Objective: There is a growing interest in how people living with dementia may achieve good outcomes and be resilient despite their health challenges. Understanding what might be important for resilience in this population is largely untested theory. Methods: The analysis draws a subsample with cognitive impairment (N = 579) from two waves of the Cognitive Function and Ageing Studies Wales study, a nationally representative study of community‐dwelling people aged 65+ in Wales. We constructed a measure of mental health resilience (MHR) defined as no depression, no anxiety and high well‐being. Drawing on a resilience framework, we tested univariate and cumulative effects models of the factors that enable MHR, and then examined whether MHR is important for reducing loneliness over time. Results: Across both waves of data 22% (n = 121) met the criteria for MHR. The cumulative effects model found the odds of MHR were greater for male gender, higher self‐esteem, greater social resources and no subjective memory complaints. Controlling for these significant predictors, MHR significantly predicted lower total and sub‐scale scores for loneliness at wave 2. Sensitivity analysis shows these effects held at lower levels of cognitive function when the Mini‐Mental State Examination score was <25, but not at <23. Conclusions: This paper addresses a gap in research regarding the conceptualisation and measurement of resilience when facing cognitive impairment. Understanding what aspects of a person's life might enable good mental health despite cognitive impairment—to be resilient—could inform effective strategies for friends and families, along with health, and social policy and practice.