<p>How reliable is the evidence for new forms of treatment in ophthalmology and what does each study phase actually demonstrate? Phases I–IV not only stand for different developmental stages but also particularly for different targets: from the first safety assessment and dosage results to the proof of efficacy and to data from treatment practice; however, especially in ophthalmology the assessment of clinical trials is particularly challenging. Bilaterality, invasive application routes, complex functional measurement and the increasing importance of imaging biomarkers make the planning, performance and interpretation difficult. In addition, there are discipline-specific sources of bias, due to the choice of the analytical unit, masking problems or the limited informative value of structural surrogate points with no clear functional relevance. The article shows how ophthalmological studies can be correctly classified and critically evaluated though all phases. The focus is on study design, choice of endpoints, regulatory context and typical methodological pitfalls. The aim is to provide a practical orientation for the clinical routine and to strengthen the foundation for evidence-based treatment decisions.</p>