The first public opinion poll probably was taken shortly after the Neanderthals learned about fire: “Thag like wooly mammoth burned or raw?”
Throughout history, humans have attempted to aggregate the opinions of many individuals in order to gain insight into social, political or economic issues. Since the early 1900s, the business world has formalized such efforts into market research techniques for gathering information about brands, markets and customers. However, the traditional tools for conducting such research are no longer adequate in a Web 2.0 world.
With email, blogs, text messages, peer-to-peer networks and various social media platforms, consumers have unprecedented avenues for communicating, interacting and sharing their experiences and opinions — positive or negative — regarding any product or service. This activity is generating text-heavy unstructured data at a rate so accelerated that it overwhelms attempts to manually monitor all the information.
To stay abreast of what is being said about their brands across these channels, more organizations are implementing text analytics, one of the fastest-growing areas in the business intelligence (BI) realm. Also known as text mining or sentiment analysis, these applications use various mathematical, statistical, linguistic and pattern-recognition techniques to allow automatic analysis of unstructured information as well as the extraction of high-quality, actionable data.
Analysts say adoption of text analytics technology and applications has begun in earnest, and is expected to be rapid over the next few years as both vendors and enterprises invest in addressing a variety of business opportunities. Software and service revenues for text analytics now total $835 million globally with anticipated annual market growth rates of 25 percent to 40 percent expected, according to Seth Grimes, an analytics strategist with Washington D.C. consultancy Alta Plana Corporation.
“Text analytics helps forward-looking organizations gain new insights into customers’ perceptions, motivations and plans — whether the goal is to boost customer satisfaction, product quality and sales or to reduce churn,” said Grimes.
Text analytics has been a recognized IT sector for more than 15 years, yet it remains one of technology’s best-kept secrets. Long utilized in the publishing, research and scientific communities to structure and extract meaning from both long- and short-form communications, text analytics is capturing the attention of enterprises and contact centers. Although the market is fragmented, the prevalence of social media is driving increased enterprise interest.
Organizations are utilizing text analytics for search and retrieval, multichannel analytics, document tagging and publishing, market research, threat detection and reputation management. E-discovery and regulatory compliance are two areas in which there is expected to be extensive application of text analytics in the near future. Associated with these are the areas of bankruptcy settlements, due-diligence processes, and the handling of data rooms during a takeover or a merger.
The range of uses continues to expand as companies invest in text analytics to structure their social media communications and as vendors invest to enhance their applications for usability, support and robustness. Industries at the forefront of text analytics usage include automotive, education, e-commerce, financial services, government, high technology, insurance, retail, social media and telecom.
“Any organization contemplating or currently implementing a social media strategy should be thinking about how text analytics can help achieve their goals,” said Donna Fluss, president of DMG Consulting, a strategic advisory firm specializing in contact centers. “There is an enormous opportunity for companies to differentiate themselves by utilizing text analytics or text analytics-enabled applications and sentiment analysis to capture and structure data on what is being said about them, over what channel, when and by whom.”
Text analysis differs from traditional search tools that require a user to know what he or she is looking for. Text analysis attempts to discover information in a pattern that is not known beforehand, through the use of advanced techniques such as pattern recognition, natural language processing and machine learning. By focusing on patterns and characteristics, text analysis can produce better search results and deeper data analysis, thereby providing quick retrieval of information that otherwise would remain hidden.
“The business insights that can be mined from customer conversations are every bit as valuable as financial data to large corporate brands,” said Ian Bonner, CEO of text analytics vendor Attensity. “Making sure that data is both highly accurate and easily accessible is a top priority for any organization that wants to leverage customer conversations as a business asset.”
While current solutions tend to focus on the English language, some text analytics applications can handle other languages. Due to large investments by the U.S. government, languages such as Arabic, Farsi, Urdu, Somali, Chinese and Russian are well covered. Vendors with the most mature solution sets include Attensity, Basis Technology, Clarabridge, Lexalytics, SAS, Temis and ZyLAB.
“The global nature of today’s consumer is forcing companies to think more broadly about a universal view of their customer, looking across cultures, time zones and media,” said Darren Jaffrey of Clarabridge, which recently upgraded its product to include French and Portuguese natural language programming.
It might have been fairly easy to establish a consensus of opinion in Stone-Age society, but modern man often must sift through an overwhelming amount of information to gain accurate insight. Text analytics helps by processing massive amounts of unstructured information and presenting it as structured data that can be more easily analyzed and understood.
“Text analysis can help organizations better understand their communities of customers, fans, advocates and colleagues by surfacing commonly used phrases,” said Cliff Figallo, editor and moderator of Social Media Today. “Revealing the juxtaposition of key terms across hundreds or thousands of posts and conversations would reveal deeper levels of shared experience and sentiment. It would also bring more understanding of disagreement and conflict within communities, telling organizations how to better address and serve people with varied attitudes toward an organizations products and services.”
