Scenario analyses have been used in multiple science-policy assessments to better understand complex plausible futures. Scenario archetype approaches are based on the fact that many future scenarios have similar underlying storylines, assumptions, and trends in drivers of change, which allows for grouping of scenarios into typologies, or archetypes, facilitating comparisons between a large range of studies. The use of scenario archetypes in environmental assessments foregrounds important policy questions and can be used to codesign interventions tackling future sustainability issues. Recently, scenario archetypes were used in four regional assessments and one ongoing global assessment within the Intergovernmental Science-Policy Platform for Biodiversity and Ecosystem Services (IPBES). The aim of these assessments was to provide decision makers with policy-relevant knowledge about the state of biodiversity, ecosystems, and the contributions they provide to people. This paper reflects on the usefulness of the scenario archetype approach within science-policy processes, drawing on the experience from the IPBES assessments. Using a thematic analysis of (a) survey data collected from experts involved in the archetype analyses across IPBES assessments, (b) notes from IPBES workshops, and (c) regional assessment chapter texts, we synthesize the benefits, challenges, and frontiers of applying the scenario archetype approach in a science-policy process. Scenario archetypes were perceived to allow syntheses of large amounts of information for scientific, practice-, and policy-related purposes, streamline key messages from multiple scenario studies, and facilitate communication of them to end users. In terms of challenges, they were perceived as subjective in their interpretation, oversimplifying information, having a limited applicability across scales, and concealing contextual information and novel narratives. Finally, our results highlight what methodologies, applications, and frontiers in archetype-based research should be explored in the future. These advances can assist the design of future large-scale sustainability-related assessment processes, aiming to better support decisions and interventions for equitable and sustainable futures.Nadia Sitas, Ryan Blanchard, and Patrick O'Farrell were supported by SwedBio at Stockholm Resilience Centre funded by the Swedish International Development Cooperation Agency (SIDA). Zuzana Harmáčková and Reinette Biggs were supported by the GRAID programme funded by the Swedish International Development Agency (SIDA). Reinette Biggs was also supported by the South African Research Chairs Initiative (SARChI) (grant 98766) and the Swedish Research Council (grant 621-2014-5137). Eefje den Belder was funded by the Ministry of Agriculture, Nature and Food Quality of the Netherlands. Support to Paula A. Harrison was provided by UK Department of Environment, Food and Rural Affairs and the EU-funded IMPRESSIONS project (Grant Agreement 603416). We would also like to thank our reviewers for their constructive comments in shaping this paper.