Existing studies on customer behavior lack quantitative and high efficiency study, their technologies rely heavily on hardware. Therefore, the information of consumers in offline stores was insufficient, which made enterprises unable to accurately track consumers. However, computer vision (CV) is an expert in identifying and tracking people’s behavior, and its function is suitable for investigating enter-store customer behavior. Therefore, the aim of our study was to develop an offline consumer behavior portraying system based on CV. Then we used this system to investigate enter-store consumption behavior. We selected 71 shoe stores in China, then installed the system in store for a three-month data collection, and evaluated the impact of customer's age, gender, enter time, and region factors on enter-store behavior in China. Through our system, we successfully study ways to improve the purchase conversion rate of enter-store consumers, which could guide enterprises to adjust better marketing and operation strategies.
Real Time Impact Factor:
Pending
Author Name: Jingjing LI, Keyu HOU, Wei XU, Jin ZHOU
URL: View PDF
Keywords: customer behavior, computer vision, off-line retailing, enter-store data collection
ISSN: 1583-4433
EISSN:
EOI/DOI: https://doi.org/10.24264/lfj.2
Add Citation
Views: 1
