Analytical goals may be defined as those standards of performance required to facilitate optimal patient care. Review of the considerable amount of published work on glucose analyses allows delineation of the following current goals: For plasma glucose: within-laboratory between-batch imprecision--CV less than or equal to 2.2%; inaccuracy - no bias; linearity - 1 to 28 mmol/L; detection limit - 1 mmol/L; turnaround time - 30 min (in emergency situations). For urine glucose (quantitative): within-laboratory between-batch imprecision - SD less than or equal to 0.1 mmol/L; inaccuracy - no bias. For urine glucose (qualitative or semi-quantitative): negative - should be negative; positive - should be positive +/- one positive colour block; detection limit - 2.8 mmol/L. Moreover, each laboratory should develop goals for the many pre-analytical factors affecting glucose analyses. While the above goals are not always achieved in current practice, they are attainable and are worthy targets to strive to reach. The general principles discussed here for glucose analyses are applicable to other analytes of interest to the clinical biochemist.