The new generation of powerful instruments is reaching sensitivities and temporal resolutions that will allow multi-messenger astronomy of explosive transient phenomena, with high-energy neutrinos as a central figure. We derive general criteria for the detectability of neutrinos from powerful transient sources for given instrument sensitivities. In practice, we provide the minimum photon flux necessary for neutrino detection based on two main observables: the bolometric luminosity and the time variability of the emission. This limit can be compared to the observations in specified wavelengths in order to target the most promising sources for follow-ups. Our criteria can also help distinguishing false associations of neutrino events with a flaring source. We find that relativistic transient sources such as high-and low-luminosity gamma-ray bursts (GRBs), blazar flares, tidal disruption events, and magnetar flares could be observed with IceCube, as they have a good chance to occur within a detectable distance. Of the nonrelativistic transient sources, only luminous supernovae appear as promising candidates. We caution that our criterion should not be directly applied to low-luminosity GRBs and type Ibc supernovae, as these objects could have hosted a choked GRB, leading to neutrino emission without a relevant counterpart radiation. We treat a set of concrete examples and show that several transients, some of which are being monitored by IceCube, are far from meeting the criterion for detectability (e.g., Crab flares or Swift J1644+57).