Humana reduces call center volume with text analytics

Health insurer Humana's provider call center handles more than 1 million calls a month with inquiries ranging from eligibility and coverage to more detailed conversations regarding specific claim benefits. As part of maintaining customer service excellence, Humana wanted to find new ways to improve provider access to information. By using SAS® Text Analytics, Humana was able to extend beyond the structured information and analyze the unstructured call center text notes to provide more convenience to providers while lowering internal operational costs.

Mining text for answers
Very little value was found in the structured data captured during an inquiry with a provider. The call center agent would select categories to classify the reason and details of the call; however, the categories selected were heavily weighted to just a few default selections. Eighty percent of all inquiries fell into the top two or three categories out of 18,000 possible combinations. Humana needed a more reliable method of understanding the inquiry and what information was helpful in resolving it. That information was buried in the call center agent's unstructured, freeform electronic notes.

Reading a million call notes a month was simply not scalable. And given the jargon and shorthand used by the call center agents, the Humana staff simply couldn't count words. "Since we didn't have any preconceived notions as to which, out of all the possible thousands of combinations of categories, were the most important to our providers, we needed a way to analyze the entire collection of call center notes and have the data tell us what was important,'' explained Greg Hayworth, the Humana scientist in charge of the project.

Finding a solution with SAS®
Humana chose SAS Text Miner to deliver a scalable and reliable process for classification of provider calls. "Using a pass-through query from SAS to our data warehouse, we didn't need to replicate information,'' Hayworth said. "We used the SAS Text Miner parsing node to automatically identify parts of speech. We also extracted entities such as phone numbers, addresses and titles."

Humana then used the SAS Text Miner filter node to eliminate ubiquitous terms and rarely used terms. "We generated synonym lists that could be stored and reused on similar projects," said Hayworth. The next step was to use SAS Text Miner's clustering algorithm to generate meaningful clusters of terms. "We could further refine as needed to customize to our needs. These clusters formed the specific categories of provider calls that we wanted to address."

Helping staff members understand the results
"We wanted to create specific reports to share with our internal audiences, so we imported the mined scores into SAS® Enterprise Guide®," said Hayworth. SAS Enterprise Guide is a point-and-click, menu- and wizard-driven tool enabling data analysis and results publishing. The results of the project: Humana determined nine meaningful categories that providers typically call about and added information to the provider Web pages to address many of the most common questions. The company also improved training for call center agents in these areas. Based upon the implementation of the analysis, Humana experienced more than a 10 percent reduction in calls per claim.

The results illustrated in this article are specific to the particular situations, business models, data input, and computing environments described herein. Each SAS customer’s experience is unique based on business and technical variables and all statements must be considered non-typical. Actual savings, results, and performance characteristics will vary depending on individual customer configurations and conditions. SAS does not guarantee or represent that every customer will achieve similar results. The only warranties for SAS products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with SAS as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of SAS software. Brand and product names are trademarks of their respective companies.

Copyright © SAS Institute Inc. All Rights Reserved.