Quality control in mass reared arthropod natural enemies is a multi faceted concern; however, there is always the need to assess one or more quantitative traits. Because it is usually not possible to measure all the individuals in a population, sampling is necessary. It often suffices to classify traits of interest as either greater or less than some acceptable level and when this is the case, the performance of a sampling procedure can be judged using an operating characteristic or OC. The OC can be thought of as a series of points, each of which provides the probability that a sample will lead to classifying the trait of interest as less than or equal to some predetermined critical value. It is common when sampling biological populations that sample variances are large and relatively large sample sizes are required to gain acceptable sample precision. For mass-reared arthropod natural enemies, a different pattern seems to apply. For these organisms, a manufacturing paradigm of quality control seems most appropriate and sample sizes can often be quite small and still yield acceptable precision. This pattern occurs because variation in the parameter of interest has, in the systems we have examined, varied more among batches of natural enemies produced by one or more companies and this source of variation can be omitted from consideration. Examples include classifying the abundance of phytoseiid mites in shipping containers, classifying the numbers of parasitic wasps on distribution media, and assessing the flight capability of parasitic wasps.