Recommendation engine for personalised products

How we helped Magneds customise a recommendation engine for personalised products.

Situation

Magneds offers tailor-made loyalty and savings programmes to retail customers, ranging from advice and content creation to implementation and monitoring. A retail customer, one of the largest dairy producers worldwide, requested to extend their long-lasting loyalty programme with a personalised product service.

Approach

We have implemented a custom recommendation engine that suggests personalised cashback or new products geared towards a specific customer segment. We used a collaborative filtering algorithm that detects customer and product profiles based on previously redeemed products. Throughout the project, several A/B-tests were designed and executed to test the efficacy of the recommendation engine to other internal recommendation methodologies.

Results

The A/B-tests proved that personalised recommendations (1) were more actively redeemed compared to other methods, (2) could be effectively used to promote cross-sales, and (3) were able to reactivate inactive customers, each of which is important KPIs for the retail customer. Based on these insights, the retail customer decided to expand their marketing strategy towards more personalised advertisements.

Case Studies