Carrefour’s operational data store its fit for the future with MongoDB

INTRODUCTION

To access and disseminate real-time information to its customers and managers, Carrefour has chosen a MongoDB ecosystem coupled with cloud architecture.

Regulatory requirements

In Romania, well-known French retailer Carrefour is completing new IT infrastructure that will support e-commerce and the company’s operational data store (ODS)—a database designed to centralize information from diverse sources for simpler analysis and reporting.

This is particularly relevant now that European regulatory constraints, collectively known as EURINV19, require customers to have automatic access to electronic invoices. EURINV19 is a project co-funded by the European Health and Digital Executive Agency (HaDEA) of the European Commission. Aimed at helping businesses and public entities adopt and comply with the European standard for electronic invoicing, EURINV19 facilitates seamless, cross-border e-invoicing within the EU.

Therefore, the first objective of Carrefour’s architecture upgrade is to automatically send receipts to customers’ smartphones in seconds. The second objective is to store all customer invoices in a modern cloud environment database set up for analytics.

THE CHALLENGE

Real-time insights

Until recently, customer receipts were sent to Carrefour’s data analysis platform, Google Cloud’s BigQuery, which is AI-enabled and fully managed. In collaboration with Google Cloud storage space, it enables analysis of an impressive amount of information. However, the tool is not designed to perform real-time operations, so Carrefour chose the MongoDB environment. “Thanks to this decision, we will be able to carry out transactions in the database and deliver the customer’s receipt to their smartphone application in a few milliseconds,” said Guillaume Blaquière, Group Data Architect at Carrefour.

Additionally, to fulfill European obligations concerning the creation of automatic and electronic invoices, Carrefour chose to use MongoDB Atlas because of its ability to use and extract data in real time. Traditionally, an electronic invoice is in a semistructured format that includes formal and generic data as well as specific information—such as a customer number or a purchase order number. When filed in a traditional database, this information is not usable. However, with MongoDB Atlas, Carrefour’s IT team can access it quickly and easily.

Simpler, faster, and more economical

The versatility of the MongoDB environment and its integration with Google greatly simplified the implementation of Carrefour’s ODS. In fact, MongoDB Atlas directly integrates into a virtual private cloud, which enables Carrefour to share or consolidate information, such as business results, between Romania and France securely. In addition, MongoDB Atlas enables the creation of real-time dashboards of store revenues. This means that managers can access all relevant information at a granular level and sort it in various ways, such as by department, by type of product, or by the hour. “It’s very simple and very fast,” said Blaquière. “MongoDB Atlas totally changes the way [we manage our data].”

This managed, cloud-centric solution will enable the Carrefour IT team to focus on value-added functions, such as selling products, optimizing the supply chain, and providing receipts. This is particularly valuable because Romania is a small country without a large pool of IT talent. Choosing a solution managed by MongoDB and Google greatly simplifies infrastructure management.

OUR SOLUTION

A mature and scalable solution

“While the MongoDB Atlas solution is easy to use and deploy, it also has the advantage of being scalable with increased or reduced computing power according to needs,” said Pierre Delivet, Sales Director at Google Cloud.

In addition, MongoDB can provide Carrefour’s IT team with in-depth information on its use of virtual machines and computational load and recommendations on increasing or reducing power. Access to all this information, and the resulting decrease in cloud computing costs, played an important role in Carrefour choosing MongoDB.

OUTCOME

Carrefour modernized its architecture to prepare for the future. Soon, it will be able to upload data directly to MongoDB and then replicate it in traditional databases or in BigQuery for analytics. This will simplify Carrefour’s IT architecture, making it ready for generative AI.

Moreover, MongoDB Atlas Vector Search will enable Carrefour engineers to use vector search and embeddings, making it possible for Carrefour to index the information for easy AI access according to user queries.

This new infrastructure in Romania is already of interest to Carrefour teams in Belgium, Spain, Brazil, and France. It will facilitate new business activities, such as Carrefour Links, which will enable business-to-consumer brands to launch communication campaigns via a wide range of online and offline channels, including Carrefour’s own website. The aim of this particular project is to monetize data—by capitalizing on the website’s traffic—as well as to offer hyper-personalization to customers and manage loyalty programs, with the ultimate goal of generating more revenue.

With the support of MongoDB and Google, Carrefour is continuing to migrate its tools to the cloud and continue the company’s digital transformation. “In the end, we will be more efficient with our customers, and our employees will flourish with new technologies,” concluded Blaquière.

To find out more, click on the MongoDB Atlas.