The cloud computing model enables organizations to employ policies for the automated provisioning of computing resources. The impact on the quality, such as performance or cost, of such policies is often not known in advance for complex, large, and highly distributed cloud applications. Software architects lack a feasible approach to evaluate scaling policies for their cloud application quantitatively. While approaches exist in the literature, they are costly and require a high effort. This paper first explores the variability in style and configuration of scaling policies for cloud applications. Second, we propose modeling and the use of terminating simulations to evaluate cloud scaling policies. The approach aids the architect in understanding and explaining their dynamic behavior and the existing tradeoffs. Third, we conduct simulation experiments on a representative case study model to show the approach’s feasibility. On a model, we evaluate the performance, cost, efficiency, and complexity of three different policies: two are centralized policies, and one is decentralized; one is aware of the application architecture, whereas the two others are agnostic. Results show that the policies improve the performance for the selected model and scenario. However, no significant difference among them exists in terms of performance. Besides, other metrics highlight the present tradeoffs across policies. All in all, the case shows that the approach helps architects refine the style and find an appropriate policy for their context.