Purpose <p>There exist many ecosystem service impact models in the field of life cycle assessment. Increased interest in model development has come alongside increased data quality and methodological detail. However, the methods implemented thus far are service-specific and non-comparable, hindering their widespread adoption. This review assesses existing ecosystem service impact models in life cycle assessment, and provides a set of overarching discussion points to begin the process of harmonization for these models moving forward.</p> Methods <p>We conducted an expert-opinion literature review of existing models, looking specifically at models which fall within the methodological scope of life cycle assessment. We then grouped these models according to common methodological approaches we identified in the literature. Finally, we define unifying characteristics and highlight pros and cons for each approach, with the aim of guiding the field towards specific best practices.</p> Results and discussion <p>Our review highlights large differences in methodologies across existing ecosystem service impact models. We grouped existing ecosystem service models into three overarching methodological approaches—land use impact, extended ecosystem quality, and inventory-based models—signifying some common understanding around ecosystem service model development. Nonetheless, we see large methodological differences across services modelled, impact categories addressed in each model, ecological realm coverage, indicators used to measure service impacts, and in how impact models fit into the standard supply-demand framework of ecosystem services. Such inconsistencies demonstrate the complexity of ecosystem service modelling, as well as the range of applications for state-of-the-art impact models.</p> Conclusions <p>In order for life cycle assessment to comprehensively cover ecosystem service impacts, in all ecological realms, model developers need a much more unified framework moving forward. In order to achieve such a framework, we need to address not only some basic methodological inconsistencies, but also to more clearly define the purposes for including ecosystem services in life cycle assessment. We suggest the creation of a common set of guidelines for future model developers.</p>