Arterial hypertension represents the most important cardiovascular risk factor with a direct responsibility for a large share of cardiovascular mortality and morbidity in the world. Despite the wide availability of antihypertensive therapies with documented effectiveness, blood pressure control still remains largely unsatisfactory in large segments of the population. Guidelines for the management of arterial hypertension suggest the preferential use of five classes of drugs—angiotensin-converting enzyme inhibitors, angiotensin II type I receptor inhibitors, calcium channel blockers, thiazide/thiazide-like diuretics, and beta-blockers—recommending the use of combination therapy, preferably in pre-established combinations, for the majority of hypertensive patients. The evidence of a non-negligible heterogeneity in the response to different antihypertensive drugs in different patients suggests the opportunity for personalization of treatment. The notable phenotypic heterogeneity of the population of hypertensive patients in terms of genetic structure, behavioural aspects, exposure to environmental factors, and disease history imposes the need to consider all the potential determinants of the response to a specific pharmacological treatment. The progressive digitalization of healthcare systems is making enormous quantities of data available for machine learning systems which will allow the development of management algorithms for truly personalized antihypertensive therapy in the near future.