Mining for the Gold Within Text

Text analytics is slowly making its impact felt as companies that have established data warehouses are finally taking the next steps required to achieve richer, more granular customer insight. The technology now exists to enable companies to extract valuable information from material that was once abandoned in the black hole of business: not only internal data such as customer emails, contact center notes, online chats, and survey responses, but external data as well, including blogs, customer-discussion boards, and online forums. Recognizing the importance of this functionality, text analytics solution provider Attensity yesterday unveiled its on-demand Voice of the Customer (VoC) analysis software. Attensity, which likes to refer to itself as the "business intelligence for unstructured data," has a patented Exhaustive Extraction engine that does not require the client to write any rules to define what he's looking for. "You don't always know what customers are going to complain about," says Michelle de Haaff, the company's vice president of products and marketing. "[You] don't know what will be a problem." Companies that are focused on the impossible -- trying to know what they don't know -- often miss out on the granularity, she adds. With Exhaustive Extraction, facts about customer behaviors, activities, and events are automatically pulled out of unstructured data without any predefinition, she explains. That, combined with text analysis, allows companies to recognize trends, make connections and queries, and ultimately take corrective or -- even better -- preventive action. In general, Attensity organizes data along the following lines:

• Facts are extracted and mapped onto relational tables;
• reports and dashboards allow easy navigation of top issues;
• anomaly reports reveal issues that have developed since the last time instance, which may have been just a few hours ago; and
• alerts are delivered to bring attention to concerns that fall outside of what's considered to be a normal occurrence.

According to research conducted by the consulting firm Hurwitz & Associates, 70 percent of text-mining users ranked "voice of the customer solutions" as their top text-mining application. Despite the strenuous processes involved in companies' text-mining initiatives, the study showed that users found tremendous value in the software and reported an unwillingness to ever do without it again, according to Fern Halper, a partner at Hurwitz & Associates. Text analytics technology, he says, "is critical, given the amount of unstructured information that's out there," Halper says. Attensity isn't the only vendor offering this capability, or even the only one doing it on-demand. As more vendors enter this space -- making the technology increasingly user-friendly -- more users will undoubtedly begin to see the value as well. However, the problem isn't that companies don't understand the value, or that they don't know that there are solutions available -- it's that they're so tangled up in structured data that they can't even begin tackling unstructured. "They don't have the budget or the IT resources to deal with more," Halper says. "It's unfortunate because there's so much insight to be gained." Halper suggests other hurdles companies face when exploring the use of text analytics:

• processing, cleaning, and accessing data;
• developing rules (a burden that Attensity claims is bypassed with its technology); and
• overcoming limited out-of-the-box functionality.

But even in just the six months following her firm's study, Halper has already seen significant improvements in the software products on the market. For those who have adopted text-mining technology, she says, "the value far outweighs any obstacles." If it's not already at the top of the company to-do list, companies are going to have to face this imperative sooner or later, she adds. It's not an easy task, but technology is more than up to the task. Whether the information is within your corporate walls or out in the infinite expanse of the Internet, Halper says, "you're going to end up with unstructured data no matter what."

Related articles: Feature: Too Much Pork for Just One Fork Shiny, clean data and solid lead qualification will help satisfy all by closing the nutritional sales-and-marketing info loop. With Clementine 12.0, SPSS Goes Deep New data mining release offers improved functionality for a broader base of users. Real ROI: Better Info Leads to Better Campaigns SPSS Software provides quick analysis and answers. Feature: 12 Tips for Generating Rich Data Here, a guide to uncovering the bounty buried in your data warehouse. Feature: 7 Strategies for Profiting From Customer Data With every contact center interaction your company gleans valuable customer information. Here are the best ways to turn that data into revenue. Uncovering Business Intelligence: Outlines for Text Mining New tools help to unite unstructured data with what already is known to glean more complete views of customers. Viewpoint: Listening to the Voices of Customers Text mining helps companies discern true customer sentiment. Viewpoint: Hear Customers Knocking? Time to listen -- text mining needs greater adoption by customer service departments. Viewpoint: Creating True Communication With Customers Relationships, even eCRM relationships, are a two-way street.

Related Articles

Attensity Buys Biz360 for Social Media Monitoring

With a goal to "open up the enterprise," the semantic analytics company purchases the global listening solution provider for an undisclosed sum.

Attensity Group to Tackle Unstructured Data

The three-company roll-up aims to enhance customer service by using voice-of-the-customer feedback.

Text Links May Make the Best Marketing

Overall, 25 percent of online visitors will click a text link first; younger consumers, however, may be more graphically inclined.

Roar of the Market

Attensity's new Market Voice solution strives to harvest customer insight from both direct and indirect channels.