Creating Visibility into KM Programs
New initiatives in knowledge management are common in both technical and transactional support organizations. The goals are generally to improve customer satisfaction while increasing the capacity to handle cases. However, it is difficult to judge success and set strategy because of the lack of “visibility” into the effects of these initiatives. The following is a set of best measurement practices for judging new KM projects, and a metrics dashboard to augment current measurement plans.
Organizations often lack the content to support their contact center’s ability to quickly answer questions or to allow customers to serve themselves. To address this gap, customers often try to clean up and add new content. One popular organizational model is “knowledge-centered support” (KCS), which shifts the content creation responsibility to the representatives themselves. KCS is an excellent idea; the problems lie not with the programs themselves, but with the inability to accurately measure and tie these programs to strategic goals.
Let’s examine the proposed measurements:
1. Measure and group direct labor on problems, not just elapsed time. For complex technical support centers, there are multiple interactions or activities required to solve problems. When there is no visibility into the components of direct labor, it’s difficult or impossible to know the components of agent activity. For example, when agents are first asked to contribute content as they close cases, direct labor may increase until some of the contact gaps are filled. As coverage increases, direct labor will fall below preexisting levels because more agents are solving problems with existing knowledge. If there is no visibility into how the components of direct labor change (talk time, after-work time and administration time), the initial increase in direct labor can be mistaken for failures in the program.
2. Group or categorize calls according to the amount of real labor required. Understanding the direct labor making up average handle time scores is a great first step. However, if calls are grouped into buckets simply by the amount of labor each case takes, you can develop a deeper understanding of problem types, allowing you to focus on where to spend content efforts.
3. Understand the differences between shorter and longer calls. Call-center managers often believe that short cases are more likely to be self-served, while long cases require additional expertise. In general it’s a good assumption. Even though there may be many short calls, these short calls may not account for a significant amount of overall costs. Alternately, even a small number of long calls may be disproportionately expensive. Long calls often involve simple problem-solving, but with multiple steps and long wait times between the steps. In some cases, it’s easy to create self-service content for these problems. Without visibility into the different kinds of calls, these special cases are often ignored.
4. Assess adoption of your KM program. If your processes are not being followed there will be no benefit. If representatives are not creating content when they create a solution, the team can’t benefit.
5. Measure the availability of and gaps in content. If a key part of your strategy is to create relevant content, you need to know how much you have, how much you need and your progress toward those goals.
6. Understand the potential for self-service among the groups of calls. Once you understand differences between short and long calls, it’s important to understand the potential for self-service. Take a random sample from each group and do a simple assessment of the potential for it to be handled by self-service. This helps you understand the capacity-improvement “potential.” If you are creating content customers are using to solve problems, this “potential” will fall over time as more and more of the cases are solved online.
Editorial note: This article contains a chart which can be viewed in the PDF version online or on Page 4 of "Best Practices in Customer Relationship Management", July/August 2007
KNOVA, a Consona CRM solution, maximizes the value of every interaction throughout the customer lifecycle. Built on an adaptive search and knowledge management platform, KNOVA’s suite of applications integrates with CRM implementations to help companies increase revenues, reduce service costs and improve customer satisfaction. Industry leaders including AOL, Ford, HP, Novell, McAfee and H&R Block rely on KNOVA’s award-winning service resolution management applications to power an intelligent customer experience on their Web sites and within their contact centers. For more information, visit www.knova.com.