Attensity Makes Text Analytics Smarter
June 14, 2011

My research into customer analytics shows three important things: Text analytics are at the early adopter stage; companies still use spreadsheets as their main tool for analysis; and to move companies away from spreadsheets vendors must offer tools that are as easy to use as spreadsheets. That’s no easy task, given the huge volume and varied types of text data companies are generating and the complexity of analyzing unstructured text. However, the research also indicates that this challenge will be met, and a new software release from text analytics vendor Attensity is the type of product that can help companies overcome these challenges.  

As with so many things today, the driver is social media. Companies at the very least want to monitor what consumers are “saying” about them on social media, and what consumers are saying produces a lot of unstructured text. Whether tproducts are called text analytics or social media analytics, companies are beginning to deploy systems that allow them to tap into social media and produce analyses of what consumers are saying about their brands, products and services. As they do this, companies are realizing they have a wealth of other sources of text data – email, forms, surveys, Web scripts and others – that can be analyzed to form a bigger picture of their customers, which is now popularly called the Voice of the Customer. 

This need for text analytics plays to the strengths of Attensity. The company recently announced Analyze 6.0, built on the existing Attensity Data Grid and its powerful natural language processing engine, which can deliver insights, discover trends and analyze sentiment from text-based customer interactions. 

Major enhancements in this new version make the product easier to use. A new exploration environment gives users the ability to create new types of analysis using drag-and-drop techniques. Users can select the data they want to analyze, filter it by several different types of parameters (such as names, products or scores), design a dashboard that meets their individual requirements either from scratch or using one of the provided templates, and then drop the required reports into the dashboard. Users can select from a set of standard reports or build reports to individual requirements. The software provides extensive visualization tools: Information may be displayed as tables; as bar, pie and line charts; as category trees (such as types of complaints within the category Complaints); and as gauges and something Attensity calls word clouds, which display the key words in a document with color-coding and variable font size. Word clouds allow users to get a quick overview of the content of selected documents, which in turn helps them create or refine categories of interactions based on frequently used words. As one user told me, it allows them “to see what they need to know so they can create the correct analysis rules.” Last but not least, new capabilities allow users to share or distribute reports and analysis in multiple forms using one-click access to email. 

The software also supports a number of functional enhancements, starting with new metrics about percentage totals by category or subcategory. The new “hotspotting” feature, which goes beyond normal trend analysis, allows users to see any significant change between the current analysis and a historical trend. For example, some words or phrases may have just started to be used, or a phrase that once occurred 10% of the time now appears 25% of the time. This guides users to look for a significant event that might have brought about the change. The final major functional enhancement allows users to use social media information to enhance their profiles of customers, tracking for example what other products a customer has (and maybe is complaining about). As well, more technical enhancements improve data security and system administration. 

In this major release, Attensity has recognized that different generations of people use new sets of words, phrases and acronyms, and that different industries have their own sets of terms. It has enhanced the application’s word-spotting capabilities to recognize text and social media slang and put it into the context of other content. Attensity also announced a banking industry version that recognizes many specific banking phrases and idioms. This version also includes additional multichannel analysis and enhanced workflow capabilities, which for example allow interactions to be delivered to the person best equipped to deal with them. 

All of these enhancements add up to a major advance in how Attensity Analyze can be used to derive insight from many forms of text-based data. It addresses key issues for users by making the software simpler to use and includes features that go beyond what spreadsheets can provide.  

Recently I wrote that it is time for companies to change the way they handle customer interactions. Gaining insight into what customers like or dislike using text analytics is a good place to start. Has your company recognized the benefits text analytics can deliver? Please come and tell us your thoughts. 

Regards 

Richard Snow – VP & Research Director

 

 Copyright © 2013 Ventana Research All Rights Reserved :: Privacy Statement :: Contact Us ::