TongSun

“Our predictive models go beyond traditional demographic attributes, and allow for deeper actionable insights.”– Tong Sun, director of Scalable Data Analytics Research Lab at PARC, A Xerox Company

By Tong Sun

One of the central ideas around our customer care platform  is the idea of customer-centric care across voice, chat and social media. This idea moves beyond solving an individual’s particular issue, allowing a brand to understand better who the customer is and increase the value of interaction.

During the past year, we’ve seen how brands use social media engagement and analytics to interact with their customers. Our researchers in the Scalable Data Analytics Laboratory used their expertise in text mining, machine learning, and predictive modeling to create a social media tool and platform that quickly and accurately extracts the sentiment of customer comments and will soon be able to review them in the context of ongoing conversations. The platform then assesses, prioritizes and delivers the insights to the best contact (a customer sales agent or a sales or marketing team), so they can address issues or opportunities.

The social media analytics platform has been incorporated into the offerings of WDS, A Xerox Company. Through this tool, WDS has been able to move beyond the market’s simple keyword-centric approach toward a much more granular level of accuracy where sentiments — such as sarcasm or abbreviated wording — can be accurately analyzed and routed.

Understanding the Customer

Joined by researchers from the Xerox Research Centre India and the Xerox Research Centre Europe, we now are working on ways companies can gain a deeper understanding of who their consumers are. We have developed inventive methods and tools to analyze user-generated content on social media channels to predict not only a person’s demographic attributes (such as age, gender, parental/marital status, education level and profession), but also their personality and life events that might trigger certain behaviors during their interactions with brands.

We’ve also worked with Xerox customers in the retail and financial service domains. We developed the concept of “customer journey” that illustrates the steps customers go through when engaging with a brand. Our predictive models go beyond traditional demographic attributes, and allow for deeper actionable insights that can be generated to strengthen the brand’s ability to communicate with customers. We are exploring how certain life events such as “graduating,” “getting married,” “expecting a baby,” or “getting a job,” can be used as triggers to purchase or upgrade a motor vehicle, for example.

Churn is another “event” in the customer journey. We’ve developed ways to predict the warning signs that indicate a customer might leave a brand or switch to a competitor. One type of sign might be frustration with a product, customer service or Web support.

Subscribe to Simplify Work and receive email updates when we publish a new article.

Billion Dollar Ideas is a series of three articles that let you in on the people and ideas behind Xerox R&D  labs around the globe. These, and other articles about Xerox innovation, are available in the Simplify Work Innovation category.