Joint R&D&I projects and programs increase performance but entail project-related risks; namely, concurrent projects may share resources, delays may be inherited by later projects, and costs may overrun. With a project portfolio, one can better plan budgets, timelines, and risks, manage the resource-sharing of parallel projects, and map content-related projects as one program. However, large ... [Show full abstract] (publicly funded) research programs are rarely designed and managed as portfolios. This paper discerns the structure of the European Union-funded 7th Framework Program from a project management perspective. The proposed method estimates the structure based on the content, collaboration, and duration, which enables very accurate mapping of a naturally evolved, unstructured portfolio using limited publicly available data. We tested the sensitivity of overruns/delays, relative costs, and outputs to changes in the distribution of single projects, multi-projects, and programs. The results indicate that supporting programs leads to an improvement in performance, while the duration increases only slightly. The relative costs and output can be improved by moderately increasing the proportion of multi-projects and programs without the risk of overrun or delay. Although we used data from an unstructured portfolio, we believe that the results are valid for R&D&I portfolios in general. To address research policy, we suggest that understanding framework programs as portfolios of projects and managing their structure support collaboration and increase output with only a marginal impact on the cost and duration.