When a new drug is developed, one of the first requirements is to establish the correct dose. Unfortunately, in dose-determination studies, not enough lessons have been learned from the past. Pilot studies are often planned without sufficient statistical power, due to an insufficient number of patients and highly variable blood pressure measurements. In the development of the new angiotensin converting enzyme (ACE) inhibitor benazepril, crossover trials were used to obtain useful information. At the end of phase II of the benazepril development, a double-blind crossover study was carried out with 25 patients, and the results made it possible to redefine the 12- and 24-h effects of benazepril in comparison with placebo. Moreover, the crossover trial allowed an investigation of the biological effects of the treatment. In further work, the efficacy of 10 mg benazepril, administered once a day, was confirmed in comparison with captopril and enalapril, with a beta-risk of less than 20%. Since this crossover study yielded reliable data, and there was no carryover effect, a similar crossover design was used to study the interaction between benazepril and nifedipine. In the past, mistakes were made and many antihypertensive drugs were administered in high doses, with no further beneficial effect on blood pressure and an increased risk of side effects. Work described in this paper shows that fewer but better designed and implemented studies can improve the efficiency and value of dose-finding studies for antihypertensive drugs.