Objective: To critically review the literature regarding workplace breast-feeding interventions and to assess their impact on breast-feeding indicators. Design: A systematic review and meta-analysis was conducted. Electronic searches for workplace intervention studies to support breast-feeding, without restriction on language or study design, were performed in PubMed, CENTRAL, CINAHL, Embase, Web of Science, Business Source Complete, ProQuest-Sociology and ProQuest-Social Science to 13 April 2020. A meta-analysis of the pooled effect of the programmes on breast-feeding indicators was conducted. Results: The search identified 10 215 articles; fourteen studies across eighteen publications met eligibility criteria. Programmes were delivered in the USA (n 10), Turkey (n 2), Thailand (n 1) or Taiwan (n 1). There were no randomised controlled trials. The pooled OR for exclusive breast-feeding at 3 or 6 months for participants v. non-participants of three non-randomised controlled studies was 3·21 (95 % CI 1·70, 6·06, I2 = 22 %). Despite high heterogeneity, other pooled outcomes were consistently in a positive direction with acceptable CI. Pooled mean duration of breast-feeding for five single-arm studies was 9·16 months (95 % CI 8·25, 10·07). Pooled proportion of breast-feeding at 6 months for six single-arm studies was 0·76 (95 % CI 0·66, 0·84) and breast-feeding at 12 months for three single-arm studies was 0·41 (95 % CI 0·22, 0·62). Most programmes were targeted at mothers; two were targeted at expectant fathers. Conclusions: workplace programmes may be effective in promoting breast-feeding among employed mothers and partners of employed fathers. However, no randomised controlled trials were identified, and better-quality research on workplace interventions to improve breast-feeding is needed.