The many antihyperglycemic preparations are best used for type 2 diabetes in a logical sequence, using combinations of agents, with clear targets for glycemic control. On the basis of long familiarity, proven benefit and known side effects, and low cost, the sulfonylureas, metformin, and insulin still deserve to be the standard treatments. As shown in the central shaded area of Fig. 4, standard treatment begins with monotherapy and progresses to oral combination therapy and then to two oral agents plus basal insulin. Several triggers for deviation from the standard methods are identified (see Fig. 4). The incidence of each of the conditions that require early individualized treatment has not been studied, but it seems reasonable to estimate no more than 10% each for a strongly symptomatic presentation, inability to use a sulfonylurea or metformin, inability to use insulin, or an early need for prandial therapy. If this estimate is correct, approximately two thirds of patients who are diagnosed with type 2 diabetes should do well with standard therapy for up to 10 years using the standard methods shown. Eventually, many more will need individualized treatment to maintain glycemic control. This scheme is certain to evolve as further information on the nonglycemic benefits (or hazards) of the various therapies appears and as new treatments are released. Notably, agents that mimic or potentiate the effects of gastrointestinal peptides, such as amylin and GLP- 1 analogues and dipeptidyl peptidase IV inhibitors, are likely to alter the current algorithm. For now, systematic application of the scheme (see Fig. 4) should improve the success of treatment greatly from its currently disappointing level.