An overlooked source of variation in evolvability resides in the social lives of animals. In trying to foster research in this direction, we offer a critical review of previous work on the link between evolutionary speed and sociality. A first set of studies emerged that we grouped under the “selectionist perspective”. These studies conceive social interactions as knitting forces of an environment with specific selection pressures. Social pressures, in turn, promote evolutionary change in the characters they depend on. Studies in this perspective have investigated cladogenetic effects of sexual selection, while non-sexual contexts have been relatively ignored. We grouped a second set of studies grounded on population genetics under the “populationist perspective”. Such studies regard social interactions and the social units arising from them as a dividing force that splits a population into smaller parts, tweaking the number of reproducing individuals and unbalancing the contribution of both sexes. The consequences of these effects on the effective breeding population size are made responsible for changes in the speed of neutral evolution. Despite the manifold approaches used to measure sociality and evolutionary speed, there is support for both the selectionist and populationist perspectives on anagenesis. On the contrary, evidence for cladogenetic consequences is mixed. We suggest six areas for im-provement to cope with the current situation: 1) Conceptually separating the potential for evolutionary change from its realization. 2) Considering that under social competition, a single axis of variation is unlikely to explain reproductive success. Acknowledging the existence of alternative social tactics could enrich the current framework. 3) Address both the selectionist and populationist perspectives simultaneously. Social selection strength and Ne consequences need to be assessed using 4) as many axes of social variation as possible and 5) in both sexes. 6) Considering the evolutionary covariances in communicative systems might improve the validity of tests for the current framework. In addition, we develop predictions for how variation in each social dimension and component might affect evolutionary speed. Continuing to refine the theory and evidence on social effects on evolutionary speed might come at a benefit not only for the current issue but also for the domains it integrates.