Iryna Savchenko
FinTech Engineering Writer at INSART
11 June 2019

Managing Quick Company Scaling in FinTech: Case Studies

To win the market and acquire clients, FinTech startups need to be able to scale their businesses. This means adapting to new generations and capability to serve more clients as well as expanding software functionality and accommodating changing loads. Rob Freedman, Director of Marketing at Advisor Software (ASI), says this about the WealthTech industry:

“The wealth-management sector is experiencing disruptive changes [because] clients now expect to be served digitally, be it via a self-directed tool or with the help of an advisor. Wealth managers who leverage these technologies future-proof themselves and lay the groundwork for success in terms of scale, client services, and business growth.”

His words can pertain to any FinTech area. We talked to a number of CTOs of FinTechs, mainly in the WealthTech sector, to learn how they solve these issues and what they do on both the technology and business sides to enable their business and the businesses of their clients to grow.

From microservices to quick releases

For a technology company, the ability to rapidly expand software functionality is key to success. This is why continuous integration and deployment (CI/CD) methodology is becoming so popular in many FinTech companies. Rishi Srivastava, Director of Engineering at ASI, told us how they successfully carry out software development to allow clients—financial firms—to scale their businesses.

According to Srivastava, the ASI platform is a blend of legacy and modern products. To make these work efficiently, they employ microservices and use Amazon ECS and Fargate to make architecture scalable. They have very small development teams that service small components of the system.

They also understand that a one-size database approach doesn’t fit their highly distributed application and moved to microservice- and container-based architecture. This made the platform highly scalable and enabled quick releases. The ASI’s development and DevOps teams apply CI/CD to automate the software release process and deliver updates more quickly.

“When we get the requirement, we just transform that concept or requirement to reality [in] the fastest possible way.”

Freedman agrees that fast deployments save their clients money and headaches.

APIs and business

To ease the burden on financial experts, many FinTechs provide APIs. These allow experts to integrate new services into their own software so they don’t have to switch among several applications. Thus, some financial firms focus on their businesses and leverage technology. Tradier is one such company.

Jason Barry, CTO at Tradier, shared with us some details. Tradier’s APIs are designed to handle 200–250 millisecond responses, though they usually come back in fewer than 100 milliseconds. Integration with Tradier gives users many capabilities including access to historical and real-time market data, tools to manage user and account data, and trading on simple and complex equity and option orders.

Their approach made the platform popular among financial professionals. According to Barry, their client base continues to grow significantly; over the previous year, they had more than 200% growth. Today, Tradier executes nearly 300 million API calls per month, which is a few hundred thousand transactions per hour.

To keep up with this growth, Tradier had to improve its architecture, and it is now driven by AWS comprising over 50 servers/services. Understanding that most of the company’s sustained growth is long term, they decided to scale the system—but not on a daily basis.

“We’re scaling usually for weeks or months, and it makes it much more predictable from a billing perspective.”

RAM-based data storage

Some FinTechs aim to provide excellent, unique services to their users, whereas others decide that the software space is already crowded, and that there’s no need to build what has been done by others. The latter companies present themselves as digital hubs and integrate into any platform that might be useful for financial companies and enterprises. Trizic is an example of such a FinTech company.

To become a hub for integrating across many data sources, Trizic integrated with all custodians—data aggregators, quote providers, etc.

We talked to Steve Mays, when he was the company’s CTO, about the challenges they faced. Mays pointed out that they started with business process automation. The development team began scaling Trizic’s initial architecture. They used queueing-based microservices. This gave Trizic the ability to arbitrarily turn on additional Docker containers if a given service needed more performance or fault-tolerance at a very low cost and in an entirely automated manner. Queues provided fast, RAM-based data storage and strong assurances that data is only read once, while allowing for multiple services to take on work from that queue in parallel.

To enable their clients to scale their own business, Trizic collects feedback and reflects on those insights when polishing its services.

“Because the roadmap is flexible, we don’t create releases and try to cram features into it but prioritize client needs and minimum viable versions of features instead.”

Today, the team is focused on improving existing integrations with custodians, trust systems, and CRM as well as bringing new integrations online quickly.

Scaling down

Unlike many FinTech startups, InvestEdge started from creating a complex system with a broad base of services for ultra-high net worth investors. According to Roland Collins, CTO at InvestEdge, today, the company builds middle-office solutions to improve advisor’s efficiency. This enables advisors to ensure that everything is done correctly.

The company has designed a distributed processing engine—a computing framework that handles high-volume workloads in a horizontally scalable way.

“A lot of systems started out in the brokerage world and then moved up in complexity. We started with the ultra-high-net-worth world, and we’re building down in complexity.”

Takeaways

We have observed a few cases of how FinTech companies handle scaling platform functionality and increasing loads, as well as what they do to enable their corporate clients to grow.

Have you tried any of these approaches, or was your way quite different? What challenges did you meet? What would you do differently? Share you experiences with the FinTech CTO community!