Many small airports in the United Kingdom are competing for traffic. Ideally, forecasts for these airports should be made by applying behavioral models of airport choice to national forecasts. Although the UK Civil Aviation Authority has attempted this approach, it is difficult to retain sufficient accuracy to deal with the smaller airports. A two-stage method of analysis has been proposed that combines the simplicity of a" step-down" method with the accuracy of discrete choice models. The stability of regional shares of national traffic has been demonstrated for subcategories of traffic. The adequacy of discrete choice models to predict local airport shares, when applied to traffic generated within a region, has also been demonstrated by means of a case study. The two-stage approach is therefore recommended for consideration by airports wishing to determine their potential traffic in a competitive market.