Urban neighbourhoods are defined as much by their commercial character as their residential; retail services not only provide material needs for those living nearby, but less-tangible social and cultural capital as well. It is reasonable to expect, then, that excessive churn in these businesses can threaten the stability of a neighbourhood. Using a longitudinal data set on mixed-use neighbourhoods in New York City, we test whether or not neighbourhoods of varying circumstances and characteristics experience different degrees and types of retail turnover. Results suggest that there are meaningful differences in retail turnover across neighbourhoods. Retail turnover is directly associated with the type of business activity, commercial infrastructure and the neighbourhood’s consumer profile. However, when all three sets of factors are considered simultaneously in a regression analysis, consumer-related characteristics explain turnover more than those related to the local commercial environment. Specifically, businesses that provide necessity and more frequently consumed goods/services are more stable and chain establishments are more likely to venture into markets with some housing price discounts, growth potential and possibly less organised opposition. Neighbourhoods with less (and more heterogeneous) general retail (as opposed to food service) concentration, as well as bigger businesses, are more stable. More importantly, bigger households and higher shares of white residents are most strongly associated with less retail churn, and population growth is the strongest predictor of more turnover.