Our research focuses on combining marketing research with algorithm- based recommender systems research in order to test the suitability of recommender systems in the communication strategy of a mass retailer. Personalization is very popular; however, recommendations are still rather rare in offline settings. This doctoral dissertation addresses this fact by assessing the effectiveness of a personalized communication strategy using collaborative recommendations. These recommendations take the tastes of users similar to the given customer into account. Specifically, we investigate the impact of personalized collaborative recommendations and personalized collaborative messages included in a marketing communication in the mass retail sector. Avenues for further research would be that mass retailers could imagine to communicate in real time with their customers in store using mobile devices or to construct a personalized customer relationship with the self-scanning tool in store.