Friday, October 23, 2009

How Semantic Technology and Text Extraction can help in listening to "Voice of Customer!"

Businesses succeed by getting, keeping, and growing customers. Customers are the only reason you build factories, hire employees, schedule meetings, lay fibre-optic lines, stock inventory, write software  or engage in any kind of business activity. "Without customers, you don't have a business. You have a hobby." These are not my words but they are explained very nicely by Don Peppers and Martha Rogers in their famous book - Return on Customer. The book is basically about developing a business metric around your customers that can drive better management and higher stock price.

We know that it is becoming increasingly hard to keep this customer as they are no longer hungry for what a business has to offer. They have more choices than ever as far as developed and emerging economies are concerned! At the same time, Social media tools have made it possible for hundreds of millions of people on this planet to express their opinion. And everybody has an opinion about something! If that "something" is not a person or social/political/economic  issue than chances are that it is likely to be a product or a service! It is becoming very important for businesses to understand what their customers are thinking and saying about their products and services. They know that they will lose this customer as customers' shift loyalties very quickly if they are not heard. Businesses call it "Voice of Customer" and have been trying various ways in last few decades like polls, surveys, incentives and some other ways to understand that. There have been many issues with most of the earlier approaches as it is not an exact science and new regulation and privacy issues are making it harder. Marketing and support personnel have been trying hard to keep tabs on real-time conversations regarding their products, services and brands in blogs, Web forums, wikis, microblogging services including Twitter and Facebook. The advancement in text analytics is bringing  a complete new approach to hear VOC.

In a nutshell, "text analytics" is about processing unstructured information in text and transforming it into relevant information. It is a highly sophisticated technology which uses concepts like clustering, categorization, advanced pattern recognition, machine learning, AI,  language identification, semantics, linguistics and relationship extraction . The text processing can extract entities, concepts, events and sentiments from the text and marry it with structured data in databases and data warehouses to get useful business insights. Pharma and intelligence community have been the earliest adapters of this technology as they have compelling need and ROI is very high. In of my earlier blogs, I have also explained how Open Calais, using text analytics, is helping Semantic Technology.

Attensity Group, one of the remaining leaders, after Clearforest and Inxight were acquireed by Reuters and Business Objects respectively, text analytics space, has announced a very interesting product called Attensity Cloud. They built this product by combining their patented text analytic technology with Radian6 platform. Radian6 provides a platform to develop a dashboard, metrics, filtering and segmentation for social media data. According to Attensity Group, the Cloud plaform also uses sematic web technology along with text analytics technology. The Attentisty Cloud can tell things like:
  • Which conversations are relevant and active
  • Which products are most talked about e.g hamburger vs. salads in case of Mcdonald
  • Sentiments behind each product
  • Intentions to buy product
  • Top issues, cries for help
  • product suggestions
  • Eearly warning of issues
  • Actionable facts
Morever, you can integrate this platform with CRM and e-service to get value across all channels. I think that it is a very good move as it is a very smart idea to tie text analytics technology with semantic technology and applying it to solve a business problem. It can allow organizations to monitor, analyze and respond to social media in more effective and faster way.

Since, the platform is introduced very recently so there are no real customers. It will take a while to understand the power and usefulness of this plaform and how it is working out for Attensity's customers. Nonetheless, the concept is good because it will probably help in customer retention if businesses can integrate, configure and deploy Attensity Cloud successfully along with CRM systems. Because, In the end, any customer needs just few basic things:
  • They want you to know who they are
  • They want you to know what they need
  • They want a reliable way to communicate with you
  • They want you understand their definition of value
  • They want you to know what they want next
  • They want you to support them if they trust you and buy from you
  • And they don't want you to ask the same questions
I believe, despite doing all of these things, it is still a challenge to retain a customer if you don't innovate. Do you know the number of customers Apple stole from its competitors, who thought they were doing everything right to hear VOC, by introducing iphone? But thats a different discussion for a future blog!


  1. This is an excellent article that I found on TipTop, the new semantic search engine that processes tweets in real time. Since you are clearly an expert in this field, I would love to get your feedback. I can be reached at shyam at

    Thank you.

  2. Please consider opposite statements. Customers:
    - don't want you to know who they are,
    - hate your illusions as to what they want,
    - want a reliable way to avoid communication from you,
    - ....

  3. Great blog post! There are quite a few vendors offering -either- text analytics -or- social media analysis -or- CRM, few other than SAS can deliver all of these capabilities. Wouldn't it be better to have one solution to the problem? Most I.T. managers say they'd much rather consolidate. Which users are up for managing all the integration?!

  4. Insight 1: Not every customer is online and even if they are not every one of them makes herself/himself heard.
    Insight 2: Everything that applies AI should be approached with caution. AI tells you what _it_ thinks it sees and not what it actually is.
    Altogether, using such tools may help you a lot but you have to keep in mind that you're observing a segment of your customers through a somewhat blurry lens.