How OM Bank built a digital bank and scaled to 1M customers with Redpanda

Challenge

Building a data streaming foundation for a digital-first bank

South Africans have plenty of banking choices. Established financial institutions, fintechs, and newly licensed digital challengers are all competing for customers who already bank with more than one provider. OM Bank set out to do things better.

“One might argue we don’t need another bank in South Africa,” says OM Bank’s Chief Technology and Data Officer Suvania Moodliar. “And we don’t need one. But customers do deserve better ones.”

That’s the purpose of OM Bank: to combine transparent, affordable banking products with data and technology that help customers “bank clever” and gain value through their financial decisions.

OM Bank is part of the Old Mutual Group, which has served customers for more than 180 years. The group operates in 12 countries across two regions, with more than 580 locations and ATMs in South Africa alone. While the role of a bank hasn’t changed, Suvania argues that banks have evolved in how they show up for their customers.

“We need to make sure that we're addressing different customer needs in the moment,” she says. “No one wakes up in the morning and goes, ‘I'm really excited to fill out a home loan application.’ What South Africans say is, ‘I want to buy a home for my family. I need to send my kid to school to get an education,’  and a bank is an enabler.”

Customers access services primarily through the OM mobile app, where they can automate savings, send money instantly, access credit, and speak to Themba, a digital financial advisor.

As a result, event-driven data sits at the heart of OM Bank’s operations. Every login, deposit, expenditure, and card swipe tells the bank how customers behave. These insights inform critical decisions like when to engage customers, what to offer them, and which channel to use.

Before OM Bank launched, the data team had a rare opportunity to design and build their data architecture from the ground up. And as a regulated bank, OM Bank had to keep its raw data within its own environment, with firm control over how it was stored and accessed.

“We designed the architecture so that it could serve the customer in the mobile app with real-time insights,” Suvania says. “We also wanted to bring that same data to rest and use it for reporting purposes.”

OM Bank was building for rapid growth, so their data streaming architecture had to handle increasing data volume without compromising resilience. Under South African banking regulation, Redpanda is classified as business-critical, which holds the platform’s performance and resilience to the same standard as the bank’s own.

Why Redpanda

A reusable event hub running in the bank's cloud

OM Bank considered Apache Kafka® but ultimately chose Redpanda Cloud with Bring Your Own Cloud (BYOC) on AWS. Redpanda offered the bank more control and flexibility while assuming the operational burden of running the streaming infrastructure.

“Being in control of data infrastructure and storage makes creating integration patterns quite easy. BYOC gave us a lot of agility and speed.”

OM Bank ran a proof of concept, validating the platform’s stability, data quality, ordering, and performance across the data platform and real-time applications. The teams also tested failover and worked through how the platform would scale.

“The technology itself—the ease of use, the user-friendliness, the way we can execute an implementation—was proof in the pudding in us getting up and running in a couple of weeks,” Suvania says.

In adopting Redpanda, OM Bank gained:

• Data control. Redpanda runs inside OM Bank’s AWS account, giving the bank control over access and visibility into where their data is stored.
• A reusable event hub. Instead of creating individual point-to-point integrations, the same data can be consumed wherever it’s needed.
• Simple integrations. Keeping the infrastructure within OM Bank’s environment makes it easier to connect partners and microservices within the wider Old Mutual Group.

Redpanda serves as OM Bank’s primary ingestion layer, pulling data from partner systems into their data estate. The data is then aggregated and enriched before being sent to a third party for recommendations and personalization, and ultimately back to the OM Bank customer recommendation team for engagement programs.

By combining real-time events with historical data, OM Bank can understand where customers are in their financial journey and respond accordingly. A customer’s behavior informs when the bank engages them, what it recommends, and whether that message is better delivered through the app or another channel.

“Every time that piece of data is reused, you’re getting value 10, 20, 30 times over for the same infrastructure spend.”

As OM Bank grew, Redpanda continued to align its support with the bank’s changing data and governance needs.

“The Redpanda team has been very deliberate in understanding the bank's business strategy and the different phases in which we're scaling our customers,” Suvania says, “so they could validate if their solution is sized to support what we need.”

Result

One million customers, zero unplanned failovers

Just over a year after opening to the public, OM Bank has passed the one million customer mark on a single Redpanda cluster. The bank now onboards between 2,500 and 4,000 new customers a day. Redpanda has continued running without a service incident or unplanned failover, and customers continue to share positive feedback about the mobile app's simplicity and responsiveness.

As a young bank, OM Bank treats its use cases as experiments rather than fixed plans. Redpanda gives the team the flexibility to experiment: if an event isn't adding value, they can enrich it, change it, or try something new without rebuilding their pipelines.

The architecture is also opening the door to new possibilities. OM Bank is enhancing their data lakehouse with Redpanda’s connectors and integrations while exploring how the platform can support their AI roadmap, from traditional machine learning models to generative AI.

OM Bank has built the reusable event hub they envisioned from the beginning, and they did it with Redpanda. Suvania compares Redpanda to the body’s blood vessels: one core system carrying data wherever it’s needed most, from the data platform for regulatory reporting and analytics to microservices and partner systems, all on the same infrastructure investment.

“As a new bank, we're curious—it's about progress, not perfection. Redpanda enables us to have flexibility of experimentation that makes the number of use cases possible."

Book a demo to see Redpanda in action and chat with our experts.