Monday, March 14, 2011

Text Analytics Put to Use

  Text Analytics is the newest way of analyzing all digital data and compiling the information based on parameters set by the user. Text analytics works like data mining. The software can be programmed to search out key words, phrases, etc. and compile all the information that comes with each of those tagged words. Experts in this industry can then go in and create reports and recommendations based on the reports to the clients. Text Analytics was very helpful for JetBlue in 2007 when they left thousands of people stranded on JFK runways. They received 15,000 emails the following day (almost quadruple what they normally see). They needed to respond quickly but they had an information overload. Thankfully, they had begun using text analytics before this event and were able to use the technology to understand the main problems and were able to quickly solve the problems. Additionally with text analytics companies are able to combine structured and non structured data together and understand situations even better. For example, JetBlue can combine structured info like flight times, fares, etc with the results of surveys and interviews with their customers to get a better understanding of their consumers and economic influences. Text analytics is a powerful tool that will become a major measurement of success/ need for improvement for all companies. 
To read more about JetBlue's Text Analytics, check out this link


Currently, customers have hired consultants to interpret the data in ways the client can understand. I recently read an interview of Tom Anderson, a consultant and an expert in Text Analytics who says, within the next five years companies that specialize in text analytics consulting will not have an advantage in the field because other market research firms will adapt the softwares and be able to do the same thing as part of their packages to their clients.(read full interview here) I see Text Analytics becoming like Web Analytics; where individuals can set up the software and have the education and information to be able to interpret their own data. When this will happen I do not know, but I think it will end up that way. 

No comments:

Post a Comment