Species delimitation is the process of identifying species-level biological diversity. Recent advances in genetic data collection have led to an explosion in the number and variety of methods for species delimitation. However, most studies use only a few of these methods, often without justification. Because the parameter space relevant to species delimitation is vast, each method makes simplifying assumptions that may not hold in all cases. Researchers should apply a wide range of methods and trust results that are congruent across methods. Incongruence may indicate differences in the power to detect cryptic lineages or violations of method assumptions. Inferences should be conservative, as failing to delimit species is often better than falsely delimiting entities that do not represent actual evolutionary lineages. Species delimitation methods are applied to data from sexually reproducing organisms. Existing methods range from nonparametric to highly parameterized models. An ideal method would use a parametric approach, as most methods operate by fitting models of historical diversification to data. However, the parameter space relevant to species delimitation is large, and existing tools limit exploration to a subset of this space by making simplifying assumptions. Researchers should use a wide range of methods and trust results that are congruent across methods. Simulation studies can help evaluate method accuracy, but results are conditional on specific data attributes. For example, Camargo et al. (2012) found that BPP and a custom ABC method were more accurate than spedeSTEM in their simulations, but spedeSTEM was less accurate in other simulations. Species delimitation methods can be categorized as discovery or validation approaches. Discovery methods do not require sample assignment, while validation methods do. Many researchers use combinations of these approaches to delimit species in complex systems. For example, Leaché & Fujita (2010) used a two-stage approach, first using Structurama to assign individuals to groups and then validating these groups with BPP. This approach helps mitigate the shortcomings of individual methods. Validation approaches like BPP and spedeSTEM have limitations. BPP depends on the accuracy of the guide tree, while spedeSTEM assumes that all shared polymorphism is due to unsorted ancestral polymorphism. Inaccurate guide trees can lead to incorrect species delimitations. Researchers should use multiple methods and consider the results for congruence. Incongruence may indicate issues with one or more methods. Species delimitation should consider genetic and non-genetic data, such as morphology and behavior. Integrating these data can improve the accuracy of species delimitation. Species concepts, such as the general lineage concept, have influenced recent approaches to species delimitation. However, it is important to define a species concept when reporting species delimitation studies. Species delimitation is a vital part of evolutionary biology, bridging phylogenetics and population genetics. Researchers should use a wide range of methods and trust results that areSpecies delimitation is the process of identifying species-level biological diversity. Recent advances in genetic data collection have led to an explosion in the number and variety of methods for species delimitation. However, most studies use only a few of these methods, often without justification. Because the parameter space relevant to species delimitation is vast, each method makes simplifying assumptions that may not hold in all cases. Researchers should apply a wide range of methods and trust results that are congruent across methods. Incongruence may indicate differences in the power to detect cryptic lineages or violations of method assumptions. Inferences should be conservative, as failing to delimit species is often better than falsely delimiting entities that do not represent actual evolutionary lineages. Species delimitation methods are applied to data from sexually reproducing organisms. Existing methods range from nonparametric to highly parameterized models. An ideal method would use a parametric approach, as most methods operate by fitting models of historical diversification to data. However, the parameter space relevant to species delimitation is large, and existing tools limit exploration to a subset of this space by making simplifying assumptions. Researchers should use a wide range of methods and trust results that are congruent across methods. Simulation studies can help evaluate method accuracy, but results are conditional on specific data attributes. For example, Camargo et al. (2012) found that BPP and a custom ABC method were more accurate than spedeSTEM in their simulations, but spedeSTEM was less accurate in other simulations. Species delimitation methods can be categorized as discovery or validation approaches. Discovery methods do not require sample assignment, while validation methods do. Many researchers use combinations of these approaches to delimit species in complex systems. For example, Leaché & Fujita (2010) used a two-stage approach, first using Structurama to assign individuals to groups and then validating these groups with BPP. This approach helps mitigate the shortcomings of individual methods. Validation approaches like BPP and spedeSTEM have limitations. BPP depends on the accuracy of the guide tree, while spedeSTEM assumes that all shared polymorphism is due to unsorted ancestral polymorphism. Inaccurate guide trees can lead to incorrect species delimitations. Researchers should use multiple methods and consider the results for congruence. Incongruence may indicate issues with one or more methods. Species delimitation should consider genetic and non-genetic data, such as morphology and behavior. Integrating these data can improve the accuracy of species delimitation. Species concepts, such as the general lineage concept, have influenced recent approaches to species delimitation. However, it is important to define a species concept when reporting species delimitation studies. Species delimitation is a vital part of evolutionary biology, bridging phylogenetics and population genetics. Researchers should use a wide range of methods and trust results that are