.
Now, a few years later, the offshoring mania has subsided and more than a few contact centers have begun to “on-shore”. In my more recent conversations with call center managers, many of them cite poor customer satisfaction as the primary reason they have moved more of their call volume back home. This realization gave me hope that their metrics might have matured as well.
Unfortunately, an informal (and very unscientific) survey leads me to believe that the most popular metric in the call center remains ‘average length of call’. The shorter, the better. I can’t speak for the rest of you but I would rather have it take 10 minutes for a rep to answer my question clearly and completely than to get off the phone quickly only to discover I have to call back for more information.
So why the continued disconnect between their objectives and their metrics? Despite a continued evolution toward a unified contact center, I think the answer primarily stems from the fact most organizations still consider their call centers as overhead, rather than a potentially strategic asset. Therefore, any investment in the call center must be justified as lowering long-term costs (an efficiency argument) instead of improving customer satisfaction, market awareness, or competitive intelligence (effectiveness arguments).
Call centers can bring this treatment upon themselves. By only reporting efficiency measures, they reinforce this notion of how they should be graded. If all senior management knows about their call center is that it answered questions more quickly this month compared to last month, how can it possibly understand its tie to the organization’s strategy?
To make matters worse, as I describe in CP Wire (web site no longer active, PDF) , most contact centers make the same critical mistakes when they report performance. These reporting mistakes are made internally during weekly operational reviews and then repeated during the quarterly organizational assessment. By making these mistakes, call centers undermine the organization’s confidence in their own performance – keeping them relegated to overhead status, rather than being viewed a strategic asset.