Analytics

Print
Customers are Knocking, but No One's Listening

1 Nov, 2007

By: Jeff Kaplan

We conducted an informal research study interviewing customer executives ranging from mid-size to Fortune 100 companies, and the analysis highlighted that although all customer service executives recognize the importance of reading and responding to customer inquiries, they don’t have the time to read through the tens of thousands of inbound e-mails or Web forum inquiries they receive each day.

What we also discovered was that the majority of these companies have customer service representatives who are only able to read on average 10 percent of all incoming service complaints. What’s astonishing is that the executives interviewed acknowledged their current customer inbound response and turnaround time was far from optimal, but had no immediate plans for change. What’s even more amazing is that there is technology available today that can automate and improve customer service response time and accuracy of resolution by leveraging text mining - the process of collecting text (referred to as “unstructured data”), extracting out key terms and concepts, appropriately routing the inquiry and summarizing the voice of customer.

Why is Text Mining Not Being Leveraged Everywhere?

Although over the past few years there has been a growing buzz surrounding text mining and how it can be applied to the staggering amounts of data generated from daily e-mails, service requests, blogs and customer forums, very few customer service departments have taken advantage of this technology to find out what their customers are trying to tell them.

To date, current processes of manually reading customer inquiries have become a futile effort. A study conducted by the School of Information Management at the University of California at Berkeley revealed that instant messaging generates around 5 billion messages a day, or 274 Terabytes of data a year, while e-mail adds another 400,000 terabytes annually. Merrill Lynch has reported that more than 85 percent of the information within an enterprise is unstructured and crucial customer service complaints are now extending beyond the corporate firewall when inquiries are not being responded to, as customers recognize their voices can be heard through peer-to-peer blogs.

Given the benefits that text mining solutions provide, such as automating the tagging of inquiries, pulling out key complaints and matching the appropriate correspondence back to the customer, why have customer service departments not embraced text mining solutions?

For starters, it is extremely challenging to find one vendor that offers an end-to-end solution. For example, most vendors tend to focus on niche areas of the text mining process, from term extraction to building domain-specific taxonomies to visualization of conceptual relationships in text.

Moreover, software ‘solutions’ that are marketed as licensed text mining suites have failed to deliver on scalability and require customers to have in-house text mining expertise. Along with unmet expectations, license price-points can vary by as much as 100 percent, contributing to a muddled market that makes potential customers leery and slows the emergence of new providers.

Criteria for Evaluating Text Mining Solutions

First and foremost, the highest rate of success is achieved when implementing text mining solutions that are delivered as a service. The main reason is that although text is very easy to manually read through one piece at a time, it does require several areas of expertise and technology to automate the process.

Text mining performed the correct way is a combination of three skills: information retrieval, linguistics and data mining. If you’re like most of the folks working in customer service and you’re fortunate enough to have these skills, you now need to also find additional time from your current day-to-day job. That’s why it makes so much sense and the ROI is much greater when you have experts providing text mining solutions as a service.

When evaluating a text mining solution, make sure it has the following functionality as displayed in Figure 1:

• It is delivered as a service from trained text mining professionals.

• Its taxonomies are domain specific – for example, an airline company would not want to use the same taxonomy that automotive manufacturers use.

• It has the ability to use clustering techniques to identify new trends, anomalies and correlations across your entire customer base.

• It scales to ensure that the solution can handle not only your volume today, but also support future growth.

• It’s automated to process customer inquiries as they come in and identify the appropriate routing and case-management assignment.

Steps to Ensure Your Customer’s Voice Stays Private

Before you embark on investing in a text mining solution to learn more about what your customers are saying, you need to first gain their trust by keeping their information secure. With the increased convenience of customer support services being offered through e-mail and Web sites, the amount of personal identifiable information (PII) being passed through the Internet is increasing at an exponential rate.

From the customers’ point of view, it’s more convenient to send a question about their credit card statement through a Web site’s “live chat” or e-mail a service complaint versus the alternative of picking up the phone and waiting to speak to a live representative.

The risk for this convenience is the requirement that you send personal identifiable information (PII) over the Internet, along with your inquiry or complaint, such as your credit card number, passwords, last activity, last purchase, etc. — all of which in the wrong hands could lead to fraudulent activity and identity theft.

There are a few recommended security procedures to help ensure your customer service messages stay secure. First, treat all customer service data as if it contains PII, because it will when your customers identify themselves and provide information to help you address their concerns. Store and manage customer service data with the same care and rigor you have in place for your customer billing database. Also, ensure there are proper internal network security, authentication and access controls to manage who can retrieve this data.

Second, protect your customers starting at the Web form they use to write in. Enable technologies like SSL, which is a mature industry standard for preventing unwanted snooping of Web pages as customer input travels back and forth between your Web server and the customer. Remember that e-mail outside your organization typically cannot be protected like this, so while it might be convenient for some tasks, it is not safe for messages that may contain PII. You may want to notify your customers about responses or updates to their concerns but direct them to your Web site for the details, where it can be shielded from prying eyes.

Third, reduce theft risk from employees, a common customer service center problem, by limiting employee access to customer service data to just the issues or messages they are working on. In addition, log employee access to customer service data and regularly audit that access to identify any red flag behavior. Finally, put processes in place to prevent large quantities of customer service data from reaching outside the data center, such as on employee computers, where it could be stolen. Try to avoid retaining any of this information on employee laptops that leave your business site and head out into the real world where your background checks and cardkeys have no effect.

Put it All Together

Everyone understands that it’s more expensive to acquire a new customer than keep an existing customer. On top of that, brand loyalty continues to suffer as customers today more than ever have increased purchasing options. Factor in that many customers feel like their suggestions or complaints are not being heard and you now have a recipe for customer retention problems. The good news is that text mining provides a way for companies to gain competitive advantage by being able to identify new service trends, address their customers’ inquiries in a timely manner and make their customers feel like their information is protected and business is valued.

Print