This paper aims to demonstrate the advantages of a sequential approach to decision making in climate policies, This sequential approach is necessary because of the uncertainties regarding both the avoided costs of climate change and the costs of GHG emission abatement, This assessment focuses on the minimization of learning time, the reduction of technical and institutional inertia, and the possible unexpected non-linearities in abatement curves. A stylized representation of so-called ＇no-regret＇ short-term potentials, of long-term innovation trends and of backstop technologies is used, We show that if we do take care of the inertia constraints in the decision process, both policies (namely the no-regret and the innovation process exploitation) becomes crucial in terms of minimization of the learning time.