This health e-Commerce company has more than 120 years of experience in medical supplies, health products and services. In addition, it is a company that is present in many countries and has more than 150,000 customers a year and about 20,000 references available. 

This entity is both a distributor and a manufacturer. And it is characterized by its high commitment to its customers' requirements. Likewise, this e-Commerce is in a continuous process of improvement in its projects. And all of this, from a solid base of social responsibility and betting on innovation and diversification. 

Undoubtedly, online channels are key to this company's sales strategy. However, due to the breadth of its catalogue, it is difficult for them to offer the products best suited to the needs of each customer. Hence the requirement to have a recommendation engine that advises each user on the products that may be of most interest to them.
 

Challenge

As we have indicated above, this company has more than 20,000 references. This large number was preventing users from viewing items that could potentially be of interest to them. In fact, the company was missing out on sales opportunities and required changes to its business system.

For this reason, in the last years, it has been implementing different strategies trying to improve the click-through and conversion rates of its e-Commerce: widgets with best-selling products, most popular products, new products... 

However, they had the great challenge of offering each user a completely personalized product proposal according to their specific health problem. To do so, they decided to go with Emergya and Google Cloud technology as a solution. 

Solution

The strategy followed was to analyze information on user behavior on the website. Thanks to this data collection system, it was possible to establish a score of interest in each product. 

Based on the data collected and using Machine Learning clustering techniques on GCP products, a recommendation system based on collaborative filters was implemented on the website.

This mechanism is basically responsible for recommending those products that are most likely to be relevant to the user and that are similar to others that have already caught their attention.

Results

After implementing the recommender on the home and product pages of the e-commerce site, and after finding that the click-through rate was 25% higher than with the product widgets they already had, the site's KPIs improved significantly:

• Depth of navigation increased by 12%.
• Overall website conversion rate improved by 0.5%.
• The bounce rate on product landing pages decreased by 10%.

With these results and, after applying Emergya's know-how and Google Cloud's cutting-edge technology, we managed to improve the customer experience and achieve the goal of increasing the conversion rate.