There is widespread understanding the individuals vary in their responses to different types of interventions aiming to reduce obesity. A major goal of precision nutrition is to target interventions towards those who will respond best. This assumes that the variation in response is repeatable and has a physiological or genetic basis. Individual differences are often presented as ‘waterfall diagrams’ where the individuals are ordered by the magnitude of weight loss. These show that on some diets a small population of people actually gain weight, while others lose tremendous amounts. In humans, the underlying reason for these differences remain unclear but may reflect issues of adherence to the prescribed diet/intervention. In mice we can place individual animals onto diets and there are no issues of adherence. A major question then is what the waterfall diagrams look like when we do this? Do these patterns support or undermine our goals with respect to personalizing nutrition interventions in humans? May talk will present data on these latter two issues.