This study examines the impact of both gender and gendered agglomeration - viewed as a circumstance in which firms with a given gender composition tend to establish themselves in specific cities - on innovation dynamics in terms of knowledge spillovers. Our framework considers gender composition of ownership, management, and workforce, as well as various types of innovations including product, process, marketing, aggregate innovation, and R&D. The paper employs an ordered probit and a two-stage Heckman probit models with instrumental variable to address both selection and reverse causality issues between gender and innovation. Using data from the 2016 World Bank Enterprise Survey on 3,921 African firms, this study yields four main new findings: (i) both male and female agglomeration influence positively the probability of innovating, (ii) the positive effect of, having a woman as manager, having fewer woman among both the firm's owners and workforce, is contingent upon the variety of innovations within the firm, (iii) in term of human capital, the nature of the agglomeration does not matter for innovation, and (iv) R&D expenditures decisions are made by the firm’s ownership rather than its management.