We're all familiar with the traditional outsourcing model of using external service providers to handle IT infrastructure and maintenance in hopes of cutting costs and freeing in-house IT staff to focus on higher-value activities unique to the business. But something similar is happening with data analytics: Companies are starting to supplement their in-house analytics capabilities by using external providers.
At a recent meeting of the Society for Information Management's Advanced Practices Council, researchers Gabe Piccoli (University of Pavia) and Federico Pigni (Grenoble Ecole de Management) described a startup called Versium—one example of a company offering this new breed of data-analytics services. Versium can combine a business's own data with Versium's collection of customer data and apply predictive analytics to better understand, find and retain customers.
Versium's data warehouse has over 300 billion online and offline observations about consumers, such as purchase interests, social-media behavior, demographics, education level, family status, financial rating and life changes that might trigger new purchases. These attributes are combined with enterprise data to produce predictive scores and consumer intelligence.
Predictive scores include fraud scores (who is trying to scam us?), churn scores (who is most likely to cancel?), social influencer scores (which customers affect peers' behavior?), wealth scores (what is the predictive buying power of my consumers?), shopper scores (who are discount shoppers vs. full price?), and recommendation scores (which offers should be sent to which consumers?).
At the council meeting, Barbara Wixom, an expert in business intelligence at MIT's Center for Information Systems Research, offered other examples of companies getting data and analytics from external providers—either while they build their internal capacity or in lieu of doing so. She cited the rental-car company Hertz, which supplements its in-house analytics resources and data warehouse with external services.
Hertz outsources the selection and provision of non-Hertz data, as well as the processes of modeling and cleansing data, hosting and managing data, and gleaning insights from that data. These outsourced capabilities allowed Hertz to let customers swap or upgrade their reserved rental car on their mobile phones.
Hertz also supplements its in-house capabilities with software from IBM and Mindshare Technologies for a "voice of the customer" analytics system that examines thousands of comments from Web surveys, emails and text messages so the company can quickly pinpoint and resolve customer problems.
Using a set of linguistic rules, the system automatically categorizes comments with descriptive tags like "vehicle cleanliness," "staff courtesy" and "mechanical issues," freeing location managers from having to tag them manually. The system also flags customers who request a callback from a manager or who mention Hertz's customer loyalty program.
Analytics outsourcing can speed up the delivery of new services, provide access to advanced technology, and give access to data scientist skills that are notoriously hard to acquire and retain in-house. And it could be a short-term solution while the company figures out its data-analytics strategy of the future. But be careful: You'll need contract terms that protect competitive information and practices.
Are you using external providers to strengthen your customer engagement? Are you plotting both a short-term and a long-term data analytics strategy? Your competitors most likely are.