Customer Experience (CX) and Dominant Path Analysis – Part I

This is the first article on Dominant Path Analysis, in a multi-part series, where we take a deeper dive into the process of improving the customer experience in the contact center through the analysis of Interactive Voice Response flows.

Customer Experience (CX) has become the competitive differentiator in today’s business world. In recent surveys, it’s shown that good CX is a bigger differentiator than price, product or convenience. This is due in large part because of commodity hardware and software, which eliminates price point as a competitive advantage.

The Interactive Voice Response (IVR) world is somewhat different, though. In many cases, an IVR solution is complex and costly. Historically, the goal of most IVR’s was/is to eliminate people calling into the contact center and taking up the time of customer service agents, and, for many customers, this is perceived as a negative.

To overcome this negative perception many companies have begun serious reviews of their IVR solutions. They seek to understand the caller experience in order to so that they can refine the process and create an improved CX experience, and ultimately customer satisfaction, retention and loyalty.

A Top-Down Dive Into Your IVR

The common approach to this review is to take a top down dive into the IVR.  This is referred to as Dominant Path Analysis. Figure 1 shows a diagram of a simple IVR.  In technical terms this is referred to as a tree.

In this example, the IVR team at First Federal Bank might have an assumption about which path is the most frequently used, and thus should be optimized first.  The important thing at this point is to look at the data to understand what is really happening.

Typically this is done by analyzing  “bread-crumbs” (textual representations of the path that callers take through the IVR), here are two examples:

welcome_pp, main_menu_d, checking_d,

welcome_pp, main_menu_d, cardnumber_d

There are four obstacles to doing this in many organizations:

• The data is not available
• The data is not accessible
• The data is not properly partitioned
• The data set is sparse

Not Quite as Simple as a Tree

To further complicate matters, most IVR’s are not a tree structure, but are in fact what Computer Scientists/Mathematicians call Directed Acyclic Graphs (DAG). Figure 2 shows a simple example of what that looks like while Figure 3 shows a sample of an actual IVR.

Looking at Figure 3, one could imagine how difficult dominant path analysis would be by doing simple bread crumb analysis.  The analyst would have to have deep knowledge of the IVR and have the data in a format that is accessible and in a usable format.

Up Next In this Series

The next article in this series will be a deep dive into data analysis using an actual case study…

Getting Help to Improve IVR Flows and Your Customer’s Experience

This deep data analysis is something most companies seek in an experienced consultant. This is something we do every day at PTP, but also bring the deep understanding of the customer experience, the processes and departments involved, and the holistic view of the customer. It all ties together to create that seamless experience that has become the main differentiator for companies to thrive and succeed.  You can learn more here, or simply contact us to see if we can partner to help you deliver the most strategic CX solution to improve revenue.