Web interventions are gaining popularity in trying to change a person's behavior. However, poorly designed intervention websites will affect the learning process of a person, what more to remember the content that they have learnt. The objective of this paper is to discuss the design of a web intervention using Gagne's Condition of Learning Theory and cognitive dissonance. The usage of learning theory enables the designers to develop the intervention website based on users' learning capabilities, thus placing users as priority. This ensures that users are able to maximize learning and recall of its content when faced with the decision to smoke or not. We are planning to recruit target users who are smokers, ranging from the 18 to 22 age range. Smokers will be categorized according to a baseline survey. Each category of users will go through the web intervention of different content. It is hoped that by combining the abovementioned theories, smokers are able to reduce their intake. © 2009 Springer Berlin Heidelberg.