It is increasingly recognised that a large amount of potentially valuable information created during the engineering design process is contained in less formal sources, such as log books and emails. Hence there is a requirement to improve the capture, management and re-use of these sources. However it is first necessary to understand the nature of the information in order to effectively achieve this. The research reported in this paper deals with the creation of an approach for assessing the information content of emails, in particular, those associated with design projects. The development of this approach, grounded in a wide range of existing literature and derived from iterative applications to an industrial dataset is presented. The approach considers three perspectives: what the subject of the email is, why the email is being sent and how its content is expressed. The application of this method to a 700 email corpus from a software design project is described. Key findings from this are presented and discussed, with respect to the character of the content and the project lifecycle.