There is great interest in understanding the value of data. However, its non-rivalry and externalities imply that private markets and market prices (if they exist) will not reflect social value. What’s more, the value of any given data depends on the uses to which it can be put. This paper applies a form of discrete choice analysis widely used in marketing to elicit users’ preferences and the marginal value of data. Given the widely-acknowledged importance of data in the economy, public data sets are an important part of the national economic infrastructure, we selected a prominent open data set. The method delivers estimates that could be used by public bodies for purposes such as estimating the social return to the data they provide or setting prices for ‘freemium’ access in cases where they are required to recover some of their costs.