Four milking regimes using automatic cluster removers (ACR) were tested over a 19-week period, from mid to late lactation. Each treatment group consisted of 16 slow-milking cows. The milking regimes used were: an ACR setting of 200 ml/min (Control); raised ACR setting from 200 to 500 in steps of 100 ml/min (Raised ACR); raised ACR as above in conjunction with pre-milking teat stimulation (Raised ACR+Stim); and terminating the milking when an ACR threshold of 200 ml/min was reached or when a predetermined maximum milking duration was reached (Timer). All incremental treatments were applied in blocks of 6 or 7 weeks duration. ACR thresholds were raised from 200 to 500 ml/min without observed loss of milk production when compared with controls. However, even up to an ACR setting of 500 ml/min there was little reduction in the group milking duration even when used with teat stimulation. In contrast, the Timer treatment resulted in a 34% reduction of the maximum milking duration for the group without significant loss of milk yield. For all groups, including Control, strip yield was occasionally very high and highly variable. Willingness of cows to enter the milking platform, behaviour during milking, teat condition and incidence of mastitis were similar for all treatment groups. The results indicated that simple truncation of milking at a predetermined maximum duration could be a most potent and inexpensive method of milking a herd more quickly. Such a method could be employed by using a simple timer in any dairy regardless of the level of sophistication of the milking system.