Text Analytics: Knowing Customer Sentiment Has Bottom Line Benefits

Every major financial institution records its inbound calls, but few actively mine the data for insights into their customers. As much as 80% of customer data available to financial institutions (FI) currently goes unexamined, according to Gartner Research. Yet, information gleaned from customer data can help reduce attrition, operate call centers more efficiently and cross-sell with greater success.

Recent academic papers across different industries indicate that text analytics (TA) can help increase retention and renewal. Exactly how much improved customer retention adds to the bottom line will vary by company. But in today’s competitive marketplace, no one disputes that it is cheaper to stem the outflow of existing customers than to acquire new ones.

The key to unlocking customer insights is text analytics, a data mining technique that examines text communications or transcripts of call center conversations as a person would. A new generation of natural language processing software allows FIs to analyze and quantify tremendous masses of data quickly. First, by identifying recurring subjects, a topic analysis can prevent problems proactively. For example, if customers are calling frequently with billing questions, a statement redesign could be the solution.

Second, TA can determine customer sentiment through analysis of the emotional content of customer communications. By comparing attitudes across its customer base, an FI can respond to individual clients more effectively, while marketing to different segments more precisely.

[For more on how capital markets organizations are using analytics and big data to improve customer service and uncover market opportunities, read: Demand for Deep Analytics Challenges Data Managers.]

Two factors could propel the growth of TA and linguistic software in customer-related areas. First, FIs seek to manage call centers more efficiently and drive revenue through more effective cross-selling. Second, regulatory requirements of the Consumer Financial Protection Bureau (CFPB) have created additional expenses around consumer complaints.

TA technology is mature and is already being used in financial services, particularly in mitigating compliance risk. Some FIs use customized TA to identify rogue trading activity or dishonest brokers who reveal nonpublic information. Customer-related uses have the potential for greater long-term benefits, especially in sales, marketing, customer service and online experience.

Redefining the Customer Experience
Text analytics technology could revolutionize customer experience strategies by helping to support a three-step process that analyzes customer interactions, highlights problem areas and even suggests possible solutions.

1. Textual analysis: Analyze the content of customer interactions -- call center transcripts, surveys, emails and tweets -- to produce a deep, quantitative understanding of customer attitudes and dissatisfaction. One research study showed that customers who used highly emotional words -- positive and negative -- tended to be the most loyal.

2. Data integration: Combine the analysis of customer interactions with existing predictive models based on banking and credit history datasets to determine the impact on product usage, account closings and other key measures. Without TA, executives miss the huge pool of comments and complaints describing customer attitudes about the company, its products and employees.

3. Real-time approaches: The final step is to use these customer analyses to develop more refined customer experience strategies. By combining text analysis with voice recognition tools, sentiment analysis can occur while the customer is on the telephone. When run through a decision engine, it’s possible to deliver customized sales and service scripts to the customer service representative (CSR) that are adjusted to the customer’s emotional state -- all in real time.

Realizing TA’s promise
In many cases, CSRs need help uncovering needs and selling a large selection of increasingly complex and regulated financial products. With TA technology fully in place, the CSR’s first response might be acknowledgment of a problem, rather than cross-selling. A customer calling about a replacement debit card is not ready to hear a pitch for a new mortgage product.

Likewise, a caller who is upset about service and billing problems is probably not willing to be up-sold whole life insurance.

Text analysis allows financial services providers to move beyond the mere pushing of products to immediately taking the action that is best for the customer. TA software can make financial advisors, brokers or CSRs more efficient by suggesting relationship-building actions, instead of standard cross-selling messages that stand little chance of success. Greater call center efficiencies carry implications for more fair incentives based on success rates with qualified and receptive prospects.

Beyond the basic CSR-customer interface, text analytics enables financial firms to refine their customer experience strategies. For example, quantitative understanding of customers permits differentiated service for micro-segments of the customer base. Marketing chiefs can better recognize how more profitable customers differ from less profitable ones, and seek greater revenue by delivering extra value more efficiently.

Managing Expectations
There are two caveats, however. First, TA software is not a solution by itself. As with most technologies, intelligent and creative use of the new tools is the key to successful business results. When skilled linguists adapt TA software to create a company-specific taxonomy, these tools become win-win for FIs and customers.

Second, personal privacy may become a hot issue as linguistic technology spreads. On the regulatory front, the CFPB could pay more attention to the issue of customer privacy. Therefore, it is vital that FIs properly manage every customer’s digital identity, through clear policies, in order to prevent unwanted push back from consumers and regulators.

As financial institutions embrace the era of big data, text analytics can be an important element of successful customer relationships. Through quantifiable information about customer attitudes and areas of concern, FIs can address emerging trouble spots more effectively. Contact centers can be organized more efficiently to deliver real-time solutions. Longer-term, text analytics allows innovative companies to develop and refine customer experience strategies that will drive revenue growth and advance larger business goals.

About The Authors: Heidi Boyle and Bernhard Klein Wassink are Principals and Avril Castagnetta is a Senior Manager in the Financial Services Office, Customer Advisory practice at Ernst & Young LLP in Chicago and New York.