Clarabridge Spans Sentiment
October 24, 2010
My perception of Clarabridge is that the company was a more friendly version of MicroStrategy’s query tools. I guess I was incorrect. I read “Execuvue Software Tracks Guest Sentiment” and learned about my perceptual shortcomings. According to the write up, Apptech “is now leveraging text analytics to include guest sentiment tracking.” For me the key factoid in the write up was contained in this passage:
The Execuvue Sentiment Engine measures the two primary sources of guest satisfaction information: brand data, including email surveys and brand comment cards, and Web data from social media sites and hotel review sites. The system collects data from all sources, including written comments and, in conjunction with Clarabridge, uses a natural language processing program to extract written speech prompts and then converts them to measurable scores. Additionally, it provides drill down capabilities so operators can view individual guest comments that make up the sentiment metrics.
A quick trip to the Clarabridge Web site alerted me that “Clarabridge is the leading provider of text mining software used by many global 1000 companies to improve customer experience management.” When I last looked closely at Clarabridge, I watched as the firm demonstrated how its system could provide information about a retail chain’s inventory.
Like other content processing companies, Clarabridge has verticalized; that is, instead of pitching a one size fits all platform or framework, the company is focusing on solving a problem. Like Lexalytics and Attensity, that problem is figuring out if content expresses happy or sad thoughts. Masssaged with some numerical sage and basil, a licensee will be able to head off trouble or ride a surge of interest quickly.
A Clarabridge document identifies these features of its system:
- Automatic linguistic “reading” of text and ad-hoc searching and filtration;
- Categorization of the text at detailed sub-document, sentence, and clause levels;
- Identification of varying levels of positive and negative sentiments and what they relate to;
- Analysis of root cause, emerging issues, and trends, and;
- Capability to drill down to the original text to understand any areas of interest.
The pricing model is interesting because it appears to be a per document charge. Clarabridge refers to a “verbatim” which is unclear to me. Each processed verbatim rings the cash register at 15 cents. Taxi meter pricing is a valid method, but the fees can escalate unless close attention is paid to the document processing flow.
Stephen E Arnold, October 24, 2010
Freebie
Comments
-
- Subscribe to Beyond Search
Feature archive
News archive
-
