Once you know why someone turns on the TV, it becomes much easier to suggest to them exactly the content they want to see. More on this project regarding contextual information for recommender systems at http://bit.ly/1DBpLIO . Findings from 'Empathic Products', an exploratory research project with VRT O&I on mood and context-aware media recommendations. Mood, social company, activity level and other context variables as detected by the Sentiance mobile sensing SDK were found to be good proxies for need states, and whilst the current state of the art recommender systems perform optimally when purely engagement is the goal, there is a challenge still when the broader TV experience becomes more important than content engagement. This is exactly where context information becomes detrimental.