LAUNCHMongoDB 8.3 is built for the sub-100ms retrieval & zero downtime AI demands. Read blog >
AI DATAStop fighting your data layer. Get the memory & retrieval agents need to scale. Read blog >

Kiabi leaves behind legacy limits for cloud leadership

Kiabi sought to escape 70% uptime and failing legacy Oracle servers by modernizing its data core to support global scale and omnichannel retail.

Photo of company employees.

Their Challenge

Kiabi sought to escape 70% uptime and failing legacy Oracle servers by modernizing its data core to support global scale and omnichannel retail.

Our Solution

Kiabi migrated to MongoDB Atlas on GCP, building a multi-region cluster to secure its sales repository and loyalty data with 100% resilience.

Outcome

Kiabi achieved 100ms response times and high availability, using 13 MongoDB Atlas clusters to scale through its most successful sales year ever.

industry_enterprise

Industry

Retail

atlas_product_family

Product

MongoDB Atlas

atlas_for_edge

Use Case

Modernization

THEIR CHALLENGE

From on-premises legacy limits to cloud growth

Kiabi is one of Europe's most recognizable value fashion brands. Over four decades, the French retailer has grown into a 33-country operation with more than 640 points of sale worldwide. With a network spanning France, Spain, Italy, North Africa, and the Emirates, the brand processes a massive volume of transactions to keep pace with its mission to make quality designs accessible.

But by 2021, Kiabi’s technical infrastructure was at a crossroads. While the business was rapidly evolving to meet new consumer trends, such as the secondhand market and omnichannel shopping, its backend was anchored in the past. The company depended on on-premises infrastructure built entirely on Oracle databases and legacy application servers.

“There were power outages, machines were breaking down all the time—it was hell to maintain,” recalled Romain Poiré, Developer Experience Engineer and Lead of the Development & Digital Lab at Kiabi.

Under constant pressure to keep the hardware running, the IT team frequently dispatched staff to resolve server issues, creating significant operational pressure. With an uptime of around 70% in the previous environment, payments could be processed in-store without ever being successfully sent back to the central system, so staff had to painstakingly re-integrate all transactions.

Beyond resource issues, the legacy architecture also struggled with the sheer scale of modern retail, particularly during peak seasons. “We can multiply them as much as we want,” said Poiré. “But when we get to the sales or the back-to-school periods, the physical machine ‘explodes’ after a while—it has its limits.”

To break this cycle, Kiabi launched a ‘move to cloud’ initiative, but knew that simply shifting virtual machines wasn't enough; it needed a fundamental change in how data was handled. MongoDB emerged as the catalyst for change, moving from a technology used only by a few developers to the cornerstone of Kiabi’s future vision.

Kiabi logo
“Today we have a huge number of needs around document transactions and we know that MongoDB can do it natively, easily, and with many advantages.”
Romain Poiré
Developer Experience Engineer and Lead of the Development & Digital Lab at Kiabi.

OUR SOLUTION

Building a modern sales repository on MongoDB Atlas

To future-proof the business, Kiabi migrated its infrastructure to Google Cloud Platform (GCP). Central to this transformation was the company’s decision to adopt MongoDB Atlas for mission-critical services, including the sales repository and a revamped loyalty program.

The sales repository is the heart of Kiabi’s data flow; it captures every transaction from 640 stores and the kiabi.com website, and also calculates points for their loyalty program. Faced with 10,000 transactions per store daily during peak periods, Kiabi needed a data platform that offered native document flexibility and seamless scalability. 

“Likewise, resilience is extremely important,” said Poiré. “I can't imagine having to tell clients, ‘You’ve lost the 600 points you accumulated throughout the year—your discount has disappeared’.”

To build internal momentum, Kiabi appointed ambassadors to champion the transition, and supported their advocacy with a quick-win project to demonstrate the platform's power: while their legacy setup required nearly a week for deployment and monitoring, the team proved they could achieve full self-sufficiency on Google Cloud and MongoDB in just two days.

While the shift was initially met with some resistance from database administrators (DBAs), the developer community was quick to adapt. To bridge the gap, Kiabi worked with MongoDB to provide intensive training, helping DBAs realize that MongoDB's operating principles were a natural evolution of their existing expertise. “We sought support from MongoDB to drive change across different levels of the company,” said Poiré.

Kiabi implemented an active, multi-region MongoDB Atlas cluster architecture. To ensure maximum resilience and meet strict Disaster Recovery Planning (DRP) requirements, the primary cluster resides in Mons (Belgium), with secondaries in Germany and France. This setup enables the system to handle the massive write-heavy load of retail receipts while maintaining high availability.

"Today we have a huge number of needs around document transactions, and we know that MongoDB can do it natively, easily, and with many advantages," noted Poiré.

Kiabi logo
“As of today, we haven't encountered the problems we had before. We've won a lot in scalability, security, and speed - especially when developing structural architecture that works well and can be duplicated.”
Romain Poiré
Developer Experience Engineer and Lead of the Development & Digital Lab at Kiabi.

OUTCOME

Resilience at scale and 100ms response times

The transition to MongoDB Atlas has fundamentally changed how Kiabi handles its most critical data. The retailer now runs 13 production clusters built on a scalable data infrastructure that automatically adjusts to meet the intense demand spikes of the fashion retail calendar. 

The impact on performance and reliability has been transformative, moving from 70% uptime on legacy systems to consistently high availability with MongoDB Atlas.

“As of today, we haven't encountered the problems we had before,” said Poiré. “We've gained a lot in scalability, security, and speed—especially when developing structural architecture that works well and can be duplicated.”

Indeed, the technical metrics tell a story of complete efficiency. The platform delivers lightning-fast performance, maintaining response times of around 100 milliseconds even during peak traffic, ensuring a seamless customer checkout experience. It supports nearly 500GB of live data, securely replicated across three regions to protect integrity and availability. Built on a modular architecture, the system also enables rapid innovation, allowing new capabilities—such as a loyalty points program managing millions of rewards—to be integrated with complete resilience.

This new technical foundation was put to the test recently during Kiabi's most successful year for sales on record. MongoDB Atlas's agility enabled the team to quickly develop a receipt-retrieval application to support new French laws on digital receipts and radio-frequency ID security.

“We have completed the retail equivalent of a heart transplant,” said Poiré. “With MongoDB, we’ve transplanted sub-modules, and we continue to transplant more sub-modules on top. It’s modular and easily adaptable to many things.”  

By moving away from the Oracle foundations of the past and embracing a developer-driven, cloud-native approach of the future with MongoDB Atlas, Kiabi has ensured that its technology is no longer a constraint but a catalyst for its global growth.

Run MongoDB without the operational burden

Atlas is the simplest way to deploy MongoDB. Get global resilience, push-button scalability, and advanced security.
Learn More
Illustration of a database stack

Explore more success stories

View all stories
Novo Nordisk logo
With Video

Novo Nordisk

This Danish pharmaceutical giant became the first in the industry to generate a complete clinical study report (CSR) in minutes with generative AI and MongoDB Atlas.

Read more
Toyota Connected logo
With Video

Toyota Connected

See how Toyota Connected migrated to Atlas and AWS to enhance reliability for its safety platform.

Read more
L'oreal Groupe logo
With Video

L'oreal Groupe

Discover how L’Oréal improves app performance and velocity with MongoDB Atlas.

Read more

Take the next step

Get access to all the tools and resources you need to start building something great when you register today.
Get StartedTalk to an expert
Illustration of a database.