Expert's Angle: Analytics and the Future of Contact Centers: Success Based Analytics and Quality Model Analytics

What changes are in store for the contact center in terms of analytics? This article series explores some of the major trends that will impact contact centers of all types over the next ten years. Read the previous article.

One of the over-arching topics implied throughout this article series has been how to use success metrics in order to predict the future behaviors of clientele as part of the analytical model. Here, we will discuss how to turn success metrics into actionable analytics, and explore how moving from a task-based quality model to a behavior-based quality model will open analytic possibilities to improve the overall operations of the contact center.

Matching Caller Behavior to Future Interactions

The Analytical Approach essentially consists of taking common variables and behaviors to come up with predictive models and using past behaviors to predict future behaviors. Another component of this approach is capturing – and using - common demographics and characteristics to further develop segmentation models. These variables within the process will give the analytical contact center the cornerstones for developing treatment models that will increase the reach of the contact center and decrease the overall cost-per-contact.

Being able to predict future behavior of clients, as well as preferred contact channel interactions, should help illicit a positive response from clientele as they will see the contact center as less obtrusive and more attuned to their specific needs and preferences. The future of contact center interactions will be strongly based on specific and customized messages delivered during IVR delays all the way to providing relevant and timely information to the contact center and the business’s clientele. Other industry experts and articles offered on this site have described the “Voice of the Client” as being the ‘Future’ of the contact center and the customized, analytical approach described here is in line with this thinking. By customizing the channels by which we are able to reach out to the public, as well as the customized message, we are making the interaction process much less intrusive and more meaningful to the client and reacting to their implied preferences and needs.

Success-Based Analytics and the Agent

While the analytical contact center dedicates a considerable amount of effort and attention to predicting the future success of a campaign, a similar amount of effort should be paid to investigating the Key Performance Indicators (KPI) that are used to measure the success of the operations and the agents. Analyzing outcome-based KPI will provide key insight into what makes an operation successful and how that success can be repeated with other contact centers within the business, or within other teams within that contact center. The Quality Model is one of the often-overlooked aspects of analyzing your own contact center.

Quality Model Analytics

Most of the Quality Models active in today’s contact centers are task-based. Simply put, a task-based model asks the evaluator whether the agent performed the required task. Did the agent verify the caller? Did the agent enter the interaction notes correctly? In some cases, the evaluator can score how well an agent performed these tasks on a scale of 1-5. However, this type of model is fraught with inherent risks:

• Are the evaluators aligned with what has been agreed on for range scoring?
• Does the task-based model fully capture the strategic intent of the contact center?
• Is it possible to account for every type of interaction within a task-based model?

It is important to bear in mind that moving to a behavior-based Quality Model requires that the actual strategic intent of the contact center is taken into account, as well as whether or not the agent involved has achieved that strategic intent. By scoring on agent behavior rather than task, the contact center is better able to understand some important aspects of its Operations.

A great example of this is analyzing behavior scores for the center to determine where possible gaps in training may exist, as well as what future topics for refresher training can be placed into the continuous development plan. When analyzing behavior-based models for common errors, the contact center may also find process and procedures that are being misunderstood or which simply don’t work, either for the agents or for the client. The act of thoroughly utilizing and analyzing the information at hand provides a holistic view of the contact center.

Conclusion

One of the common challenges with any contact center is the potential for "analysis paralysis," because there is so much information available for review. The amount of data that needs to be waded through to uncover some of the gains we have discussed in this series may only seem daunting. However, the contact center needs to develop a strong governance program that includes all key stakeholders.

It is not unusual for contact centers to seek out advice from vendors who specialize in harnessing and analyzing the 'Big Data' to provide the keys to the segmentation models. This is very specialized work, but the payoff for the effort can be quite high. The contact center should be able to increase its reach and cut down costs through automation, as well as increase its overall success rates. Improving the relationship with client base is also an immeasurable affect that can be realized as a customized approach to the client can be achieved through this enhanced understanding of the 'what and 'how' messages and interactions can be delivered.