The paper is based on a 5-year follow-up study (2007 to 2012) of a random sample of 349 small business enterprises that operate in and around the city of Pretoria in South Africa. Data were gathered from each of the businesses on socioeconomic factors that were known to affect the long-term survival of small businesses. The objective of the study was to identify and quantify key predictors of viability and long-term survival. Pearson's chi-square tests of associations, binary logistic regression analysis and the Cox Proportional Hazards Model were used for screening of variables, and for estimating odds ratios and hazard ratios of key predictors of viability and long-term survival. The study found that 188 of the 349 businesses that took part in the study (54%) were not viable, and that the long-term survival and viability of small businesses were adversely affected by lack of entrepreneurial skills, lack of supervisory support to newly established businesses, and inability to operators running newly established businesses to acquire relevant vocational skills.