Evidence-informed innovation goes through flanking research and untracked paths, together with didactic professionalisation, after failure and opposition, and always with a controlled degree of ‘unreasonableness’.