This study is concerned with how impact from research and innovation (R&I) programmes is accounted for in impact evaluation reports. Establishing causal links between a research funding instrument and different effects, poses well known methodological difficulties. In the light of such challenges, textual accounts about causal links ought to be carefully written. Nevertheless, impact evaluation reports have a tendency towards unwarranted simplification as far as impact inferences are concerned. In this study, we illustrate how such simplifications – versions of the narrative device ellipsis – are accomplished. Using examples from three Swedish impact evaluation reports, we focus on the constituent components of longer impact accounts, that of the impact argument, to analyze the various ways that impact is narratively achieved through simplification. We believe this analysis can contribute to the methodology of impact evaluation, as well as spread light on some the difficulties in the historiography of innovation in general.