IDENTIFYING PREFERENCES THROUGH MOUSE CURSOR MOVEMENTS: PRELIMINARY EVIDENCE

Identifying customers’ preferences is a challenging task with significant practical implications for online shopping. Current methods often put considerable burden on the customers through such methods as questioning, so the process could benefit from a more accurate and less intrusive estimation of how customers weight product attributes, particularly in the initial purchasing phase. Our goal is to derive attribute weights automatically by recording and analyzing cursor movements. We conducted an experiment to confirm the suitability of the proposed design, and found a highly significant correlation between the time people spend investigating a product attribute and their self-reported importance rating. Our proposed Web page design might also reduce the risk of information overload.