In today's social media whirlwind of a world, the fact that customers still pick up the phone to call companies deserves attention. Each call is a one-on-one conversation with happy customers, unhappy customers and prospects seeking information. These incoming conversations are an often overlooked, but easily accessible, and some might argue supremely valuable, source of consumer insight.
Contact centers contain rich customer feedback, in agent notes, captured emails and online chats. From this data, a company can glean product or service issues, enhancement requests and even competitive information. Notes accompanying calls to a marketing or sales hotline provide insights on campaign effectiveness or messaging, and calls into consumer affairs hotlines help provide an early warning to quality, health, or safety issues.
There are several advantages to mining contact center content.
- The data is proprietary, which offers some degree of market advantage over public data that is available to all viewers.
- The feedback is more likely to come from true customers or prospects, making it highly relevant.
- The CRM record captures associated information about the caller and products/services, which makes correlating feedback to specific products, services, or customer types easier. The CRM systems can capture the customer's location, specific product or service experiences, even social/demographic insights about the customer (loyalty, gender, age group).
With so much benefit within contact center feedback, why aren't more companies utilizing this valuable source of insight? Historically, the agent only read qualitative feedback during the call. While some companies attempt to read the feedback, extrapolating insights from the thousands of calls that occur each day in a contact center is a time-consuming, expensive task.
Text Mining Creates Efficiencies
More recently, however, text mining / text analytics technologies permit the automation of processing feedback from contact centers for sentiment extraction and issue classification. In the last two years, leading retail, telecom, travel, entertainment, and financial services companies have adopted text analytics solutions to more fully analyze contact center data and generate insights and returns on this investment that create true competitive advantage.
How it works
Customer feedback in the contact center, along with structured data associated with the call (customer type, product, location detail, etc.) is run through mining and transformation algorithms that extract product issues, service issues, customer sentiment, and root cause determination of issues. Using techniques including Natural Language Processing (NLP), machine learning/classification, and "sentiment extraction," text and data are effectively translated into quantitative insight and scored, correlated, and presented to business analysts for review in reports, dashboards, and time-critical alerts.
Text mining effectively automates what would otherwise be a manual process, so an analyst, instead of spending hours and days reading individual call records, can track issues across 1000's of calls at a time by examining reports and alerts that distill the feedback into actionable insights.
- A global retailer uses text analytics to mine their customer affairs hotline to track perceptions of private label products vs. brand name products, product safety and quality issues, and customer perceptions of food product quality, health, and nutritional value.
- A leading fast-food company tracks their customer support hotline and uses text mining to identify causes customer satisfaction, employee service issues, and promotion effectiveness/feedback on new products launches.
- An international hotel chain extracts specific causes of negative customer experiences at underperforming properties with text mining, and then uses a performance management process to manage the underperforming properties and make concrete performance improvements.
Real World ROI:
- Cost Avoidance: Technology automates a manual, time-consuming process, so many companies find they spend less on market research and customer support than they would otherwise, and the savings accrue in months, not years.
- Continuous Improvement: Turning the qualitative (text-based feedback) into quantitative (performance scores, sentiment scores) helps create measures for performance tracking and improvement. These measurements allow an enterprise to calculate and correlate the financial impact of customer experience improvements on customer profitability, loyalty, and willingness to promote a company's products or service.
- Creating Strategic Value: Customer feedback is a gift, and insights derived from text analytics often find hidden insights otherwise missed -- ideas for product improvement, early identification of a significant problem before it escalates to a support burden, or the insight to make a strategic investment based on customer desires or interests.
Your contact center has a truly strategic asset -- your customer's needs, suggestions, and passions. Tap into it.
About the Author
Sid Banerjee (firstname.lastname@example.org) is the cofounder and chief executive officer of Clarabridge, a leading provider of text mining software used by many Global 1000 companies to improve customer experience management. Banerjee is a highly respected thought leader within the text mining community, with more than 15 years of business intelligence leadership experience.
Please note that the Viewpoints listed in CRM magazine and appearing on destinationCRM.com represent the perspective of the authors, and not necessarily those of the magazine or its editors.
You may leave a public comment regarding this article by clicking on "Comments" at the top.
If you would like to submit a Viewpoint for consideration on a topic related to customer relationship management, please email viewpoints@destinationCRM.com.
For the rest of the March 2010 issue of CRM magazine please click here.