It has always been challenging to select breeding stock to ‘improve’ traits they do not express themselves. The classic case must be our inability to select bulls based on their milk yield; we need to rely on their mother’s, sisters’ and daughters’ performance to suggest how future offspring might shape-up. This traditional approach, relying on progeny testing means a bull is typically 5-6 years old before his daughters are milking and his ability is confirmed with any confidence. Another test, perhaps more pertinent to low-input or organic dairy production, is identifying robust breeding animals who’s offspring can cope with varying inputs and disease challenges, without resorting to drug treatment. We don’t want to expose our breeding animals to pathogens to see how they cope before selecting them as AI studs. Genomic breeding could offer solutions to both these dilemmas, allowing us to assess potential breeding stock from the time they are born, as well as their likely ability to stay healthy, despite diseases they or their offspring might encounter. This presentation will explain the principles of genomic breeding in the dairy sector and outline how is has been evaluated to improve functional traits in organic dairy cows, under the recent EU LowInputBreeds project.