It is now commonplace for Universities to record lectures with video cameras. Indeed there are several off-the-shelf systems, which Universities can purchase to provide this type of functionality, e.g. Echo360, Panopto etc. There are also several distribution outlets available, such as iTunesU and YouTube EDU, which Universities can use to distribute this recorded media to students. However, the capture of standard lecture material with these systems can only provide partial support to learning. Material recorded in this way can be engaging for students who attended the original lecture, but has less efficacy for students who are seeing the material for the first time. To be truly effective learning mechanisms in their own right, these new recording systems need to address two key issues. Firstly, current lecture material is overwhelmingly designed for the live lecture theatre audience. Consideration is rarely given to how these materials will support learning when viewed as stand-alone learning resources. Secondly, as lecture theatres are rarely designed for video capture, the off-the-shelf recording systems are often severely limited by the environment, equipment and resources available. Lighting and camera position are key considerations that have a big impact on the quality of the captured material, but are generally restricted by the environment required for the live audience. This paper reviews these two key issues and presents both a framework for the production of teaching material targeted at video capture, and the bespoke recording system developed for online learning in the School of Informatics at the University of Sussex. Additionally the paper covers analysis of download rates, qualitative staff and student feedback and lecture attendance and shows that using this framework has a significant effect on the student interaction with recorded material. Other types of online support such as providing copies of lecture slides are also discussed and a tangible improvement in engagement over these techniques is shown.