In this study, we aim to develop and improve an attribute classification method for pedestrians using a plantar pressure value. At present, many retail businesses use various methods (e.g., member's card, direct mail, questionnaire, manual classification by staff members) of collecting customer information. However, these methods have some problems. One of the problems is the instability of collecting customer information. The member's card cannot cover all customers. Moreover, manual classification includes dispersion by individual difference. However, using pressure sensors has advantages. One of the advantages is that the pressure sensor does not violate the targeted person's privacy, because the pressure values from the sensors cannot identify one individual from a large indefinite number. Another advantage is that the pressure sensor can be measured without being affected by the environment (e.g., lighting, field of view, focus range). Our method of attribute classification uses four feature quantities, which are the CoG (center of gravity in the X-axis and the Y-axis), plantar area, and total pressure value. In the experimental results, the classification accuracy for gender was 72.19% and that for three age groups was 42.33%.