Related Services
GCP, Airflow, Github, Data Technology
Industry
Retail
Company
Xebia Data
fonQ is a popular webshop in and around the Netherlands for the home and living space, known for its wide range of furniture and lifestyle products. The company offers a variety of items from well-known suppliers and places a strong emphasis on its own private-brand products. These in-house brands are set to play a key role in fonQ's future, and as such, there is a pressing need to refine the pricing strategy for their products to stay competitive and maximize profitability.
The Journey Towards Data-Driven Pricing
Before this project, fonQ's pricing of private-brand products involved a manual, rule-based process. While this method provided a good foundation, it was time-consuming and could not utilize patterns in historical sales data, resulting in infrequent and less-than-ideal price adjustments.
In collaboration, fonQ and Xebia streamlined the transition to automated price recommendations through a structured approach and continuous feedback loops. To make this happen, we assembled a team with all the needed experts – data analysts, scientists, pricing professionals, and reporting specialists – who would manage the entire process from end to end. Due to this setup, we were able to go from a proof of concept to a fully integrated minimum viable product in no more than three months, from local to cloud environments and manual to continuous deployment practices.
Initial resistance arose over the model's suggestion to lower prices, as it seemed contradictory to increasing profit margins. However, multiple live tests demonstrated that the model's strategic price reductions led to a significant margin boost. This effectively discarded the doubts and underscored the value of data-led pricing decisions.
Fig 1: Absolute margin before and after going live with the pricing model. Slight differences before going live were caused by random variations and small tests we executed to validate the model. A ~25% improvement was observed for the selection of products for which we went live.
Margins from included products rose by an estimated 25%
The Final Outcome
The machine learning pricing system has significantly enhanced fonQ's pricing agility, enabling weekly adjustments that align closely with market conditions. This advancement has increased fonQ's monthly absolute margins from included products by an estimated 25% and has allowed the pricing team to make informed, data-driven decisions. The price recommendations were generally lower than previously adopted, so customers now benefit from more competitive prices, improving their shopping experience. This successful implementation exemplifies the synergy between fonQ and Xebia, setting a new standard for innovation in retail pricing strategies.