Münster (upm/ch)

Romance or action film?

Automated group recommender systems can help consumers make better choices

What's it going to be – a romance, an action film, or how about a comedy this time? If you want to get a movie from a video store or go to the cinema, you're spoilt for choice. Picking the right film can be tough, especially when more than one person is going to watch it. Scientists at the University of Münster show that new, automated recommender systems can help consumers choose a movie which all members of a group will enjoy. Such systems therefore have great potential for companies wanting to increase customer satisfaction and customer loyalty. The study from Münster has just been published in the current edition of the prestigious "Journal of Marketing".

Intelligent recommender systems can provide users with customized product recommendations based on personal preferences and previous behavior. Systems which generate recommendations to individual customers are available, for example, on online sales platforms such as Amazon. Group recommender systems however are not yet widely used. Presenting the results of their extensive study on the movie selection process, the authors are the first to demonstrate that such group recommender systems can indeed generate additional value for users. "Our results show that customers can make better choices and are more satisfied with a movie when they use group recommender systems," Professor Dr. Thorsten Hennig-Thurau from the Marketing Center Münster explains, who carried out the research together with Dr. André Marchand and Paul Marx. The authors also found that group recommenders have an even greater potential benefit for people who are close to each other and open to recommender technology.

Recommender systems are not only useful regarding the selection of movies, but can also be employed for other leisure activities including vacations, restaurant visits, and concerts. "Group recommender systems can be applied in a large variety of settings. Just bear in mind that most people spend a considerable portion of their free time with friends, acquaintances, or family," André Marchand points out. "I am convinced that automated decision aids will become immensely important in the future," he continues. "After all, the growing number of choices offered by the leisure and media industries is getting harder and harder to keep up with." The scientists hope that their findings will motivate retailers to focus not only on individual customers, but to also generate attractive recommendations for groups.

The study had a total of 460 participants. One member of each group was allowed to choose, from a selection of movies which none of the group members had seen, a movie to later be watched by the whole group. They were given recommendations which either incorporated only their own taste in films or also considered group preferences. A control group of participants chose a movie without the aid of automated recommendations.

Paper: Thorsten Hennig-Thurau, André Marchand, and Paul Marx (2012), "Can Automated Group Recommender Systems Help Consumers Make Better Choices?" Journal of Marketing, 76 (5), 89-109