Insights at UBC Sauder

Personalized online product recommendations may be more deceptive than they seem

Posted 2018-09-06

Beware of “personalized” product recommendations on your favourite e-commerce sites, because some retailers might use them deceptively to encourage consumers into buying lower quality products.

That’s the key finding of new research from the UBC Sauder School of Business, which explored how biased product recommendations on online shopping sites can sway buying decisions.

Biased products are those specifically promoted by online retailers. The products are typically displayed before non-promoted products on a webpage, giving consumers the impression that the results are sorted in bestselling order. This might occur if retailers would like to prioritize some products or brands over others, or they have excess inventory of a certain product and need to boost its sales.

The study found that when consumers searched for a particular product, and then received biased personalized recommendations based on this search, they had greater confidence in their product choice despite ultimately choosing a lower-quality product. But, consumers who received biased recommendations, based on what were claimed to be popular products, without inputting search criteria were more likely to venture beyond the first few pages of the site to search for alternative products.

“Today’s digital marketplaces offer consumers great convenience, immense product choice and large amounts of product-related information,” says study co-author Izak Benbasat, professor emeritus at UBC Sauder. “But there’s also a dark side of information technology that consumers need to be aware of. In this case, just because they’re searching for specific product criteria, doesn’t necessarily mean the site will return the most accurate results.”

Consumers were also largely unaware of the bias of the website’s product recommendations, revealing the risk consumers face when shopping online.

“When there’s an element of personalization, consumers mistakenly believe their decision making becomes more efficient, and they have a higher level of confidence in their purchase,” says Benbasat. “But the reality is that consumers who don’t receive personalized criteria actually choose better products more often because they do a more thorough product search.”

Benbasat adds that consumers should always do their homework to avoid getting duped by online shopping giants, whether “it’s checking reviews on multiple web sites, or better yet, finding a product review on a verified site like Consumer Reports.”

“Trust is very important in the online world, so when companies are caught ‘cheating’ consumers in any way, trust—and revenues—drop,” he says. “It would benefit companies to be more open with their tactics, and include clear disclaimers that explain why a product is displayed first.”

The study was co-authored by Bo Xiao of the University of Hawaii and was recently published in the journal Decision Support Systems. It follows up previous research from Benbasat and Xiao, which examined personalized online product recommendations and how they benefit e-commerce merchants rather than consumers. The study found that consumers who failed to further research an e-commerce site’s personalized product recommendations—even when they received a warning from the site to do so—were more likely to be misled by biased product recommendations.