There have been several failed attempts to retire the journal impact factor (see also R. J. Roberts Nature 546, 600; 2017). However, there can be value in knowing which journals are publishing papers of immediate wide interest. I therefore suggest that, rather than repealing or replacing the impact factor, its producers should rename it to reflect its intended function more accurately.

The problem lies not with the impact factor's calculation — apart from flaws such as the criteria for inclusion or exclusion — but in its misleading name, which results in its misuse. It would be better classed as an average 'short-term citability factor' that is valid for two years after publication. Journals, authors, grant agencies and institutions could then decide whether to compete in this category.

Alternatively, peer-reviewed journals could choose to be compared using other indices, such as a long-term (say, 20-year) citability factor and/or total citations over time. This might encourage editors to select papers purely on quality and originality, as they used to, irrespective of a paper's potential for popularity.

Academic leaders, administrators, policymakers and funding agencies would then be free to decide whether to base their assessments on publications in journals with short- or long-term popularity and citability.