There has been growing interest in approaches to business training that incorporate insights from psychology to develop soft skills associated with successful entrepreneurship. The empirical evidence on the causal effects of these approaches on entrepreneurs’ business outcomes is encouraging, but still not substantial enough to be conclusive. This study contributes to this literature by designing and evaluating two training programs, which are adapted to the Jamaican context. The first program provides soft-skills training on personal initiative, including the development of a proactive mindset and perseverance after setbacks. The second program combines soft-skills training on personal initiative with traditional training on hard skills aimed at changing business practices. Both programs are evaluated using a randomized controlled trial involving 945 entrepreneurs in Jamaica. Entrepreneurs are randomly assigned in equal proportion to one of the two training programs or to a control group. The research develops three survey instruments to collect information from entrepreneurs: a baseline survey, a short-term follow-up survey conducted 3 months after the intervention, and a second follow-up survey conducted 12 months after the intervention. Findings indicate statistically significant effects of the intensive soft-skills training, but not of the training combining soft and hard skills, on business outcomes in the short-term survey. The analysis of the data suggests that the main channel through which the intensive soft-skills training improves short-term business outcomes is an increased adoption of business practices. The positive short-term effects of the soft-skills training are concentrated among men and are not significant for female entrepreneurs. Neither the effects on business practices nor those on business outcomes are statistically significant in the second follow-up survey. However, the soft-skills training has persistent effects on targeted soft skills, which are measured with both self-reported and incentivized measures. Theoretical and practical implications of these findings are discussed in the paper.