Statistical optimization models were employed to optimize the adsorption of textile dye effluent onto Gracilaria edulis. Significant factors responsible for adsorption were determined using Plackett-Burman design (PBD) and were time, pH, and dye concentration. Box-Behnken (BB) design was used for further optimization. The predicted and the experimental values were found to be in good agreement, the coefficient of determination value 0.9935 and adjusted coefficient of determination value 0.9818 indicated that the model was significant. The results of predicted response optimization showed that maximum decolorization could be attained with time 131.51 min, pH 7.48, and dye concentration 23.13%. The model was validated experimentally with 92.65% decolorization efficiency. The experiment was confirmed using Fourier transform infrared spectroscopy (FTIR), high-resolution scanning electron microscope coupled with energy dispersive X-ray analysis (HR-SEM-EDX), X-ray diffraction spectrometry (XRD) and Brunauer-Emmett-Teller (BET) surface area and pore size analysis techniques. Desorption studies at various pH (2–14) were performed and a maximum of 23% of the dye was recovered from the adsorbed biomass.