Measure Outcomes, Not Activities, to Improve Customer Service

Ann All

It's no surprise most organizations employ metrics that measure activities rather than outcomes. It's a variation of the good, old-fashioned path of least resistance. It's easier to monitor average call-handling time in contact centers than it is to evaluate how the length of calls might impact customer satisfaction.


I've written about this topic numerous times. Even if organizations understand the importance of measuring outcomes rather than activities, it can be difficult to move in that direction. Some tips included in a paper produced by The Verghis Group, a consulting company focused on customer support, might make it easier. The subtitle says it all: The New World of Guiding not Grading.


Using outcome-based metrics makes it harder for employees to game the system. (For instance, contact center agents can cut their calls short to reduce their call-handling times. Never mind what this does to customer satisfaction.) It will also engage employees and get them more directly involved in service improvement efforts. From the paper:

Since [outcome-based goals] are inherently harder to measure, and harder for an individual to manipulate, it's necessary for a team member to ask questions and seek to understand before they can attempt to solve the problem.

Some of the suggestions from the paper:

  • Create a litmus statement, which the paper defines as "a short, sweet statement that describes uniquely what your team is accountable for." One example offered: "We own subscriber loyalty."
  • Create three or four high-level measures that align with overall organizational goals. They should be cross-functional and apply to all teams. A good starting point is what the paper refers to as "the golden four": increasing revenue from new customers; increasing revenue from existing customers; reducing the cost to serve customers; and customer/employee loyalty or engagement.
  • Don't focus too narrowly on any one metric as an indicator of the general state of the overall customer relationship. Many organizations do this with customer satisfaction, according to the paper. That's an especially bad idea, given the small base of customer feedback from which a satisfaction measure is usually derived.
  • Focus on no more than a handful of measures. Each should be easy to explain, with an easy-to-understand intent, and as much as possible they should be from the customer's point of view. The paper suggests that any measures that do not help with at least two of the goals should not be part of the senior team's dashboard. (They can be monitored at the project level, however.)


After doing this, a logical next step is determining a few, highly focused projects involving these metrics. The paper suggests involving frontline workers in this process:

Mix up teams in terms of experience, geography, skill levels and even across functions. Let them develop the messaging, including the "What's in it for me?" message for stakeholders. Listen to them and let them lead. Do this and watch employee morale soar-with all the goodness that emerges from that.

This approach won't be easy, the paper cautions. It's tough to give up that kind of control, especially in environments like contact centers where tight controls are usually in place. But empowering employees in this way can yield clear benefits. Last summer I wrote about American Express, which cuts its contact center turnover rates and nabbed top spots in the J.D. Powers & Associates' customer satisfaction rankings when it gave employees a more active role in improving service.


Poking around The Verghis Group blog, I found the company will offer a workshop on this approach on March 18 in the Boston area. More details, including pricing, are available online.

Add Comment      Leave a comment on this blog post
Jan 11, 2011 12:07 PM Phil Verghis Phil Verghis  says:

Hi Ann,

Thanks for the kind words. Getting to the 'litmus statement' was the most powerful part with most of my clients. Once they saw what it was they needed to do, other behaviors were much easier to change.


Post a comment





(Maximum characters: 1200). You have 1200 characters left.



Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.