Detersive processes are complex systems involving a great number of variables. To determine the effect of these variables on the washing of hard surfaces and fatty soils is the object of this work. The statistical design of experiments has been used to evaluate the influence of factors such as temperature, soil concentration and surfactant concentration on detergency. The experimental trials have been made in a continuous-flow device where the soiling agent is confined in a column filled with borosilicate glass spheres. Solutions of the commercial surfactant Berol Ó LFG61 (a mixture of alkylpolyglucosides and fatty-alcohol ethoxylates) have been employed as the wash bath. Both the design of experiments and the continuous experimental system used proved to be an effective tool for detecting the key variables in the cleaning process. Expressions were developed to simulate detergency levels as a function of the variables assayed, always inside the experimental domain. In the trials with oleic acid as the soiling agent, it was found that the temperature and soil concentration were the most important variables to take into account, while the surfactant concentration was not a significant variable. When a semi-solid mixture of different fatty acids was employed, all the variables assayed proved significant, with high detergency values being reached by combining temperature and surfactant concentrations. Results clearly show that the effectiveness of the surfactant used is influenced by the type and concentration of the soil and thus the intended application of the product being developed should be taken into account when designing detergent formulas.