Drinking water distribution systems (DWDSs) have been thoroughly studied, but the concept of thermal energy recovery from DWDSs is very new and has been conceptualized in the past few years. Cold recovery results in a temperature increase of the drinking water. Its effects on drinking water quality and biofilm development are unclear. Hence, we studied both bulk water and biofilm phases for 232 days in two parallel pilot scale distribution systems with two temperature settings after cold recovery, 25 °C and 30 °C, and compared these with a reference pilot system without cold recovery. In all three pilot distributions systems (DSs) our results showed an initial increase in biomass (ATP) in the biofilm phase, along with occurrence of primary colonizers (Betaproteobacteriales) and subsequently a decrease in biomass and an increasing relative abundance of other microbial groups (amoeba resisting groups; Xanthobacteraceae, Legionellales), including those responsible for EPS formation in biofilms (Sphingomonadaceae). The timeline for biofilm microbial development was different for the three pilot DSs: the higher the temperature, the faster the development took place. With respect to the water phase within the three pilot DSs, major microbial contributions came from the feed water (17–100%) and unkown sources (2–80%). Random contributions of biofilm (0–70%) were seen between day 7–77. During this time period six-fold higher ATP concentration (7–11 ng/l) and two-fold higher numbers of high nucleic acid cells (5.20–5.80 × 104 cells/ml) were also observed in the effluent water from all three pilot DSs, compared to the feed water. At the end of the experimental period the microbial composition of effluent water from three pilot DSs revealed no differences, except the presence of a biofilm related microbial group (Sphingomonadaceae), within all three DSs compared to the feed water. In the biofilm phase higher temperatures initiated the growth of primary colonizing bacteria but this did not lead to differences in microbial diversity and composition at the end of the experimental period. Hence, we propose that the microbiological water quality of DWDSs with cold recovery should be monitored more frequently during the first 2–3 months of operation.