Population models are often used to help understand the population level consequences of the impacts of offshore wind farms on seabirds. Metrics can be derived from these models in order to quantify the population level consequences of the impacts associated with the wind farm. Based on a previous review of assessments of offshore wind farms, we identified 11 metrics which have been used to assess the population level consequences of impacts from offshore wind farms on seabirds. However, seabird demography is often not quantified accurately and may be subject to significant levels of uncertainty. This leads to concerns that these metrics may be sensitive to assumptions about population trend, demographic parameters and density dependence. As a consequence, it is important to understand the extent to which conclusions based on these metrics may be influenced by the assumptions underpinning them. With this in mind, we tested the sensitivity of each metric to assumptions about population trend, life history strategy, mis-specification of demographic parameters, the incorporation of density dependence and whether the metric was derived from a stochastic or deterministic population model. Our analysis revealed that all metrics were sensitive to at least some of the assumptions underpinning them, but that some were more sensitive than others. We describe these sensitivities, indicating how to use each metric most effectively.