Purpose - The design of distribution networks is prone to risks due to the uncertainties associated with factors that change over time. The purpose of this paper is to present a new method to identify those factors that the structure of a distribution network is most sensitive to. Design/methodology/approach - A new method presented in this paper combines simulation and the Taguchi technique to allow a wide range of factor uncertainties to be evaluated without excessive computation time and effort. The simulation model developed is based on real world data of a European after-sales business in the automotive industry. Findings - The optimum design is most at risk due to the uncertainties associated with stock holding costs and delivery frequencies rather than customer volume changes and transport tariffs. This was found to be counterintuitive by the business managers and fore-warned them of the likely future risks. Practical implications - The analysis indicates that when developing the network a careful consideration has to be given to reliably estimating the inventory holding costs and the mechanism for determining the capital holding charge. The model is sensitive to these variables and hence a company is prone to higher risk of designing the wrong network if these variables are incorrectly estimated. Furthermore, the analysis suggests that higher customer expectations on delivery frequency have an impact on total logistics costs and therefore, companies should consider charging customers different rates for different levels of customer service. Originality/value - A useful generic method has been presented in this paper for investigating the sensitivity of a scenario. This novel approach combines the use of simulation, brainstorming, Taguchi technique and ANOVA with distribution design modelling.