Designing or renovating inheritance hierarchies in the domain of programming or in the domain of modeling still remains a tricky task. It involves integrating domain concepts sometimes with no clear frontier, finding the good abstractions and avoiding duplicated information. In this paper, we review research work that addressed this topic with the use of Formal Concept Analysis (concept lattices) since the seminal paper of R. Godin and H. Mili at OOPSLA'93. We overview the different attempts, the explored limits, and the current issues.