Natural dyes, such as annatto, are widely used in the food and pharmaceutical industries. However, the effluents from annatto dye production pose environmental challenges due to the presence of residual dye components, organic matter, and dissolved inorganic species (chloride and sulfate). This study, the first attempt in the literature to deal with real annatto dye effluent, aimed to evaluate the use of activated carbon for the removal of carotenoids (norbixin), chloride, and sulfate. A 2³ full-factorial design was employed to optimize the adsorption process parameters (temperature, pH, and adsorbent mass). The results showed that up to 90% of carotenoids and chloride could be removed under optimized conditions (low temperature and pH). Sulfate removal was more strongly influenced by the pH of the medium, with a maximum removal of 53%. The adsorption process was well-described by empirical models, though traditional isotherm models did not adequately represent the experimental data. Hypotheses that support isotherm models such as pore homogeneity, monolayer formation, constant internal diffusion were not satisfied. This study demonstrates the potential of activated carbon adsorption as a feasible method for treating food industry effluents containing high loads of carotenoids and inorganic salts.