Combining marketing models to find a strategy for increasing the profits of manufacturers and retailers, when going to the electronic market, is discussed in this paper. The strategy derives from applying the models to compute the profits in different scenarios, with a precise choice of hypotheses and parameters. After justifying the need for a model that maximizes profits when manufacturers sell products through the online channels as well as through traditional retailers, Yan gives background information and develops the model.

For a dual market, the model for computing profits is obtained using models in the literature. Two models are considered, in particular: Stackelberg and Bertrand. Using the Stackelberg model, when the manufacturer decides the online price, Yan demonstrates that in the dual channel the performance increases. In the Bertrand model, when there is no price leader in the market, the same conclusion is reached. Demonstrations of the theorems are presented in the appendix.

The rest of the paper defines the optimal profit schema, demonstrating that the optimal policy for both manufacturer and retailer is the profit-sharing policy. The conclusions translate the results in managerial policies, and point to further research to reduce the limitations of the presented study.

While the paper is clear and well written, some points are left unanswered. This is a model built over many models. Where is the data? (The original data is lost in the models that were built using models as input.) What if we use another model or a different parameterization? Where is the experimental proof? Are the conflicting attitudes of people in the marketing scenario really irrelevant, or are they important, as considered in the multiagent paradigms?