The effectiveness of constructed wetland (CW) in the treatment of synthetically prepared wastewater with a mixed dye (Congo red and methylene blue) at varying concentrations (100, 200, and 300 ppm) is assessed in this study. Two laboratory-scale wetland systems (G1 and G2), each having a depth of 35 cm and diameter 30 cm, were constructed. Both systems were planted with Canna indica as the wetland vegetation. System G1 was packed with fine gravel, crushed autoclaved aerated concrete (AAC) brick, coal, and coarse gravel, while system G2 was packed with fine gravel, crushed footpath brick, coal, and coarse gravel. The performance of the system was assessed by analyzing dye removal efficiency, pH, chemical oxygen demand (COD), temperature, and oxidation–reduction potential (ORP). A double-beam UV–Vis spectrophotometer was used to measure the absorbance of the mixed dye in the wastewater at its maximum wavelength before and after treatment. The result showed that the dye removal efficiencies of wetland systems G1 and G2 were 88.0% and 87.8%, respectively, at 100 ppm; 93.0% and 87.0% at 200 ppm; and 96.0% and 92.4% at 300 ppm of mixed dye concentration. COD removal efficiencies of wetland systems G1 and G2 were 96.0% and 95.3% for 100 ppm, 92.0% and 91.0% for 200 ppm, and 95.2% and 95.0% for 300 ppm, respectively. Kinetic analysis indicated that the pseudo-second-order model provided a better fit to the dye removal data than the pseudo-first-order model, with higher R² values (0.9722–0.9977) compared with the pseudo-first-order model (0.6887–0.9412), suggesting that adsorption-related interactions predominantly governed the dye removal process. Graphical Abstract