POWER > Case Studies > Services

Ufone Uses Advanced Analytics to Study and Capitalize on Customer Behavior

Faisal Khaliq, CIO with Ufone - Photo by Sam Phelp

To increase marketing visibility, Ufone decided to reassess how it tracked customer data and created individualized and targeted marketing offers. To that end, it decided to deploy InfoSphere Streams and the Unica Campaign Management solution in an integrated single-silo environment, most of which runs on the Power 795 in a secured environment.

Currently, marketing employees will develop marketing offers based on CDRs, which are pushed to the Streams servers. That system will then analyze the CDRs—in 30 seconds instead of the previous three hours, thanks to Steams’ “big data velocity processing,” says Khaliq—and, based on predetermined rules, recommend a reward or service to be provisioned. Following that, the Unica server will poll the database for subscribers that match the targeted provisioning outlines.

On-the-Spot Analytics - Ufone may use place-based information in the future to offer on-the-spot coupons to customers walking past particular restaurants.

The WebSphere middleware then validates the requests, sometimes by performing simple data lookups in a variety of Ufone systems, including CRM and billing. Once the validation is complete, the middleware will provision the marketing offers based on business rules either stored in the middleware or in external files and databases. The resulting responses are then fed back into the primary database for further business-intelligence reporting.

That information is analyzed to create a deeper look into customer behavior. Based on the results, marketing efforts can be further refined to improve customer responses. For example, if a customer calls Ufone, the customer service representative can see a list of specialized offers in addition to traditional information, such as name, address, package type, billing transactions and any outstanding complaints.

As Khaliq explains, the customer service representative would present an offer that is most likely to appeal to the customer, based on analytical modeling. If the customer says no to the first one, the representative would move on to the next offer and so on. That information is fed back into the analytical system, which can then be used for other customers with similar behavior—if, for example, they’re heavy texters or heavy callers or even if they’re displeased with the services. “It’s amazing how well this type of customization works,” Khaliq adds. “Once you know the likes and dislikes of a particular customer segment, you can quickly home in on the best offerings to present to them. This creates a nearly individualized experience for our customers.”

Notably, this information is available in near-real time, based on CDR history. And in today’s world of smartphones, companies such as Ufone have a bevy of data available to them about individual customers, including, for instance, most-visited blogs, fashion preferences or technology interests.

Even unsuccessful campaigns can be instructive. If the churn or opt-out rate is the same for a control group that didn’t receive an offer and a target group that did receive an offer, companies can quickly determine the campaign wasn’t a success. Organizations can then feed that information back into their analytics engine to tailor campaigns to make them more successful. “The possibilities are nearly endless,” Khaliq says.

Jim Utsler, IBM Systems Magazine senior writer, has been covering the technology field for more than a decade. Jim can be reached at jjutsler@provide.net.

comments powered by Disqus



2019 Solutions Edition

A Comprehensive Online Buyer's Guide to Solutions, Services and Education.

A Logical Approach

Taiun Company Limited improves cost and continuity with consolidation and a new disaster-recovery solution

IBM Systems Magazine Subscribe Box Read Now Link Subscribe Now Link iPad App Google Play Store