Sea level variations in the coastal zone can differ significantly from those in the open ocean and can be highly spatiotemporally coherent in the alongshore direction. Yet, where and how coastal sea levels exhibit variations that emerge as persistent and recurrent patterns along the world's coastlines remain poorly understood. Here, we use a Bayesian mixture model to identify large‐scale patterns of coherent modes of monthly coastal sea level variations from coastal altimetry and tide gauge data. We determine nine clusters of coherent coastal sea level variability that explain a majority of the monthly variance measured by tide gauges (1993–2020). The analysis of along track altimetry data enables us to detect several additional clusters in ungauged regions, such as the Indian Ocean or around the South Atlantic basin, which have so far been poorly described. Although some clusters (e.g., at the eastern boundary of the Pacific, the western tropical Pacific, and the marginal and semi‐enclosed seas) are highly correlated with climate modes, other clusters share very little variability with the considered climate modes at the monthly timescale. Knowledge of these coherent regions thus motivates and enables further investigations on the impacts of local and remote forcing on coastal sea level variability, and the extent to which coastal sea level variability is decoupled from the adjacent deep ocean.