Early-phase clinical trials with any new chemical entity have three main objectives: to establish that the drug is effective in the target disease and patient population; to assess tolerability and safety; and to select the optimum dose or doses for further evaluation. Clinical trials in migraine and other headache disorders pose additional problems because of the lack of specific objective diagnostic tests and the reliance on patient assessments of symptom relief. For these reasons particular care must be taken in selection of patients, design of the trials, choice of end-points and statistical analytical methodology. A wide range of doses must be evaluated to ensure that the dose-response curves for both efficacy and safety/tolerability are fully defined. Successful early-phase trials should then allow the optimum dose or doses of the drug to be identified for further evaluation in large-scale outpatient studies.