The potential for malachite green dye saturated effluent to severely affect the environment and human health has prompted the search for effective treatment technologies. Thus, this study was conducted with the goal of developing activated carbon from Rumex abyssinicus for the adsorptive removal of malachite green dye from an aqueous solution. Unit operations such as drying, size reduction, impregnation with H3PO4, and thermal activation were used during the preparation of the activated carbon. An experiment was designed considering four main variables at their respective three levels: initial dye concentration (50, 100, and 150 mg/L), pH (3, 6, and 9), contact period (20, 40, and 60 min), and adsorbent dosage (0.05, 0.01, and 0.15 g/100 mL). Optimization of the batch adsorption process was carried out using the Response Surface methodology's Box Behnken approach. The characterization of the activated carbon was described by SEM for surface morphology with cracks and highly porous morphology, FTIR for multi-functional groups O–H at 3506.74 cm−1 and 3290.70 cm−1, carbonyl group stretching from aldehyde and ketone (1900–1700 cm−1), stretching motion of aromatic ring C=C (1543.12 cm−1), stretching motion of –C–H (1500–1200 cm−1), vibrational and stretching motion of –OH (1250.79 cm−1), and vibrational motion of C–O–C (1049.32 cm−1), pHpzc of 5.1, BET for the specific surface area of 962.3 m2/g, and XRD for the presence of amorphous structure. The maximum and minimum dye removal efficiencies of 99.9% and 62.4% were observed at their respective experimental conditions of (100 mg/L, 0.10 mg/100 mL, pH 6, and 40 min) and (100 mg/L, 0.15 mg/100 mL, pH 3, and 20 min), respectively. Langmuir, Freundlich, Toth, and Koble-Corrigan models were used to evaluate the experimental data, in which Koble-Corrigan model was found to be the best fit with the highest value of R2 0.998. In addition to this, the kinetic studies were undertaken using pseudo-first-order, pseudo-second-order, intraparticle diffusion, and Boyd models, and as a result, the pseudo-second-order model proved to have a better fit among the kinetic models. The kinetics and isotherm analysis revealed that the nature of the adsorption to be homogenous and monolayer surfaces driven by chemosorption. Furthermore, the thermodynamics study revealed the nature of adsorption to be feasible, spontaneous, and endothermic. On the other hand, the reusability study depicted the fact that the adsorbent can be utilized for five cycles with a negligible drop in the removal efficiencies from 99.9 to 95.2%. Finally, the low-cost, environmentally benign, and high adsorption capacity of the adsorbent material derived from Rumex abyssinicus stem could be used to treat industrial effluents.