VerbaSense,  Market Research Edition.  ( for FREE trials email us at info@exasense.com ) 

VerbaSense, Survey Research Edition is a unique solution that automates coding of open-ended verbatim responses in surveys or other systematically collected content collections.   Accurate classification and coding of responses to open-ended survey questions have been an age long challenge in the market research and content coding community.  Many traditional research organizations even went as far as minimizing the number of open-ended questions they ask in market research opinion surveys because compared to closed-ended questions they are more costly to deal with.   As we look at the evolution of the internet and emergence of the World Wide Web as another significant mainstream written communication platform we can clearly see that content summarization and enumeration will continue to gain even more significance from both primary and secondary research perspectives.

On another front, consumer-generated-media and/or word-of-mouth communication platforms such as blogs, newsgroup sites, message boards, product or service review sites are already attracting great deal of interest by everyone.   Corporate Communications units, Advertising and Public Relations companies are scrambling to look for ways to leverage this newly found medium because there is no substitute to direct, candid and unfiltered consumer feedback.  It is a proven fact that well crafted open-ended survey questions in web surveys can yield richer and deeper insights into consumer's decision making calculus because respondents have total command of their articulations in an unfiltered way.    

VerbaSense currently is a service we offer for coding verbatim responses gathered by open-ended questions in market research surveys (comin soon: an ASP solution and can be utilized as fully-automated or semi-automated content coding).

VerbaSense key benefits:

• Revolutionary solution:  Unlike other automated coding technologies, VerbaSense is very simple to use and does NOT require keyword or pattern recognition training.
• Code in context:  Unlike other coding systems that typically rely on literal keyword or phrase based coding (along with complex thesaurus integration for some), VerbaSense is capable of context sensitive coding right out of the box.
• Self-Learning:  VerbaSense technology is intelligent and self-learning - the more you code the more accurate it gets.
• Significant labor savings: VerbaSense solution will significantly reduce your coding labor requirements.   
• Significant time savings: With VerbaSense you can get results within hours - sometimes even minutes - as opposed to days.
• Easy conversion path:  If you already have an established code frame(s) for your tracking survey questions we can easily adopt all your existing Code Frames into VerbaSense to insure continuity and consistency.   After the conversion, VerbaSense will continue to use your existing codes as well as statistically identify new issues or ideas that may warrant new code(s).
• Integration with your survey systems / platforms:  You can deploy VerbaSense web services option for easy and seamless integration into your environment.   VerbaSense web services can give you near-real time coding integration for your tracking studies or ongoing call center applications or online reporting systems.  Your systems send the content and VerbaSense responds with a code.   This can be particularly important if you have time sensitive coding or content enumeration needs.
• Intercoder reliability issues are neutralized, by definition.  Content coding world has always been concerned about potentially statistically biasing elements that are introduced by human coders.   Intercoder reliability is the widely used term for the extent to which independent coders evaluate a characteristic of a message or artifact and reach the same conclusion.  "It is widely acknowledged that intercoder reliability is a critical component of content quantification and analysis."   These potentially biasing intercoder elements include (but not limited to) the following:  
  • inadequate training, and calibration: You need to insure that all coders are on the same page. 
  • coder fatigue: Imagine doing the same thing over and over again for extended periods...  
  • demographic characteristics of coders  - particularly the education.   Real life example: An "account executive" gets coded as a CPA. 
  • Socio-economic characteristics of coders.   Many topics areas such as medical, financial, scientific require higher skill coding labor because articulations involve technical vocabulary and acronyms.    
  • Life experiences of coders.   In most organizations "coder" positions are filled with part time labor or off-shore labor to get it done "cheap".    
  • Language barriers.
  • Cultural and contextual familiarity.   Many organizations outsource the coding function off-shore.  While that may provide short-run cost savings, challenges emerge in other areas.  Real-life examples: "Cracker Jack" gets coded as two separate items.   "Home run" is interpreted as running home...   
• Statistical determinations:  VerbaSense platform has built in awareness of statistical concepts that are imperative for making reliable statistical inferences, and representations.   The system makes heavy use of sampling theory and probability concepts to insure that inferences are statistically valid.  
• Rigorous QA process that measures statistical reliability of codes and the coding process.   VerbaSense achieves reliability, and quality via smart utilization of technology, maintaining the delicate balance between automation and human expert review.  At the end of the day this is not magic, we just leverage technology better for faster, more efficient and higher quality research outcomes.  

VerbaSense Solution highlights:

• User can choose between fully or semi automated verbatim response coding modes.
• Automated construction of the code frame.
• Automated construction of lexicons.
• Software does not require pattern training.
• 90% base reliability. 
• No inter-coder reliability or bias issues.
• System gets smarter as N increases over time.
• New ideas and concepts are automatically detected, and if warranted, are automatically added to the code frame.
• Enumerated ideas and concepts can be input into other quantitative discovery, and causality techniques, or models.
• Backed up by disciplined team and strong ongoing committment to R&D.
• Secure hosting in world-class data center.
• Flexible business model, and upgrade path to other ExaSense solutions that preserve your investment.
• Effective training programs that helps you maximize the benefits of the VerbaSense technology. 

• VerbaSense Medical: Process medical transcripts and point out insurance codes,
• VerbaSense Law:  Process legal proceedings and point out relevant precedence categories, and cases.
• VerbaSense Education:  Automatically grade papers, classify essays.
• VerbaSense Web:  Classify web sites, competitive assessment - your web site content vs.. your competitor's web site(s)
• VerbaSense Security:  Process suspect email or web site collections, look for relationship patterns that are not so obvious to naked eye, intelligent flagging and routing for expert review.