eHealth development is not a linear process with consecutive steps. It is an iterative, flexible and dynamic process, during which constant changes can be made to development activities and their results. Consequently, evaluation should also be seen as a cyclic, longitudinal research and development activities interwoven with all development phases, without a fixed end. This means that evaluation doesn’t only take place at the end of the development process; just like implementation it is not a post-design activity. Formative evaluation already starts at the beginning of the development and continues during every development activity. Each product of a development phase can and should be critically checked, analysed and evaluated and adapted based on the results of this formative evaluation. It can take on different forms, for example, verifying outcomes of a phase with users, checking the relation with the outcomes of previous phases, or gathering stakeholder’s opinion on a specific idea. In every case, its main goal should be checking that the outcomes of activities still match the context, stakeholders, and outcomes of previous phases. Formative evaluation provides concrete tools to further improve the process and technology, in order to reach an optimal fit between technology, stakeholders and context.
Furthermore, much eHealth research focuses on evaluating the effectiveness of an implemented technology to make claims about whether the goals have been reached. Less attention is paid to outcomes related to the healthcare context and the interaction between the user and the technology, which can be seen as equally important. Just like eHealth development, evaluation should be holistic: it has to focus on the technology, users, and the context. Also, evaluation does not have a fixed end-point, since its results can be used to further improve or change a technology. Once implemented, a technology is not fixed, but can be redesigned. These improvements or changes should be evaluated as well, and again, more changes can be made based on the outcomes of the new evaluation cycle. Evaluation doesn’t have to be the end-point of the development process, since a technology is never really finished. This requires an iterative, flexible and dynamic view on evaluation.