Tremblay et al. (2016; J. Dairy Sci. 99:3824–3837) analyzed many data from North American dairy farms with automatic milking systems (AMS) for (risk) factors associated with increased milk production per cow and per robot per day. The final data set contained 54,065 observations from 529 farms collected from 2011 to 2014. The dependent variables of milk production per cow per day and milk production per robot per day were used in multivariable generalized mixed linear regression models with farm IDs as random effects and 2-way interactions selected using forward selection (Tremblay et al., 2016). Many variables and combinations of variables (interaction terms) were found to be significantly associated with the outcomes including traffic type (how cows are allowed to move among the AMS, feeding alley and lying stalls) and the number of AMS robots per pen (Tremblay et al., 2016). On average, free traffic type produced 67.21 kg (CI: 48.6–86.0) more milk per robot per day than Guided/Select; having one AMS robot per pen produced 59.82 kg (CI: 50.4–68.8) less milk per robot per day than having 2 AMS robot per pen; a AMS farm will produce 20.5 kg (CI: 17.0–24.4) of milk per robot more after 4+ years since installation compared with within the first year since start-up; Holsteins produce 216.71 kg (CI: 193.2–239.9) more milk per robot per day than Jerseys (not energy corrected) (Tremblay et al., 2016). Knowing how different factors affect milk production in AMS will help guide current and future AMS producers looking to maximize milk production and compare their performance to other AMS farms (Tremblay et al., 2016).