I will discuss what recent multiwavelength surveys tell us about the drivers for H2 formation on galactic scales (and by extension what makes the ISM good at forming stars). The question is an open one, with results in the last few years variously identifying gravitational stability, metallicity, pressure, and colliding flows as the regulating quantity. From large programs with the IRAM 30m ("HERACLES") and the VLA ("THINGS") we know the distribution of atomic and molecular gas in several dozen galaxies that have also been observed by Spitzer and GALEX. These data allow us to estimate a range of physical conditions over kiloparsec-sized resolution elements across each galaxy (e.g., the star formation rate, HI mass, H2 mass, stellar mass, kinematics, dust-to-gas ratio, and radiation field). We are using these estimates to push beyond galaxy-averaged scaling relations, testing proposed local drivers for the formation of molecular clouds and stars from diffuse neutral gas. I will give our current best answers to the basic question "where is the interstellar medium good at forming stars?" and talk about where we're going next.