Faecal egg counts (FEC) and FEC reduction (FECR) tests for assessing gastrointestinal nematode infection and anthelmintic resistance are rarely carried out on ruminant farms because of the cost of individual analyses. The use of composite (pooled) faecal samples in which equal amounts of faeces from several animals are mixed and analyzed together is a promising method to reduce time and costs, but few studies are available, especially with regard to the evaluation of different pool sizes and its application in FECR test. A total of 29 groups of cattle were investigated in Italy and France (9 to 20 animals/group). In each group, individual faecal samples from heifers (6 to 20 month old) were collected before (D0) and two weeks after (D14) anthelmintic treatment (ivermectin or benzimidazoles). FECs were determined individually and as pooled samples using the Mini-FLOTAC technique. Three pool sizes were used from all the animals of each group: pools of 5 or 10 individual samples or global pool. The mean FEC of individual and pooled samples were calculated as the arithmetic mean and FECR% was [1 − (arithmetic mean FEC post treatment ÷ arithmetic mean FEC pre- treatment)] x 100%. Correlations and agreements between individual and pooled results (FEC and FECR%) were estimated with the Spearman’s and the Lin’s correlation coefficients respectively (rs and ρc). High correlation and agreement coefficients were found between mean FEC of individual and pooled samples. Values were in the same range for the different pools (0.95 to 0.98 for rs and 0.97 to 0.99 for ρc) and indicated that any of pooling strategy was efficient. In contrast, %FECR calculated from individual FECs and from pooled FECs showed lower values for Spearman rs (0.67 to 0.80) and Lin’s ρc (0.49 to 0.74) meaning that composite sampling has a lack of reliability for estimating FECR%.