The ways in which flavour molecules interact with proteins in foods have an impact on flavour and aroma and on the (in)stability of the proteins. There is a long history of analysing these interactions using a "specific binding model" that gives values for the number of molecules, n, bound to the proteins with a binding constant Kb. However, recent progress in molecular thermodynamics forced us to reconsider this approach. In addition, there are a number of methods for determining these values and it is not at all clear whether the various assumptions behind the various methods allow legitimate comparisons between techniques. By adopting an assumption-free molecular thermodynamics approach, Kirkwood-Buff theory, we find that we gain a welcome universality, simplicity and deep understanding of what is happening at the molecular level. Here we look at three different methods for examining flavour-protein interactions (vapour pressure, dialysis equilibrium and melting temperature changes), show how historical data can be re-cast into the universal language of Kirkwood-Buff and provide a free, open-source app that can both re-analyze historical data and be a platform for analyzing fresh data. In each case, the fundamental theory is described along with a pragmatic implementation accepting the realities of experimentation. One key insight is that the n and Kb parameters of the classical binding models can be turned directly, via simple arithmetic, into the Kirkwood-Buff integrals that accurately capture non-specific flavour-protein interactions.