It is a privilege to have respected colleagues engage with our definition and ethical evaluation of overdiagnosis. In our response to the commentaries, we first deal with paradigmatic issues: the place of realism, the relationship between diagnostic standards and correctness and the distinction between overdiagnosis and both false-positives and medicalisation. We then discuss issues arising across the commentaries in turn. Our definition captures the range of different types of overdiagnosis, unlike a definition limited to diagnosis of harmless disease. Certain implications do flow from our definition, as noted by commentators, but we do not view them as problematic: overdiagnoses can become beneficial diagnoses as medical knowledge and practice changes over time; inadequate systems of healthcare can produce tragic overdiagnosis, and the effectiveness of treatment partly determines whether overdiagnosis occurs. Complexity and uncertainty in balancing benefits and harms is unfortunate, but not a reason to avoid making a judgement (ideally one that reflects multiple perspectives). We reaffirm that overdiagnosis, for the foreseeable future, must be estimated at a population level and defend the importance of good-quality risk communication for individuals. We acknowledge that a lot turns on the relevance of professional communities in our definition and expand our reasoning in this regard then conclude with a note on the difference between intentions and goals. We expect that it will be some time before these matters are settled and we look forward to continue debating these matters with our colleagues.