According to McKinsey research, 3/4 of the films and series watched on Netflix are played after the viewer is shown a recommendation that they will enjoy it. Behind the suggestions is a machine learning algorithm that analyzes their choices and taste.

It is similar with Amazon, where AI suggestions generate as much as 35 percent of purchases. Now a similar technology has been made available to over 15,000 Polish online sellers on the Shoper platform.

  • The Shoper e-commerce platform has introduced a new recommendation system based on machine learning technology
  • During online shopping, buyers are shown additional products tailored specifically to them
  • The principle of its operation is described in an article on the Shoper blog
  • Thanks to the SaaS model, the new system operating as part of the Shoper Recommendations service is available to all of the over 15,000 sellers using the Shoper platform

Online stores like Netflix

How does product recommendation work?

Artificial intelligence in Polish e-commerce — product recommendations

Recommendation systems today form the foundation of the most popular apps and VOD services. Their users, while browsing content, watching films, or listening to music, receive notifications that allow them to discover content similar to what they have already watched and matching their taste. Global sales platforms also use similar technology.

- Thanks to the analysis of customer visits to online stores, on a site that uses a so-called recommendation system, buyers are shown suggestions and product sets that may additionally interest them and increase the value of their shopping basket. Recommendations are the result of advanced algorithms and their ability to learn, and the analysis takes place in real time – the system analyzes what will interest the customer on the current purchasing path and, based on this information, proposes the best-matched products – says Adrian Słowik, Software Architect and IT Manager at Shoper.

Technology of giants available to everyone

Technology of giants for every store

More than 15,000 online stores working on Shoper software can already use such a solution. The system operates in stores as part of the Shoper Recommendations service, available for all subscription plans. This is another innovation that Shoper has introduced to its platform to improve and modernize sales.

- The system in Shoper online stores is designed so that, in product suggestions, the customer sees only those proposals that meet their search criteria during a visit to the site. This is an important element that affects the high quality of the suggestions and thus increases the chance of conversion – says Jacek Zientkiewicz, Business Development Director at Shoper. - Additionally, through the recommendation system, customer service in the store can be improved. As market analyses show, most consumers prefer to shop in those stores that present them with accurate offers and product recommendations. Online, where some buyers may lack direct contact with the seller, this type of new technology can improve everyday shopping and make the customer feel perfectly served.

Summary

The Polish e-commerce platform Shoper has made available to over 15,000 sellers a recommendation system based on machine learning, similar to solutions used by Netflix or Amazon. The algorithm analyzes customer behavior in real time and suggests products to them, increasing the chance of conversion. This is a step toward local entrepreneurs matching global giants.