One of the biggest difficulties in ensuring vaccine success against influenza viruses is the ability of the viruses to constantly evolve and avoid vaccination-induced immunity. As a consequence, vaccination against avian influenza often fails, and constant vaccine updating is too costly. In our research we examined the evolution of avian influenza viruses by establishing a novel method of selection for antigenically-different mutants. We conducted extensive transmission experiments and quantified for the first time the minimum level of immunity that animals need to have in order for a vaccine to protect them against antigenically different viruses. From our results it is clear that the antigenic differences between vaccine and virus are not the main problem in vaccine effectiveness.