The European Habitat Directive encompasses a conservation policy focused on habitats rather than single species. This policy is based on a list of « Habitats of Community Interest » that must be preserved or restored in the Natura 2000 network. The underlying ambition to conserve large-scale integrative units has strong ecological justifications, and inspires more recent initiatives such as the IUCN red list of ecosystems. Evaluating this policy is therefore pivotal to identify and reproduce best practices. For that purpose, we take advantage of the literature in political decision-analysis to define a four-fold normative framework according to which a conservation policy can be positively evaluated if: (1) it demonstrably contributes to conservation, (2) it is science-based, (3) it is operational and (4) it is legitimating. Based on an extensive exploration of the published scientific literature, unpublished reports and databases (among which the recently released European-wide evaluation of the conservation status of habitats of community interest), we identify a series of knowledge gaps plaguing this habitat conservation policy. We argue that, due to these knowledge gaps, the contribution of this policy to the conservation of habitats is unproven, it is not science-based, not operational and not legitimating. Our analysis draws heavily on the French implementation, and we investigate it to show how this example can be used by other countries, France itself and Europe as a whole, to improve this habitat policy. We then draw the constructive lessons from our analysis, identifying concrete means to strengthen the European habitats conservation policy.