Mental health services worldwide are under strain from a combination of unprecedented demand, workforce reconfigurations, and government austerity measures. There has been relatively little research or policy focus on the impact of staffing and skill mix on safety and quality in mental health services leaving a considerable evidence gap. Given that workforce is the primary therapeutic intervention in secondary mental health care this constitutes a major deficit. This study aimed to explore the impact of staffing and skill mix on safety and quality of care in mental health inpatient and community services. Exploratory, qualitative methodology; purposive sampling. Inpatient and community mental health services in the United Kingdom. 21 staff (including nurses, occupational therapists, psychiatrists, social workers, and care co-ordinators) currently working in mental health services. We conducted semi-structured telephone interviews with a purposive sample of staff recruited via social media. We asked participants to describe the staffing and skill mix in their service; to reflect on how staffing decisions and/or policy affected safety and patient care; and for their views of what a well-staffed ward/service would look like. We conducted thematic analysis of the interview transcripts. The participants in this study considered safestaffing to require more than having 'enough' staff and offered multiple explanations of how staffing and skill mix can impact on the safety and quality of mental health care. From their accounts, we identified how the problem of 'understaffing' is self-perpetuating and cyclical and how its features interact and culminate in unsafe care. We conceptualised the relationship between staffing and safety as a 'vicious cycle of unsafestaffing' which comprised: (1) understaffing (the depletion of resources for safe care provision); (2) chronic understaffing (conditions resulting from and exacerbating understaffing); and, (3) unsafestaffing (the qualities of staffing that compromise staff capacity to provide safe care). Continued policy focus on safestaffing is clearly warranted, especially in mental health as staffing constitutes both the principal cost and main therapeutic driver of care. This paper provides compelling reasons to look beyond regulating staff numbers alone, and to consider staff morale, burden and the cyclical nature of attrition to ensure the delivery of high quality, safe and effective services. Future research should investigate other mechanisms via which staffing impacts on safety in mental health settings. Copyright © 2019. Published by Elsevier Ltd.