Purpose The purpose of this study is to improve the understanding of the impact of innovation vouchers (IVs) by assessing the additionality of IV instruments supporting business-research collaboration in Lithuania. Design/methodology/approach The approach relies on studying potential input, behavioural and output additionalities created by IVs. Adopting this perspective, the paper outlines a theoretical framework that indicates the expected contribution of IV schemes and empirically tests its implications. The model is tested by analysing the Inno-vouchers LT instrument implemented in Lithuania in 2012–2014. The study relies on a counterfactual impact evaluation (CIE) and descriptive statistical analysis of survey data of the applicants for IVs. Findings The study findings are structured around three types of additionalities. Evidence on input additionality of IV projects provides mixed results, but a significant share of projects included private investment that would not exist without Inno-vouchers LT. Yet, many beneficiaries would also have implemented projects without public support. Both survey data analysis and CIE provide evidence of behavioural additionality, albeit limited. They strongly support the hypothesis that IVs increase companies’ willingness to cooperate with research organisations. Finally, evidence on output additionality is ambiguous. Long-term or indirect output additionality (measured by employment and turnover) seems to be non-existent according to CIE results. Originality/value Despite the policy relevance and use of IVs instruments, evidence on the effectiveness is limited. The authors present findings on IVs additionality and suggest avenues for further research.