While many commentators recognize the potential for Social Impact Bonds (SIBs) to encourage innovation, empirical evidence is less clear cut. We argue that for SIBs to realize their full potential as incubators of innovation they needed to incorporate a stronger element of co-creation and strengths-based working, and suggest some accompanying characteristics of such SIBs. We analyze four UK SIBs as case studies that exhibit these characteristics. We find that within the cases strengths-based service delivery models were successfully implemented through SIBs; alongside this we found extensive evidence of co-production but limited evidence of co-creation. Strengths-based working, including co-production, helped the SIBs to catalyze early stage innovation. We identify several elements of SIB design which were instrumental in supporting strengths-based practices and could also support co-creation: greater autonomy for service providers; shifting risk to investors; use of a rate card; and long-term, flexible funding. Our findings challenge those who are skeptical about the potential for SIBs to provide a setting in which people who use services and frontline service providers are empowered to re-engineer public services and contest cur- rent systems.