The Regge-plus-resonance (RPR) framework for kaon photoproduction on the proton and the neutron is an economical single-channel model with very few parameters. Not only does the RPR model allow one to extract resonance information from the data, it h
as predictive power. As an example we show that the RPR model makes fair predictions for the $p(e,eK^{+})Lambda$ and the $n(gamma,K^{+})Sigma ^{-}$ observables starting from amplitudes optimized for the reaction $p(gamma, K ^{+})Lambda$ and $p(gamma,K^{+})Sigma ^{0}$ respectively.
