We propose a novel approach to utilize low-cadence photometric surveys for exoplanetary transit search. Even if transits are undetectable in the survey data base alone, it can still be useful for finding preferred times for directed follow-up observations that will maximize the chances to detect transits. We demonstrate this approach through a few simulated cases. These simulations are based on the Hipparcos Epoch Photometry data base and on the transiting planets whose transits were already detected there. In principle, the approach we propose will be suitable for the directed follow-up of the photometry from the planned Gaia mission, and it can hopefully significantly increase the yield of exoplanetary transits detected, thanks to Gaia.