Genetic analysis is a tool that has revolutionized management of threatened plant species in wild populations over the past two decades, but it has only recently been applied to managing ex situ populations. To improve conservation outcomes, botanic gardens have started to assess the extent of wild genetic diversity conserved in ex situ collections. DNA-based studies have provided updated seed sampling recommendations and species-tailored sampling guidelines. Genetic simulations have also advanced this field of study by testing a wide range of hypotheses in a controlled environment; using simulations, one can test different sampling strategies on species with pre-defined traits and number, size, and structure of populations. In this study, we expand the realism of simulations used for this purpose. We applied simulations to determine how pollen dispersal and logistical field constraints can influence the success of sampling genetic diversity in seed collections. We found that pollen flow greatly influences collection outcomes; panmictic pollen flow results in seed sets containing 21.6% more alleles than limited and 48.6% more alleles than highly limited pollen flow, though this impact diminishes for very large numbers of maternal plants sampled. Simulations of realistic seed sampling (sampling more seed from some plants and less from others) showed a relatively minor impact (e.g., < 2 .5%) on genetic diversity conserved compared to ideal sampling (uniform sampling across all maternal plants). We conclude that seed samplers and future simulation projects must consider limited pollen flow. Meanwhile, seed samplers can be flexible with their sampling in the field, as long as the main guideline “maximize the number of maternal plants” is followed. Simulations remain a fruitful way to advance ex situ sampling guidelines.