Efficient milking systems, in terms of labour demand, capital investment and cow udder health are critical to successful dairy herd expansion. The objective of this study was to establish the effect of two primary influencing factors on efficient milking performance, i.e. parlour size (number of milking units) and pre-milking routine (full and nil) of spring-calved cows, in a single-operator side-by-side, swing-over milking parlour. Efficiency parameters investigated in a 5×2 factorial design included milk-flow and yield, row time, over-milking duration and operator idle time. Five combinations of parlour size (14, 18, 22, 26 and 30 milking units) each with two different pre-milking routines (Full: spray, strip, wipe, attach clusters, and Nil: attach clusters) were examined with one milking operator. The trial was carried out over 40 milking sessions and cows (up to 120) were randomly assigned to groups (n=14, 18, 22, 26 or 30) before each milking session. Row within a milking session was the experimental unit. The experiment was carried out at both peak and late lactation. The data were analysed with a mixed model using GenStat 13.2. The full pre-milking routine reduced time to milk let-down and milking time, increased average flow rate but did not affect milk yield. As milking unit number increased, the duration of over-milking (defined as time at milk flow rate <0·2 kg/min) increased more with a full compared with nil routine. Thus, the use of pre-milking preparation decreased milking time per cow but as parlour size increased, milking row times, as well as the proportion of cows that were over-milked, also increased, thereby reducing overall efficiency. These results have implications for milking management in single-operator swing-over, tandem and rotary parlours with seasonally calved herds.