Among endocrine disorders, Cushing's syndrome (CS) is certainly one of the most challenging to endocrinologists due to the difficulties that often appear during investigation. The diagnosis of CS involves two steps: confirmation of hypercortisolism and determination of its etiology. Biochemical confirmation of the hypercortisolaemic state must be established before any attempt at differential diagnosis. Failure to do so will result in misdiagnosis, inappropriate treatment, and poor management. It should also be kept in mind that hypercortisolism may occur in some patients with depression, alcoholism, anorexia nervosa, generalized resistance to glucocorticoids, and in late pregnancy. Moreover, exogenous or iatrogenic hypercortisolism should always be excluded. The three most useful tests to confirm hypercortisolism are the measurement of 24-h urinary free cortisol levels, low-dose dexamethasone-suppression tests, and determination of midnight serum cortisol or late-night salivary cortisol. However, none of these tests is perfect, each one has different sensitivities and specificities, and several are usually needed to provide a better diagnostic accuracy. The greatest challenge in the investigation of CS involves the differentiation between Cushing's disease and ectopic ACTH syndrome. This task requires the measurement of plasma ACTH levels, non-invasive dynamic tests (high-dose dexamethasone suppression test and stimulation tests with CRH or desmopressin), and imaging studies. None of these tests had 100% specificity and their use in combination is usually necessary. Bilateral inferior petrosal sinus sampling is mainly indicated when non-invasive tests do not allow a diagnostic definition. In the present paper, the most important pitfalls in the investigation of CS are reviewed.