Développez l'apprentissage automatique prédictif avec Flink | Atelier du 18 déc. | S'inscrire

How Alex Bank Built a Real-Time Banking Experience with Confluent

Écrit par

Most people are not jumping for joy at the prospect of taking out a loan, or even going to the bank. If anything, banking is a chore (and not a very exciting one). But what if banking was fast, simple, and easier to understand?

That’s the goal of Australia’s Alex Bank. The completely digital banking startup, founded in 2018, offers personal loans, saving accounts, and term deposits. Like many other businesses, Alex Bank had to pivot when the pandemic hit—after getting delayed for its R-ADI (restricted authorized deposit-taking institution) license, which would allow the company to start taking deposits, the startup decided to replace all of its technology, rebuild its customer origination systems, and launch its personal loan product to the mass market.

Alex Bank eventually received its full banking license in December 2022 and has since hit several milestones, such as additional product releases and recording its FY23 financial results, marking its strongest year since the challenger bank was founded five years ago.  

At the Sydney, Australia stop of Confluent’s Data in Motion Tour, Alex Bank’s Chief Technology Officer John Heaton explained how the startup’s small but nimble tech team built a completely digital platform centered on delivering excellent customer experience using event-based architecture and real-time streaming.

The customer-centric vision

Getting a loan is like going to the dentist, Heaton said. “It's necessary, it's painful. And you actually have to go a few times in your lifetime to make sure you're on the right track. But other than that, you really don't want to do it.”

Alex Bank wanted to make going to the bank easier—and more enjoyable too. But a completely digital bank means no face-to-face interactions, no in-person branches to visit. How would the team create a friendly, empathy-driven banking experience?

Alex Bank sought to create a real-time bank, designed to be data-driven vs. process-driven. By using real-time data streaming, the startup is now able to start providing context and purpose to the customer, Heaton said. “It's no longer ‘give me your ID, let me do a credit check on you.’ You can start having a conversation with the customer—and that's not a conversation as in a chatbot,” he said. “But that's a real-time conversation as in when you ask the customer for information for the application journey, and you provide context and purpose feedback as they progress through the journey.”

It also means saving customers’ time and energy—if any services or processes are down, Alex Bank is still able to complete the customer’s journey. “We’re respecting the customer's time that they need to be in this process, and gathering the information from that customer at that point in time—and then letting them move on with their day,” Heaton said. With that in mind, the startup developed around four key tenants: simple, fast, fair, and human.

With customer feedback of 4.8 on Trustpilot—commenting on the speed, automation, and ease of the digital platform—and an NPS score of 74, Alex Bank has been delivering simple, fast, fair, and human support from very capable, and friendly staff.  

The bank has also seen other impressive milestones: as of June 30, the bank has lent over $74m in personal loans to date. The bank’s lending growth has been achieved through its strong commitment to cultivating and accelerating its broker channel, with the number of accredited brokers increasing by 272% in the last six months, and a majority of Australian major aggregators now onboarded.  

In addition, the bank leveraged its ADI to launch and expand deposit products, including a term deposit. The deposit portfolio has since achieved remarkable growth, boasting a compounded monthly growth rate (CMGR) of 74% in the six months following launch and is matching funding lending growth. 

Building the “nerve center” of Alex Bank

On the outside, it may seem like Alex Bank might not have any legacy tech to manage, Heaton said—but in reality, banking is an ecosystem with a lot of legacy tech. Interoperability became key: It was important that the team built a platform that could integrate with other systems over time.

Alex Bank turned to Kafka streaming because working with microservices was frustrating, Heaton said. Having individual pockets of information associated with each microservice can lead to divergence and lack of integrity of information. So Alex Bank built one stream for all its channels (broker and direct). While the events may be in different order, it’s a single data stream, Heaton said. 

“The only thing that we do is we actually pull in sets of attributes in and out to pivot, whether it's relevant or not relevant for the product,” he said. “But the data structure, the way that we deal and interact with it, the event-based architecture is all standard across all streams, all products.”

That allows Alex Bank to map business events to transactions. A loan process, treated as a single technical event, doesn’t mean much—but with an event-based architecture and real-time streaming, Alex Bank can stitch several separate events to a single business transaction through log files, correlation IDs, transaction IDs, and more.

Real-time streaming structure also allows Alex Bank to support a robust fraud prevention strategy. A full image of the application is taken in real time for every single event that happens on the stream, Heaton said. “We purposely did that,” he said, “so that we can start to interrogate every single version of the application to see if there's an anomaly.” 

This has allowed Alex Bank to provide complete transparency to both customers and regulators. The platform handles around 250 million security events on a monthly basis (around 9 million a day), which are all curated, interrogated, and triaged in real time.

Streaming has also allowed Alex Bank to do more with less. Building the platform took roughly six months with only three to five engineers at a time. Working with Confluent, Alex Bank was able to codify all processes, so engineers can spend their time doing their best work—and not doing click ops, Heaton said.

As the business has scaled exponentially over the last six months, the digital platform has easily met the demand with adequate capacity and capability to meet future growth targets—without the need for significant effort in re-engineering. Month-over-month, the bank has continued to scale rapidly, reaching a record high for lending settlements in June 2023—10 times higher than its December 2022 monthly results. 

The reality of real-time data

Alex Bank is able to gather 3,000 attributes in real time in less than 40 seconds on a customer, with 1,800 of those attributes used in its credit-decisioning model. That allows Alex Bank to tailor an instantaneous response unique to every individual—but getting such a fast response doesn’t allow the individual to build trust with Alex Bank. “The customer has gone ‘Well, I've just filled in my name, my address, my ID, I've given you some permission to go and gather some information. And you've told me immediately no. Feels a bit dodgy here,’” he said.

With the data at their fingertips, the challenge for Alex Bank was no longer the technology—instead, the team had to focus on building awareness and education with customers, so they understood how Alex Bank was able to make a decision on their applications.

“We understand that financial decisions in real time have gravity to our customers, right?” he said. “That's a stressful moment in their life, and they need to be respected.”

Join us in San Jose in September at Current 2023, the only event that brings together the data streaming community in real time. You’ll hear more use cases like Alex Bank’s and connect with other developers, architects, and tech execs on what they’ve built. 

  • Haley Kim is a freelance content writer based in Los Angeles. She's a former reporter with several years of experience writing for tech and media organizations.

Avez-vous aimé cet article de blog ? Partagez-le !

Win the CSP & MSP Markets by Leveraging Confluent’s Data Streaming Platform and OEM Program

This blog explores how cloud service providers (CSPs) and managed service providers (MSPs) increasingly recognize the advantages of leveraging Confluent to deliver fully managed Kafka services to their clients. Confluent enables these service providers to deliver higher value offerings to wider...


Atomic Tessellator: Revolutionizing Computational Chemistry with Data Streaming

With Confluent sitting at the core of their data infrastructure, Atomic Tessellator provides a powerful platform for molecular research backed by computational methods, focusing on catalyst discovery. Read on to learn how data streaming plays a central role in their technology.