In this paper, we suggest a novel approach to program evaluation that allows identification of the causal effect of a training program on the likelihood of being invited to a job interview under weak assumptions. The idea is to measure the program-effects by pre- and posttreatment data that are very close in time for the same individual. Our approach provides useful information on both, average effects of the program as well as information on potential differences in the effects of the program by type of vacancy or by individual.