Users such as marketers, lawyers and financial traders are already tapping into content analytics tools for a variety of applications. The text analytics and content analytics tools, which mine documents, emails and other unstructured data to discern patterns and uncover insights, are designed to help companies better understand their customers, track market trends and avoid legal headaches.
Of the current users, marketing departments are probably getting the most out of content analytics today, said Seth Grimes, founder of Alta Plana Corp., a Takoma Park, Md.-based consulting firm. He added that marketing uses include determining how customers are using products, understanding long- and short-term trends around brand perception and identifying disgruntled customers via sentiment analytics techniques.
On a macro level, social media analytics tools can seek out and process large volumes of tweets and other content, such as Facebook and blog posts. From there, the text is analyzed for keywords, word frequency and wording patterns, with the findings often being presented to users in the form of data visualizations. Marketers can see, for example, what products are "trending" well and generating the most chatter on social networks. Or they can monitor customer sentiment and feedback on a new product launch.
The micro side of sentiment analytics
In addition, marketing users can zero in on specific customers for sentiment analytics purposes. Instead of analyzing large volumes of content to identify trends, they use text analytics tools to find individual tweets and online posts – typically, ones that express negative sentiment about a product or corporate brand. Marketing analysts can drill into the content to see why customers are unhappy and then forge a plan to remedy the situation.
Both types of text-based marketing analytics aim to reduce customer churn and enable organizations to respond to customer problems and other market developments before they become full-blown emergencies, said Gareth Herschel, an analyst at Stamford, Conn.-based Gartner Inc.
Another area in which content and text analytics tools are having a big impact is e-discovery, the process of locating and making available all of the documents related to a legal case after an organization is sued or otherwise becomes involved in litigation.
Complying with discovery requests used to mean an army of lawyers locking themselves in a conference room for days while they manually reviewed boxes of documents and email printouts, said Nick Patience, research director for information management at The 451 Group, a consulting firm based in New York. Under such circumstances, legal discovery often was – and still is – labor-intensive and mistake-prone, he added.
Content analytics is changing that scenario, though. Now when a company needs to turn over documents as part of a court case, enterprise search technology can identify emails, contracts and other records by keywords; text analytics software then can narrow the scope even further by analyzing the documents and determining which are in fact relevant to the case and which are not.
An open-and-shut case for content analytics?
To meet the requirements of e-discovery orders, "you have to have more than just a search," Grimes said. "You have to be able to perform analytics on the documents in order to comply with the mandates. And text analytics is really very essential to that."
In some cases, lawyers still need to vet the documents that enterprise search and text analytics tools pinpoint for e-discovery purposes, according to Sue Feldman, an analyst with Framingham, Mass.-based IDC. But using the technology can be a good first step that whittles down the flagged content to a manageable level, she said.
Another current use case for unstructured data analytics, though less widespread than the marketing and e-discovery ones, involves using the technology to make sense of financial market data and news developments in near real time in an effort to help financial traders make faster and better investment decisions.
For example, an explosion at an oil refinery in Venezuela could have a significant impact on the price of oil. But the first report of the explosion might come from a small news outlet in South America or even a local blogger – sources that most commodities traders wouldn't normally turn to for news. Text analytics technology can comb through vast amounts of unstructured content – press releases, SEC filings and even news stories on oil refineries in South America – to alert traders to important events that can impact when they buy or sell.
The sooner such information is understood, the quicker decisions can be made, said Feldman, who added that such monitoring "is something you can't do with any other tool."
ABOUT THE AUTHOR
Jeff Kelly is a freelance writer.