Analytics: Text mining for customer insight

March 12th, 2010

By Sid Banerjee, CEO, Clarabridge, Inc.

Sid Banerjee, CEO, Clarabridge

RESTON, VA – In just a few years, text mining has evolved away from the domain of programmers and computational linguists to becoming a mainstream business solution for listening to, analyzing, and acting on customer experiences and feedback.

Customers are more verbose than ever before, and they leave text-based feedback in more and more places, creating opportunities for companies to better understand their customers’ expectations, experiences, and needs.

Today’s text mining solutions collect, categorize, and calculate sentiment– letting companies better understand the feelings, emotions, and needs of millions of customers as they evaluate, purchase, and use products and services.

Consider the typical customer

Consider the typical customer looking for a product or service. She may research a company or product online, read review sites, company and competitor descriptions, and peruse social media sites like Facebook and Twitter.

After purchase, she may fill out a survey to describe the sales experience. If she has questions about a product or service, she may call a 1-800 number, email or chat with a representative, or even use Twitter to convey frustration or satisfaction with the experience. And over a lifetime of customer experiences, she may buy more, upgrade, downgrade, or simply shift her loyalties – and each experience will generate text in one or more “listening posts” inside a company and on the internet.

Feedback a “gift”

Each piece of feedback from a customer is a “gift” – information that can be analyzed, trended, monitored, and used to improve product and service experiences. Leading companies are learning a lot by applying text mining to customer experience feedback.

A bank, after acquiring a regional competitor, is using text mining to track survey and call center feedback to ensure the migration experience doesn’t negatively affect customers and give them reason to defect.
A leading electronics retailer is tracking customer support inquiries to track the quality and ease of use of new HD TVs, computers, and consumer electronics, and using text mining to rate and reward the vendors whose products are easy to use and preferred by consumers.
A leading quick serve food retailer is using text mining to analyze customer surveys and customer hotline calls to monitor food quality, service quality, and collect feedback on marketing campaigns.

Companies use text mining because it helps generate a business return on investment. Without text mining, companies must manually read, categorize, and analyze feedback, and the manual process is nearly always more expensive and more time consuming than the more automated text mining solution.

Most importantly, text mining helps identify and create experiences that produce satisfied customers. Satisfied customers are more loyal, buy more, and recommend products and services – and ultimately the return on creating more satisfied customers is a more successful, profitable, and viable business.

About Sid Banerjee

Sid Banerjee is the co-founder and chief executive officer of Clarabridge, the leading provider of text mining software used by many Global 1000 companies to improve customer experience management (CEM). Sid has amassed more than 15 years of business intelligence leadership experience and is a highly respected thought leader within the text mining community. Clarabridge is privately held with headquarters in Reston, VA. For more information see the company’s Web site: Clarabridge.

Or visit the Clarabridge blog.

This guest blog post is part of our ongoing focus this month on Analytics. Next month we’ll be looking at emerging markets and social networking. If you have an idea for a guest blog on either topic, contact TechJournal South editor Allan Maurer: Allan at TechJournalSouth dot com.

Join hundreds of Marketing Professionals and Internet Execs at Atlanta’s premier Digital event – Digital Summit 2012
www.digitalsummit.com