Few studies have examined the impact of 'sit less, move more' interventions on workplace performance. This study assessed the short and mid-term impacts of and patterns of change within, a 19-week workplace web-based intervention (Walk@WorkSpain; W@WS; 2010-11) on employees´ presenteeism, mental well-being and lost work performance. A site randomised control trial recruited employees at six Spanish university campuses (n = 264; 42 ± 10 years; 171 female), assigned by worksite and campus to an Intervention (IG; used W@WS; n = 129; 87 female) or an active Comparison group (A-CG; pedometer, paper diary and self-reported sitting time; n = 135; 84 female). A linear mixed model assessed changes between the baseline, ramping (8 weeks), maintenance (11 weeks) and follow-up (two months) phases for the IG versus A-CG on (i) % of lost work productivity (Work Limitations Questionnaire; WLQ); (ii) three scales for presenteeism (WLQ) assessing difficulty meeting scheduling demands (Time), performing cognitive and inter-personal tasks (Mental-Interpersonal) and decrements in meeting the quantity, quality and timeliness of completed work (Output); and (iii) mental well-being (Warwick-Edinburgh Mental Well-being Scale). T-tests assessed differences between groups for changes on the main outcomes. In the IG, a multivariate logistic regression model identified patterns of response according to baseline socio-demographic variables, physical activity and sitting time. There was a significant 2 (group) × 2 (program time points) interaction for the Time (F [3]=8.69, p = 0.005), Mental-Interpersonal (F [3]=10.01, p = 0.0185), Output scales for presenteeism (F [3]=8.56, p = 0.0357), and for % of lost work performance (F [3]=10.31, p = 0.0161). Presenteeism and lost performance rose significantly in both groups across all study time points; after baseline performance was consistently better in the IG than in the A-CG. Better performance was linked to employees being more active (Time, p = 0.041) and younger (Mental-interpersonal, p = 0.057; Output, p = 0.017). Higher total sitting time during nonworking days (Mental-interpersonal, p = 0.019) and lower sitting time during workdays (WLQ Index, p = 0.013) also improved performance. Versus an active comparison condition, a 'sit less, move more` workplace intervention effectively reduced an array of markers of lost workday productivity. NCT02960750 ; Date of registration: 07/11/2016.