Background Clinical team training has been advocated as a means to improve delivery care, and failed extractions is a suggested variable for clinical audit in instrumental vaginal delivery. Other activities may also have intended or unintended effects on care processes or outcomes. Methods We retrospectively observed 1074 mid and low vacuum extraction deliveries during three time periods (prevalence periods): Baseline (period 0), implemented team training (period 1 and 2) and monitoring of traction force during vacuum extraction (period 2). Our primary outcome was failed extraction followed by emergency cesarean section or obstetric forceps delivery. Results The prevalence proportion (relative risk) of failed extraction decreased significantly after implementation of team training, from 19% (period 0) to 8 % (period 1), corresponding to a relative risk of 0.48 [0.26–0.87]. The secondary procedural outcome complicated delivery (duration > 15 min or number of pulls > 6, or cup detachment > 1) was decreased in period 2 compared to period 1, RR 0.42 [0.23–0.76]. Secondary clinical (neonatal) outcome were not affected. Conclusion Clinically based educational efforts and increased monitoring improved procedural outcome without improving neonatal outcome. The study design has inherent limitations in making causal inference. Electronic supplementary material The online version of this article (10.1186/s12884-019-2257-z) contains supplementary material, which is available to authorized users.