Background: Cochrane reviews are not well designed for readers who want to locate the key information quickly. ‘Summary of findings’ (SoF) tables are intended to address this problem. Objectives: To develop a template for SoF tables that summarize the main findings of Cochrane reviews, including what is known about important benefits and harms, what is not known about these, and how sure we can be about this information. Methods: Our starting point was the SoF table developed by the GRADE Working Group. First, we iteratively redesigned the SoF table based on user testing to ensure that the format was understandable, usable and useful for health professionals (or others) with some knowledge of systematic reviews. We also consulted an Advisory Group consisting of over 40 people with a range of different stakeholder roles in The Cochrane Collaboration to ensure that the information was not oversimplified or imprecise. We evaluated the effect of the improved SoF table on users’ understanding of reviews in two randomised trials comparing Cochrane reviews with and without tables. Results: We arrived at a format that eliminated most of the problems we found through user testing. In the second trial (N = 33), there were large differences in the proportion that answered questions about the risk in the control group (44% vs. 93%, p = 0.0003) and the risk in the intervention group (11 vs. 87%, p 0 0.0001) correctly; and in the proportion that agreed or strongly agreed that it was easy to find information about the quality of evidence for the main outcomes (24% vs. 73%, p = 0.001). Eighty-eight per cent agreed or strongly agreed that Cochrane reviews should have SoF tables, and 84% agreed with the format. Conclusions: SoF tables can help make Cochrane reviews more useful for readers with some previous knowledge of systematic reviews. Issues needing further work include improved presentation of continuous outcomes and variations in table complexity for different audiences.