As online shopping continues to dominate the retail landscape, ecommerce platforms are increasingly facing the challenge of how to manage low-quality products.
These products, often overlooked until consumer complaints pile up, can tarnish a platform’s reputation, and negatively impact sales. However, research from the University of Illinois Urbana-Champaign and Penn State University offers a new alternative.
The study introduces a “Two-Stage Classification Model” designed to proactively identify potentially problematic products before they lose consumer favor. This system, developed by a group led by Anton Ivanov, a business administration professor at Illinois, predicts future product ratings and flags items that are likely to perform poorly based on emerging trends in customer feedback.
How to manage low-quality products on ecommerce platforms
“We’ve created a framework that can forecast how consumer ratings of a product might evolve, allowing platforms to take action before a product’s reputation deteriorates,” Ivanov says. “Instead of waiting until a product is already losing popularity, our system identifies negative trends early on, helping platforms maintain their reputation and ensure a positive shopping experience.”
The model’s predictive capabilities could bring change to ecommerce platforms, particularly as consumer reviews and ratings become key decision-making factors. Ivanov emphasized that the ability to identify and remove low-quality products before they reach a critical threshold is crucial to sustaining trust and customer loyalty.
“Platforms need to act quickly to remove low-quality products and act against suppliers who continually provide poor items. Failing to do so could harm the platform’s credibility and erode consumer confidence,” he says.
The researchers assessed their model using data from over 2,800 electronics products, analyzing 800,000 consumer reviews. By using a deep learning sequence model, the group forecasted future product ratings and identified negative trends earlier than traditional single-stage methods. Their approach proved more dependable and accurate, allowing for better prediction of a product’s potential decline in quality perception.
This initiative-taking model could help ecommerce platforms maintain high product standards and prevent the widespread sale of subpar goods. It also could be particularly beneficial for managing third-party sellers whose product quality can vary widely, the study showed.
“The ability to anticipate which products are likely to be poorly rated in the near future will empower ecommerce managers to act swiftly, protecting the platform’s overall reputation,” Ivanov says.
The researchers suggest that ecommerce platforms could integrate this tool to automatically flag products for further scrutiny, helping them stay ahead of potential issues and make data-driven decisions to protect their brand and consumer trust.
Sign up
Sign up for a complimentary subscription to Digital Commerce 360 B2B News, published 4x/week. It covers technology and business trends in the growing B2B ecommerce industry. Contact Mark Brohan, senior vice president of B2B and Market Research, at mark@digitalcommerce360.com. Follow him on Twitter @markbrohan. Follow us on LinkedIn, X (formerly Twitter), Facebook and YouTube.
FavoriteThe post New model helps ecommerce platforms proactively identify low-quality products appeared first on Digital Commerce 360.
from Digital Commerce 360 https://ift.tt/U2OLjMC
0 Comments