The observational evidence of billion solar masses black holes being already in place less than 1 Gyr after the Big Bang poses stringent time constraints on their growth process. The formation of massive seeds and/or the occurrence of super-Eddington accretion episodes may contribute to the solution of this problem. Employing detailed numerical simulations and analytic techniques, we develop a general model to describe the rapidity of their growth. We then predict the time evolution of the spectrum emerging from the host halo, which allows us to estimate their detectability and put the first constraint on the mass density of this population of high-redshift objects. The next step is to detect them: the case of the Lyman alpha emitter CR7 is presented, along with a new photometric method to select black hole seeds candidates.