Self-adaptability enables a system to adapt itself tochanges in its execution conditions and user requirements in orderto achieve particular quality goals. However, assuring that theadaptation goals are satisfied poses complex challenges. Werecently developed the ActivFORMS approach that aims to tacklesome of these challenges, but further research is required toevaluate the approach. This paper presents the results of a study inwhich we applied ActivFORMS to a mobile storytelling applicationthat employs a social recommender. The initial version of theapplication used a static recommender that could not deal withchanging environment conditions, or take into account preferencesof users. To that end, we added a self-adaptive layer on top of theapplication. The study results show that self-adaptationsignificantly increases the quality of recommendations comparedto the initial version by: (1) enabling the social recommender toadapt to the quality of user input and unavailability of the GPSservice, and (2) making the recommender adaptive to userpreferences. Providing guarantees for these adaptation goals iscrucial in this domain from a business perspective. The studyresults show the feasibility and effectiveness of ActivFORMS for apractical application; but they also underpin the need for anintegrated verification approach for self-adaptive systems thatcombines offline with online verification.