This study investigates systematically and simultaneously a wide range of influences on workplace productivity loss. Workplace productivity has been widely discussed in literature for many years, yet most of the existing studies focus on a narrow pathway of effects; this can potentially result in overestimating the examined influences due to capturing unobserved effects of omitted variables. In this study we examine various productivity determinants (workers' characteristics, lifestyle, commuting, health and wellbeing, and job and workplace environment) in a single combined model, after investigating their interrelations, to re-assess some of the previous findings and provide new insights into the subject. This is made possible through utilising a unique and extensive dataset of nearly 30,000 employees in the UK collected in 2017 and developing a comprehensive Structural Equation Model (SEM). Our results generally confirm the previous findings but also highlight the necessity to consider a broad range of factors as some of the initially strong influences (e.g. of work engagement) become statistically insignificant once additional variables are introduced into the model. Overall, personal characteristics, particularly physical and mental health, as well as job characteristics such as stress at work explain most of the variance in productivity loss, while lifestyle or working patterns are less important.