Six Sigma for Services: The Great Divide
By Phil Verghis on June 2, 2008(taken from the June 2008 Verghis View newsletter)
Major philosophical differences exist between managers who run large-scale operations - ones with a massive number of incoming calls - and operations with a lower volume of quite complex calls.
One big disparity revolves around metrics. High volume support centers generally concentrate on metrics that show how smoothly their operations are running. These managers know that slight changes in efficiency make a significant difference in a unit’s profitability. The folks running complex support centers, on the other hand, tend to dismiss Six Sigma and focus on things like customer satisfaction or loyalty.
Is there a happy middle ground?
I spoke recently with Geetha Panda, Hewlett-Packard’s worldwide head for service delivery excellence. She explained a pre-emptive escalation model they’ve developed in their Bangalore service delivery center. The model was based on a study of the relationship between output metrics (customer SLAs and KPIs) and the input metrics that directly or indirectly affect them.
This derives from Six Sigma, in which you focus on the things that you have control over (e.g., staff availability), and ignore the things that you don’t (e.g., customer satisfaction). By ensuring a good end-to-end understanding of the implications of decisions made along the way, a strong alignment is created. This reduces variation and unpredictability, which in turn helps ensure that the outcomes are more likely to be statistically “in control.”
Geetha cited one example of how this approach has transformed HP’s Bangalore center. They took over an exceptionally high-profile, at-risk account from a native English-speaking center, and successfully converted them into fans again. She credits their pre-emptive escalation model for this win. In fact, the model has been so successful that HP is now extending it to other centers around the world.
As I reflected on my conversation with Geetha, I wondered why many in the support world - especially those who run lower volume, high complexity support centers - tend to dismiss Six Sigma. I would guess it’s the apparent focus on process and metrics to the exclusion of everything else. Think about it. Complex technical support inherently has a “people first, process second, technology third” hierarchy. Six Sigma’s priorities seem to be process first, followed by people and then technology.
But is that really true? A careful analysis of Six Sigma shows it’s not. In fact, one of its key tenets is gathering input and suggestions from those actually doing the work.
Are you successfully using any aspects of Six Sigma in your support organization? Let me know. I’m particularly interested in lower-volume, high-complexity support organizations using Six Sigma.
RSS Entries
Powered by
We are not formally using Six Sigma today however we’re beginning to send our Front Line Managers to Six Sigma training later this year to see what all the fuss is about. Our Call Center is very process oriented and have a firm implementation of ITIL Incident, Problem and Change Management, plus a philosophy of continual improvement.
One thing I disagree with in your article is tossing out things like customer sat scores because you can’t control them. In our Call Center we monitor very closely those things we can control - availability, handle times, average wait and service levels and we routinely far exceed our targets . We use our customer sat rate to validate that our customers like our performance. Generally our customers provide things like testimonial comments that you just can’t get by looking at statistics. Moreover, we ask our customers to rate our Technician’s performance in our surveys. This helps to separate dissatisfaction customers may have with a particular system from how they feel about the Tech. The customer sat stats appear on our section scorecard which attempts to provide a balanced view of performance.
Hi Tony,
I actually agree with your point about not tossing things like customer sat out. It is still important to measure customer sat (and loyalty). The difference between what you do and what most people tend to do is that they fixate on ‘fixing’ customer sat - something they have no control over - instead of the issues that directly influence it.
Sounds like you are using a number of techniques (ITIL, Six Sigma, Balance Scorecard) to get a good handle of what your customers want. Nice job!
Peace,
Phil