This document presents guidelines on how to setup enriched video experiences. We provide user-centric guidelines on the named entities that should be detected and selected to effectively enrich video news broadcasts. This is presented in the form of a user study. We selected 5 news videos and manually extracted the candidate entities from various sources, such as the transcript, visual content and related articles. An expert was asked to also provide interesting entities for the videos. The resulting 99 candidate entities were presented to 50 participants via an online survey. The participants rated the level of interestingness of the entities and the usefulness of information from Wikipedia about these entities. Analysis of the results shows that users prefer entities of the type organization and person and have little interest for entities of the type location. They also indicate that subtitles are not enough as a source of interesting entities and that the amount of interesting entities can be improved by the combined use of subtitles with entities extracted from related articles or entities suggested by an expert. The expert suggestions showed to be more accurate than any other source of entities. Wikipedia seems to be a suitable source of additional information about the entities in the news, but should be complemented with additional sources. We provide engineering guidelines on how to present, aggregate and process content for TV program companion applications. We describe the content processing pipeline that was developed in WP3 to feed the content for the Linked News and Linked Culture demonstrators. This shows how content from the Web can be re-purposed to enrich videos by extracting the core display content and presenting it in a uniform way to the user.