One of the strongest predictions of the cold dark matter paradigm is the hierarchy of structure down to Earth-mass scales. However, individual self-bound clumps of dark matter--"halos"--are difficult to detect directly. Instead, we use galaxies as lampposts for halos. By counting galaxies, our goal is to measure the underlying population of dark matter halos. In this talk, I describe two results that seem completely at odds with each other in measuring the population of small halos. I argue that the resolution to the problem is a better mapping between galaxies and halos. I will show what my group is doing so far to address the problem, and what opportunities lie ahead in the wide-field surveys of the 2020's.