Cloud Native: the time has come for banking
How can you compete with the agility of a FinTech startup or an online-only bank when you are a long-standing financial institution? While traditional banking IT systems may be large enough to securely process millions of transactions each night, their static set-up prevents them from efficiently addressing the concerns raised by the Revised Payment Services Directive (PSD2).
On top of that, it also hampers adoption of the appropriate measures needed in our digital and real-time era. With this in mind, how can banks prepare for the advent of open banking? One solution is the introduction of new practices in application production and deployment: welcome to cloud-native architecture.
What are the challenges and stakes surrounding this transformation?
The emergence of online-only banks has led to a similar trend as that witnessed in the telecoms sector, whereby industry standards are no longer defined by the traditional market players. This role now falls upon the new arrivals, who are quick to draw inspiration from the models used by the big names in the technology world.
A study by Simon Kucher & Partners and Research Now estimates that, in France, the market penetration of pure players in the banking sector in 2016 was around 10%. Despite being a relatively small figure, the analysis highlighted the fact that this represented a 20% rise on the year before, in a market where the most promising newcomers had yet to officially plant their roots.
Their financial tools are mobile first, accessible 24/7 from anywhere on Earth. They are underpinned by versatile infrastructure and able to quickly develop new services while complying with regulations regarding customer intelligence and fraud prevention.
Traditional banks must not waste any more time before embracing the revolution, not only to contend with this new competition, but also to meet regulatory changes that are heading in the same direction. The PSD2, for example, encourages production and distribution of banking services to be separated in order to promote an open banking model. It thus recommends the adoption of APIs that will securely expose a part of the operations recorded by the banking IT system.
On the technical side, the challenge is enormous, since banks must find a way to rapidly integrate new technological standards into systems that are historically extremely siloed. How can this transformation be tackled without reinventing the way that applications are designed, developed and implemented?
Cloud Native: the cornerstone to this transformation
Nowadays, every infrastructure project is designed as a scalable and expandable platform, whose operation costs adjust to the load. Private or public, the cloud seems a rather obvious choice. As IDC notes in its 2017 predictions of changes in the financial services industry, the adoption of cloud technology could allow top-tier banks to reduce infrastructure spend by 25% by 2019. However, there is still the need for applications capable of exploiting the full potential of such flexible infrastructure.
It is here that the cloud-native approach makes its entrance, promoting a different school for designing applications so that they work as a group of services that are developed, deployed and tested independently from each other. Each with a very precise mission, these microservices are designed to operate as isolated processes that can be deployed, monitored, exploited and updated without the need to follow any technological uniformity.
With this practice, we no longer simply manage a project but a group of elements, from design to exploitation. Using the technological foundation established for the whole project, each team is responsible for their schedule and technological choices, allowing them to find the most efficient solutions to their individual issues.
This new paradigm helps moves towards a continuous deployment model in keeping with the DevOps philosophy, where one team both designs and uses the service on a daily basis. It is also this same team which is in charge of upgrading the service in an agile manner, according to user needs and environmental obligations. The model is thus the antithesis of the traditional, monolithic design in which all business components are stuck together and deployed in one single block.
In this regard, Netflix is a textbook example. The company began its cloud transition in 2008, following a corrupt database that plagued its entire IT system for several hours. At the time, Netflix decided to go with Amazon’s public cloud (AWS). Over the seven years that followed, they progressively migrated their entire system. The organisation now public acknowledges that it could never have expanded so quickly had it not adopted this cloud-native approach and redesigned its tools and software around hundreds of microservices and an on-demand scalable infrastructure.
Banking: back to the wall or facing the future?
The entire banking industry is now faced with the challenge of shifting its IT system to a more flexible and agile model. Many have already begun this transformation, but the dynamic should be supported and backed at the highest levels given what is at stake with open banking and FinTech. This transformation requires some services to be hosted in the cloud, rebuilding core banking as a scalable platform capable of meeting customers’ (current and future) needs.
Agile methodology itself teaches that the method should depend on the context: it is therefore upon the banking sector to find the right way to lead this transformation. It is a complex path, especially for Tier 1, 2 and 3 establishments. However, everything is possible from the moment that the organisation has been revised and adjusted to embrace this approach.
Spotify, for example, has shown it’s possible to deploy the model on a large scale: its 1,500 developers are split into ‘squads’, small teams of 5 to 7 engineers under the supervision of a Product Owner who autonomously manages their project within the company. Yet, that doesn’t mean they are isolated: various talents are encouraged to regularly discuss work and feedback to move forward together.
In its 2017 predictions for the financial services industry, made in December last year, IDC estimated that 15% of banks would adopt behavioural fraud analysis tools within three years. It also forecast that half of banks would be using cognitive, robotic process automation and blockchain technology by 2020.
To remain competitive and take advantage of open banking, banks must succeed in transforming their IT systems into banking service platforms. In other words, flexible tools, based on APIs and microservices, that will enable them to open up to partners while concentrating their resources on services that bring genuine added value.
Jocelyn N'Takpe & Olivier Ducarme - Digital Architect Manager