As people increasingly use websites and digital services for shopping, leisure, entertainment and socializing, recommender systems are playing a large part in the choices that consumers make every day. Recent studies in the field by Adomavicius, Bockstedt, Curley and Zhang et al show that recommender systems are now also influencing customer behavior significantly from an economic and decision-making perspective. We explore this aspect of our blog today.
Helping customers decide
Digital marketplaces have added not just a greater variety of options to choose from, but also further pressure on the customer’s decision-making. Customers now must evaluate a larger number of products from more businesses on websites, as compared to the choices offered in a limited space brick-and-mortar shop. Researchers Adomavicius et al note that this task becomes even more complex when it comes to making choices on experiences and preference-based products like choosing songs, books or movies. Customers must be more cautious when exercising choice because mismatches cost them more in time and money.
Recommender systems are designed to benefit both buyers and sellers – it saves customers the time and effort required through pages of different products available in digital markets while businesses can use it to understand customer preferences, build brands and increase sales.
Recommendation engine algorithms make personalized recommendations based on past historical data about the customers and this results in significant benefits. For example, in the case of Netflix, more than 80% of the content that subscribers watch comes through the personalized recommendation that the service offers. Similarly, Amazon matches customers based on their past ordering, stated preferences and browsing with products that they are most likely to be looking for. In doing so, it can nudge the customer to discover and purchase new products faster.
Willingness to pay
Adomavicius et al in their study, looked at how recommendations influenced a customer’s preference and willingness to pay for music as received from digital platforms like Spotify etc. They found that customers now prefer what the recommendation system says they would like, rather than just relying on their personal experience of listening to a song. Further, they found that recommendations significantly altered the customer’s willingness to pay. As a product received a higher rating, the customer’s willingness to pay for it also increased. These findings have important implications for the entire retail industry.
Implications for business and retailers
More research by Dokyun Lee of Carnegie Mellon University and Kartik Hosanagar of Wharton examined the interaction between a recommender system and product attributes along with reviews in an e-commerce setting. Their studies have some interesting takeaways for retailers to come up with effective recommendation strategies in e-commerce. They suggest businesses whose e-commerce sites have low review volumes could prioritize recommender implementations. Their research indicates that a longer product description increases recommender effectiveness – another step that retailers could take to drive sales in the digital store.
Retail and service organizations would do well to evaluate their existing e-commerce platforms and recommendation strategies with the help of a trusted IT services vendor. Tricon Infotech is an IT consulting and software services company, delivering complex world-class custom software solutions on a variety of technology platforms to organizations across industry verticals. As a software and product development company, we offer a full range of custom software development services and SaaS solutions for a wide variety of verticals and business domains. Connect with us to accelerate your digital goals.