Text mining taps into customer sentiment

In tough economic times when vendors are faced with scaling back offerings to cut costs, an analyst with Forrester Research Inc. said text mining of unstructured data sources can reap valuable customer sentiment before drastic decisions are made.

The more structured approach of multiple-choice questionnaires can no longer suffice as they “limit the depth of insight and breadth of customer feedback,” said Bruce Temkin, vice-president and principal analyst with Cambridge, Mass.-based Forrester Research.

Text mining, on the other hand, is by no means a novel technology, but vendors are increasingly making it accessible for applications like Voice of the Customer (VoC). Unstructured data channels – in-bound e-mail, call centre conversations, blogs, SMS messages – by virtue of being text-oriented are prime targets for mining, he said, as are social media.

In a research paper entitled Voice of the Customer: The Next Generation, Temkin examines how a company can reshape its approach to gathering customer sentiment, using technologies including text mining.

As the recession has deepened, said Temkin, the correlation between customer experience and loyalty has grown, and it’s now ever more vital for a business to hear and understand its customers “because you have to make decisions about trade-offs in your company that are more painful and more far-reaching than you would if it were not a recession.”

Complicating VoC efforts is the fact that most businesses don’t even realize issues abound, given the degree to which efforts are scattered, said Temkin. “It’s not until you start looking across the entire organization at an enterprise level that you find these problems,” he said.

Awareness of text mining has experienced an uptick in the past 12 to 18 months, said Sid Banerjee, CEO with Reston, Va.-based text mining software provider Clarabridge Inc. A greater number of online surveys take place today than a decade ago, he said, yet with this plethora of digital information, “it’s hard to get through it and analyze it if you are a big company.”

With text mining, Banerjee said unstructured data sources can be mined and distilled into sentences and paragraphs about “what a person is talking about, to what degree of sentiment they are talking about that issue … and specifically how is that topic evolving over time.”

Banerjee outlined a hierarchy of content sources to which he recommends businesses apply text mining. He places the highest value on surveys because of the association with a specific transaction or event, like a shopping experience. Second, is call centre data because customers voluntarily call in to resolve issues. And online forums and news groups where like-minded individuals gather, like TripAdvisor, are also valuable.

Temkin said data sources are mostly “trending towards digital” and companies are increasingly making use of unstructured data, but that said, those companies that are aggressively exploring those channels are few and far between. While in the past, businesses would have to push customers for feedback in a particular format, he said, “the future is clearly about analyzing feedback in any form that your customers give it. That’s a trend that won’t go away.”

Related Download
3 reasons why Hyperconverged is the cost-efficient, simplified infrastructure for the modern data center Sponsor: Lenovo
3 reasons why Hyperconverged is the cost-efficient, simplified infrastructure for the modern data center
Find out how Hyperconverged systems can help you meet the challenges of the modern IT department. Click here to find out more.
Register Now