Innovative Analytics at 1&1 Telecommunication SE

Personalized marketing with next purchase predictions using hybrid recommender systems

Steadily growing amounts of enterprise data as well as the increasing individualization of marketing are the two main drivers of this project with 1&1 Telecommunication SE, which was started in May 2013.

The project’s main goal is the ability to predict customer- or contract-specific preferences at any given point in time, making it possible to play out personalized offers via different marketing channels (online, e-mail, telesales, etc.).

In order to facilitate that, the manifold enterprise data is analyzed using innovative methods of data mining, machine learning, and recommender systems. The main focus is the combination of different algorithms (collaborative filtering, content-based filtering, demographic filtering, etc.) as well as the integration of heterogeneous data sources in the context of hybrid recommender systems. By optimizing the combination weights using statistical methods, the prediction accuracy is maximized in order to be able to play out the best possible offer.

Our role is to bring the latest research innovations into the industry cooperation in order to contribute to building a customer-oriented marketing strategy.