Job creation schemes (JCS) have been one important programme of active labour market policy (ALMP) in Germany for a long time. They aim at the re-integration of hard-to-place unemployed individuals into regular employment. A thorough microeconometric evaluation of these programmes was hindered by the fact, that available survey datasets have been too small to account for a possible occurrence of effect heterogeneity. However, identifying effect heterogeneity can help to improve the design and implementation of future programmes. Hence, we use administrative data of the Federal Employment Agency, containing over 11,000 participants to analyse the employment effects of JCS on an individual level. We focus explicitly on effect heterogeneity caused by differences in the implementation of programmes, whereas we analysed these effects with respect to group-specific and regional heterogeneity in a previous paper. At first, we evaluate the effects with respect to the economic sector in which the JCS are accomplished. Second, we analyse if different types of support lead to different effects. Finally, we examine if there are varying effects which can be attributed to different implementing institutions. The results are rather discouraging and show that JCS are in general not able to improve the re-integration chances of participants into regular employment.