This paper discusses the physical and social attributes of urban density on the basis of empirical evidence gathered through an interdisciplinary research project carried out in London by a research team at the London School of Economics between December 2002 and July 2004. The background for the study was provided by the current British urban policy agenda, which -in short- seeks to create sustainable communities through a “compact city” approach, postulating the efficient use of urban land through building at higher densities and the creation of more inclusive communities by promoting mix-use and mix-tenure urban neighbourhoods and well connected by good public transport. However, this “ideal vision” promoted by compact urban development advocates has been strongly contended in a number of academic and public debates on the relative merits of urban density. Therefore, the study aimed to investigate whether high density is feasible in London without compromising quality of life and, if so, for whom and under what circumstances it can work well. To answer this question, the study sought to understand which attributes are associated with high density neighbourhoods. A case study of five areas in London was conducted, all of which have residential densities above the London average. The selection criteria included socially and ethnically diverse areas and a diverse range of building typologies in each of them. The study applied a range of interdisciplinary research techniques, namely: spatial analysis, analysis of census data, qualitative interviews with residents and key local actors on their perceptions of density, and a quantitative survey on residential satisfaction in each area. Overall, findings point out the need to re-evaluate Density as a planning tool. The study shows that higher densities support diverse urban communities at different stages of people’s life-cycles. A typology of these groups in connection to urban density is proposed. Furthermore, the paper discusses a number of more complex interrelationships between density and other parameters, such as accessibility, internal occupancy levels, car use, parking, open space, distribution of facilities, etc.