Ontologies and vocabularies play a key role when standardising, organizing and integrating data from heterogeneous data sources into Knowledge Graphs. In order to develop ontologies, different engineering methodologies have been proposed throughout the years, whose application resulted in thousands of semantic artefacts (taxonomies, vocabularies and ontologies) in a wide range of domains. But how to ensure that ontologies follow the Findable, Accessible, Interoperable and Reusable principles (FAIR) from their inception? In this paper, we review: (i) existing guidelines to help make ontologies FAIR and (ii) published FAIRness assessment methodologies and map them to the ontology development lifecycle activities. Our analysis outlines the current gaps, where no guidelines exist for ontologies to become FAIR-by-design.