The present work demonstrates the adsorption of hydroquinone (HQ) and resorcinol (R) by activated carbon based on shea residue (Vitellaria paradoxa). The adsorbent was prepared chemically by impregnation with sulfuric acid and coded by the acronym CAK-S. The central composite design (CCD) was used to optimize the main factors that influence the adsorption of HQ or R by activated carbon such as the initial concentration, the pH of the solution, the contact time, and the mass of the carbon on the expected response, which is the adsorbed quantity of the target pollutants. The optimal conditions obtained from the statistical analysis are as follows: concentration of 158 mg/L, pH 3, time of 120 min, and mass of 50 mg for the adsorption of HQ and concentration of 180 mg/L, pH 3, time of 86 min, and mass of 118 mg for the adsorption of R. The maximum quantities of HQ and R adsorbed are 45.02 mg/g and 33.65 mg/g, respectively. The analysis of variance (ANOVA) showed a good relationship between the variables involved with the coefficients of determination R2 = 98.69% for the adsorption of hydroquinone and R2 = 90.55% for that of resorcinol, which means that the model is more suitable to express the adsorbed amount according to the four optimized parameters. The experimental data obtained under these optimal conditions were simulated with two and three parameter nonlinear isotherm models as well as kinetic models. The results show that Elovich kinetic model better describes the adsorption of HQ and R, indicating chemisorption with heterogeneous active sites on the surface of CAK-S. Temkin’s two-parameter model shows that adsorption occurs on heterogeneous surfaces with a nonuniform adsorption energy distribution at the surface and Sips’s three-parameter model confirms the heterogeneity of the surface with a localized adsorption of HQ or R by CAK-S. The thermodynamics study has shown that the adsorption is endothermic (
Δ
H
0
>
0
) and spontaneous (
Δ
G
0
<
0
).