Background: Phenotyping, using caffeine as probe substrate, is a proper method to assess CYP1A2 activity. We evaluated the utility of dried blood spots (DBS) for CYP1A2 phenotyping. Results: LC–MS/MS methods were developed and validated for quantitation of caffeine and its metabolite paraxanthine in DBS, whole blood and plasma. All parameters met the pre-established criteria. While recovery, matrix effects and precision were unaffected by hematocrit (Hct), there was a Hct effect on accuracy, although for the evaluated Hct interval (0.36–0.50) it remained within acceptable limits. The phenotyping methods were successfully applied in healthy volunteers. Conclusion: Excellent method performance and highly comparable phenotyping indices in DBS, whole blood and plasma, combined with the benefits of DBS sampling, illustrate the suitability of DBS-based CYP1A2 phenotyping.