Animal behavior is extremely complex, to the extent that it may seem beyond human comprehension. Recent years have witnessed an explosion of experimental techniques that provide experimental data with unprecedented detail and quantity, making it even harder to deal with the resulting complexity. However, this revolution is also getting us nearer to the point where we can achieve a full understanding of some relatively simple animals. Just like the last pieces of a puzzle are easier to fit than the first ones, achieving this complete understanding may help us deal with the complexity. To illustrate this concept, I will present a comprehensive mapping of food preference in Caenorhabditis elegans, which shows a surprising universality in a system that was believed to be contingent on a myriad of details.