Dairy cow body size has increased over time because of breeding selection for higher milk yield, but milking stall dimensions have never been adjusted and are based on the practical experience of milking-machine manufacturers and advisory institutions. Narrow, limited milking stall dimensions might lead to behavioral changes during milking due to lack of comfort. The aim of this study was to examine the current space allowance in milking stalls on dairy farms and assess the effect of space allowance on cow behavior during milking. On 15 Swiss dairy farms, we measured clear milking stall dimensions and cow body dimensions. We calculated space ratios for length (SRlength) and width (SRwidth) by dividing the milking stall length or width by cow body length or belly width, respectively. When the space ratio was >1, we assumed that the body length or width of cow was smaller than the milking stall length or width. On each farm, 10 healthy cows were chosen for behavioral observation during 1 evening milking. We recorded rumination, elimination, and latency to enter the milking stall by direct observation. Hind leg activity was recorded using acceleration loggers. Data were analyzed using general linear mixed-effects models with farm as a random effect. Due to a strong collinearity between SRwidth and SRlength, we chose SRlength for further analysis, because it is based on skeletal characteristics. The SRlength was smallest in side-by-side parlors (1.07 ± 0.01) and largest in tandem parlors (1.18 ± 0.01). More cows had a tendency to ruminate with increasing SRlength (odds ratio: 1.8). None of hind leg activity, maximum peaks of hind leg accelerations, or latency to enter the milking stall were significantly affected by SRlength. Latency to enter the milking stall was longer for group milking parlors (side-by-side: 44.0 ± 3.2 s; herringbone: 34.3 ± 2.9 s) than for tandem parlors (19.0 ± 2.7 s). Milking parlor type had no effect on hind leg activity, maximum peaks of hind leg accelerations or rumination. The SRlength affected rumination behavior to some extent, indicating that cow comfort was positively affected by larger milking stall length. Because cow comfort is important for good milking performance, further investigations of milking stall dimensions for cow comfort and thus welfare are needed. Furthermore, the results showed that parlor type affected cow behavior, irrespective of SRlength, making future research necessary to identify the factors leading to this effect of parlor type.