Subject-based reference values have been largely overlooked in veterinary medicine. These values represent longitudinal data rather than the cross-sectional data represented by standard population-based reference values. As such they provide information about biological and analytical variation. Inherent random variation of analytes around a homeostatic set point is referred to as biological variation; data on biological variation are underutilized in veterinary medicine and have multiple applications that include setting analytical goals, predicting the utility of population-based reference intervals (RIs), assessing the value of partitioning reference values, and evaluating the significance of changes in serial results. To generate these data, relatively few individuals are sampled for a short period of time. Given the difficulty of obtaining specimens from large number of healthy individuals to establish a cross-sectional RI for many veterinary species, especially exotic species, use of subject-based RIs is a practical alternative approach for the veterinary diagnostician. Furthermore, for the majority of biochemical analytes and even many hemostatic variables, population-based reference values are less sensitive than subject-based reference values for detecting pathologic changes in an individual. The focus of this review is the clinical usefulness of subject-based reference values and diagnostic implications for their use. Implementation of the concepts of biological variation, individuality, and reference change value (RCV) may allow large diagnostic laboratories to offer more sensitive reference values to assess health and detect disease.