IBM’s Predictive Analytics Software Understands You

Pete Prestipino — Wed, 12 May 2010 14:20:00 GMT

IBM announced new software that will enable users to discover and analyze information from social media sources, merging that information with internal data to gain insights and predictive intelligence. You might be thinking that this isn’t really any different from any other analytics solution (predictive or otherwise) but you would be wrong – very, very wrong.

What makes the data mining and text analytics software unique is that it allows for the monitoring of changes in attitudes, uncovering insights and predicting key factors that will influence customer acquisition and retention campaigns. For example, companies can now extract sentiment from the use of emoticons and slang terminology that people often use in describing their view toward a product or service. Did that just blow your mind? Because it just blew mine!

Customers of IBM’s predictive analytics software can directly access text, web and survey data and integrate it into predictive models for recommendations, ultimately making better business decisions. It uses natural language processing (NLP) to allow clients to pull key concepts, opinions and categories relevant to their business from these data sources to uncover deeper customer insights.

Organizations can combine all of their structured data with textual information from documents, e-mails, call center notes, and social media sources, extracting, discovering and exploring relationships between concepts and sentiments, including emoticons and slang terminology at a specific time and through a specific channel.

A good example of this in action comes from Rosetta Stone, a provider of technology-based language-learning solutions. Seeking to proactively capture and analyze consumer responses, the company relies on IBM’s predictive analytics software to reveal trends in text responses from online customer product reviews, competitor websites and open-ended survey questionnaires and recognize why certain customers are brand promoters or brand detractors, and improve customer satisfaction, product development and marketing effectiveness.

Nino Ninov, vice president of strategic research and analysis at Rosetta Stone, said, "Predictive analytics allows us to leverage unsolicited and unbiased customer feedback and strategically improve our business. We now can also monitor competitor and industry websites, including blogs and news feeds, and other publicly available textual information to maintain a current view and better understand how the public perceives our competition."

Four Types of LinkedIn Users

Pete Prestipino — Wed, 05 Nov 2008 15:40:00 GMT

An interesting study from Anderson Analytics was released today and reveals that LinkedIn users tend to fall into four major types. You may have noticed that many LinkedIn members respond differently to various messaging strategies, so understanding their various nuances, makeup and intentions can go a long way towards using the social networking service effectively and ultimately mastering it. If you'd like to see just what kind of LinkedIn user you are, a web-based tool to predict segment type is available at Anderson Analytics.

Using predictive analytics software from SPSS Inc., Anderson Analytics found four major tupes of LinkedIn users:

"Savvy Networkers" (est. 9 million) are likely to have started using social networking earlier than others, are more tech savvy, and more likely to be active on other SNS sites like Facebook. Savvy Networkers have the most connections (61 on average) and are more likely than other segments to use LinkedIn for a wide variety of purposes other than job searching. Savvy Networkers have the second highest average personal income ($93,500) and may often have the word "Consultant" in their job description.

"Senior Executives" (est. 8.4 million) are somewhat less tech savvy and are using LinkedIn to connect to their existing corporate networks. They have power jobs which they are quite content with, and are likely to have been invited by a colleague and then realized how many key contacts were on the site and started building connections (32 on average). Senior Executives have the highest average personal income ($104,000) and have titles such as Owner, Partner, Executive, or Associate.

"Late Adopters" (est. 6.6 million) are likely to have received numerous requests from friends and co-workers before deciding to join. They are somewhat less tech savvy and are careful in how they use LinkedIn, tending to connect only to close friends and colleagues and have the fewest number of connections (23 on average). Late Adopters have the lowest average personal income ($88,000) and have titles such as Teacher, Medical Professional, Lawyer, or the word "Account" or "Assistant" in their job description.

"Exploring Options" (est. 6.1 million) may be working, but are open and looking for other job options often on CareerBuilder.com, perhaps in part because they have the lowest average personal income ($87,500). They are fairly tech savvy and use SNS for both corporate and personal interests.

Some additional findings include:

- Most users connect to people they know, including those they've met only over the phone.
- Users like the professional and business oriented look and feel of LinkedIn compared to other SNS.
- Users tend to be more senior (56% are "individual contributors", 16% are management level, and 28% are director/VP level or above).
- The majority (66%) are decision makers or have influence in the purchase decisions at their companies (decision makers also tend to be more active on LinkedIn).
- And perhaps most interestingly, the greater the number of connections the greater the likelihood of higher personal income. Those with personal incomes between $200K-$350K were seven times more likely than others to have over 150 connections.

'Net Features : predictive analytics

IBM’s Predictive Analytics Software Understands You

Four Types of LinkedIn Users