BACKGROUND: For working adults, about one-third of energy is consumed in the workplace making this an important context in which to reduce energy intake to tackle obesity. The aims of the current study were first, to identify barriers to the feasibility and acceptability of implementing calorie labelling in preparation for a larger trial, and second, to estimate the potential impact of calorie labelling on energy purchased in worksite cafeterias. METHODS: Six worksite cafeterias were randomised to the intervention starting at one of six fortnightly periods, using a stepped wedge design. The trial was conducted between August and December 2016, across 17 study weeks. The intervention comprised labelling all cafeteria products for which such information was available with their calorie content (e.g. "250 Calories") displayed in the same font style and size as for price. A post-intervention survey with cafeteria patrons and interviews with managers and caterers were used to assess the feasibility and acceptability of the intervention. Intervention impact was assessed using generalised linear mixed modelling. The primary outcome was the total energy (kcal) purchased from intervention items in each cafeteria each day. RESULTS: Recruitment and retention of worksite cafeterias proved feasible, with post-intervention feedback suggesting high levels of intervention acceptability. Several barriers to intervention implementation were identified, including chefs' discretion at implementing recipes and the manual recording of sales data. There was no overall effect of the intervention: -0.4% (95%CI -3.8 to 2.9, p = .803). One site showed a statistically significant effect of the intervention, with an estimated 6.6% reduction (95%CI -12.9 to - 0.3, p = .044) in energy purchased in the day following the introduction of calorie labelling, an effect that diminished over time. The remaining five sites did not show robust changes in energy purchased when calorie labelling was introduced. CONCLUSIONS: A calorie labelling intervention was acceptable to both cafeteria operators and customers. The predicted effect of labelling to reduce energy purchased was only evident at one out of six sites studied. Before progressing to a full trial, the calorie labelling intervention needs to be optimised, and a number of operational issues resolved. TRIAL REGISTRATION: ISRCTN52923504 ; Registered: 22/09/2016; retrospectively registered.