Text Mining Analytics
How to Use Voice Mining to Tune In to the Voice of the Customer
Voice mining technology digs into mounds of stored but often unused data, giving you ears to hear your customers. Your customers are telling you and your competitors what they want to buy and what prompted them to choose you. But manually rereading call center notes or replaying phone conversations can't possibly deliver customer intelligence in a timely fashion, so valuable information remains stuck and inaccessible.
Empowered with analytics, though, you can gain the advantage of being able to spot trends in everyday conversation threads to predict and respond to market opportunities and pitfalls. Four components are required for organizations that wish to take full advantage of voice mining technology.
Three of the four components are commercially available: audio voice-to-text convertors, text mining analytics, and business intelligence reporting and performance dashboards. The other component is the person who drives this system. Adding even a single component is noteworthy, but those organizations who implement all four in harmony will reap the most from voice mining technology.
Component No. 1: Audio voice-to-text converters
Before conversational audio data can be processed by a computer, it must be translated into electronic format. The "analysis" involved at this step is performed by a phonetic index search that automatically transforms a captured audio signal into a sequence of phonemes.
Many voice transcription systems today are supplemented with spell-checkers because words can be taken directly from a dictionary. What makes the spoken text harder to understand is the lower accuracy of the words chosen by the transcription algorithms when it comes to deciphering every "um" or "ah" sound.
Even with great advances over the past five years, there are times when these translations are impossible to read. Because the reason you are trying to "read" these spoken words is to understand and comprehend what is being said, you can get much further by taking this data and performing pattern detection and concept-clustering techniques. These are available in the second component of voice mining: text mining analytics.
Text Mining Analytics
Component No. 2: Text mining analytics
Text mining analytics technology is a result of advances in natural linguistic processing and the pervasive adoption of data mining on traditional structured data. Text mining software finds explicit relationships and discovers associations between documents.
Unstructured, free-form textual content can now be clustered by grouping the data into categories using a variety of semantic tools. These knowledge-extraction technologies can handle a huge variety of document formats in dozens of languages, and they run on sophisticated, text-parsing capabilities to transform data into a compact, information-rich structure for interactive exploration of concepts and relationships between terms and documents.
Organizations that already have predictive analytical models for their customer relationship management initiatives are finding additional gains when they add text analytics to integrate the unstructured textual data with structured data. Leading organizations are implementing these emerging technologies to find the value hidden in document collections. Those with a unified business analytics framework see improvement in the overall accuracy of their predictive models.
Component No. 3: Voice mining your own business
Sometimes, when we are faced with urgent demands, the last thing we want to do is consider investing in analytical technologies. Unfortunately, if there is no quick manner to communicate the expected return on investment or customer intelligence insights to your superiors-with the authority to act and reap the gains from those analytical insights-the door will remain closed.
The best way to overcome this obstacle is to secure access to analytic experts at the same time you address any voice mining software purchases. A trained analytical expert will ensure you not only "see" insights, but actually move on them and get the value out of predictive analytic workbenches.
One way to efficiently and effectively transfer the value is by customizing a performance dashboard or tailoring a user interface that decision makers are already familiar with. Only then will your voice mining implementation actually hear and comprehend in real time what your customers are saying.
BI Reporting and Performance Dashboards
Component No. 4: BI reporting and performance dashboards
Analyzing audio data to categorize customer segments and predict customer satisfaction levels is a multistep process-and one that our own human brains struggle to do in real time. An accurate predictive model, built from text mining and data mining algorithms, starts out by summarizing what the caller is saying now, and then goes a step further to inform you of what the caller is likely to say next.
With a BI dashboard, you can alert your staff to the recommended responses and actions to take. For example, let's say a delinquent customer calls in with a complaint. If this customer goes away, you may avoid future losses. However, if you have a very important caller (for example, one that your predictive model has identified as being a big spender or a representative of a large segment of customers), then your automated BI dashboard may prompt you to offer a special promotion or incentive to the caller-on the spot, in real time.
Voice mining technology is ready for prime time. The past decade has witnessed advances in each component. It's time to remove the "emerging" label and replace it with a market-proven tag. Not long from now, voice mining will be a "best practice" for successful organizations across the globe.
This best practice entails applying all four components in harmony. Voices are telling you how to increase satisfaction, build loyalty, reduce churn and make safer products. Technology is automating the process of giving ear to the voices that have long been muffled in mounds of data. When that information comes to life, true competitive advantage can follow.
Mary Grace Crissey is the Analytics Marketing Manager at SAS, where she applies her Master's of Science degree in Operations Research to show how text analytics, optimization and data mining can solve real world challenges. After 20 years as a military scientific officer, she continues to serve in leadership roles with professional societies including Knowledge Discovery and Data Mining (ACM/KDD) and the Institute of Operations Research and the Management Sciences (INFORMS). She can be reached at MaryGrace.Crissey@sas.com.