Mislabeling of seafood is a global phenomenon that can misrepresent the status and level of consumption of wild fish stocks while concealing the use of many other wild species or those originating from aquaculture and sold as substitutes. We conducted a DNA barcoding study in three cities within Mexico (Mazatlan, Mexico City and Cancun) and sequenced the COI gene in 376 fish samples sold as 48 distinct commercial names at fish markets, grocery stores, and restaurants. Our goal was to identify the main species sold, their mislabeling rates and the species most used as substitutes. Overall, the study-wide mislabeling rate was 30.8% (95% CI 26.4–35.6). Half of the samples collected belonged to five species traded globally (yellowfin tuna, Atlantic salmon, mahi, swai, and tilapia), most of them with important aquaculture or ranching production levels. These species were commonly used as substitutes for other species and showed low mislabeling rates themselves (≤ 11%, except mahi mahi with 39% mislabeling). The other half of the samples revealed nearly 100 species targeted by small-scale fishers in Mexico and sold under 42 distinct commercial names. Popular local commercial names (dorado, marlin, mero, robalo, mojarra, huachinango, pargo, sierra) showed the highest mislabeling rates (36.3% to 94.4%) and served to sell many of the 53 species identified as substitutes in our study. We discuss the observed patterns in relation to landing and import data showing differences in availability of commercial species and the links to explain observed mislabeling rates and the use of a species as a substitute for other species. We also outline some of the implications of establishing a labeling and traceability standard as an alternative to improve transparency in the trade of seafood products in Mexico.