Even before the COVID-19 pandemic, many FinTech companies and other non-bank competitors were on track to outpace traditional banks at creating the always-on convenience, personalized services and apps that customers were demanding. Then when the pandemic hit, banks were forced to temporarily close physical branch networks, and asked customers to migrate to digital channels. At the same time, most bank employees were required to work from home. Business Times estimated that 70% to 90% of banking staff transitioned to remote working within days of government lockdowns, paving the way for accelerated use of virtual communication.
The pandemic showed the banking industry, the accelerated need for digital transformation, as customers under stay-at-home orders desperately required access to digital, self-service tools to manage their money. “A bank chairman recently told me that there are thousands of manual banks, but in the future, there will be just a handful of digital banks. I think he’s right,” says Phil Cantor from Intellect Global Transaction Banking (iGTB), an IBM Business Partner. In fact, the IBM Institute for Business Value suggests that 59% of companies surveyed now plan to accelerate digital transformation efforts over the next two years.
Digital transformation: not as easy as entering a PIN code
For banks undergoing digital transformation, a seemingly simple business request like “open an online account” breaks down into multiple steps such as “online validation of the customer identity” to fulfill local “know your customer” (KYC) compliance requirements, which is further divided into a set of technical IT requirements cascading to the application development team. The DevOps team is then tasked with writing software extensions for existing IT applications to implement the additional capabilities needed to make the solution available to end customers.
Managing increased complexity with agile processes
To keep up with the frequent business requirements that accompany the efforts to transform clients’ banking experiences, most banks have adopted agile project planning and execution processes. The mantra of agile is to create applications in small, controllable steps, test each step before it is considered complete, then collect feedback in real time and course correct before building on top of the delivered component.
Agile stresses the need to deliver value to the end user in much shorter timeframes than traditionally achieved in a waterfall development process. People and collaboration, as well as the focus on outcomes, are more important than process and rules. Continuous planning, testing and integration are the foundation of any agile project.
Helping a real-world DevOps team succeed faster
A major banking group in Southeast Asia with more than USD 100 billion in assets and more than 15 million customers across 10 countries needed faster time-to-market for their new digital banking services. In addition, they required an improved, more granular change control process.
The client’s line of business (LOB) teams were early adopters of agile planning practices to tackle growing customer demands for their new or enhanced digital banking services. The LOB teams used Atlassian Jira software for their agile planning, while the internal IT development teams used a range of tools including, Git, Jenkins, and Artifactory to implement agile and DevOps practices.
But the banking group wanted to streamline the developer experience for both the LOB and IT teams. They required an integrated, automated DevOps platform with built-in agile planning capabilities and automated bi-directional integration to each team (see Figure 1). This would allow the teams to continue to use the tools they found most effective while ensuring seamless collaboration.
Figure 1. The agile DevOps platform
Better insight and predictability along with seamless integration with existing tools
The client decided to standardize the internal IT development team on IBM Engineering Lifecycle Management platform while continuing to support the LOB users with Jira. The IBM Engineering Lifecycle Management platform extends the functionality of standard ALM tools, providing an integrated solution that offers full transparency and traceability across all developer data. This solution also offers developer-centric capabilities such as source code management, build automation, continuous integration and continuous delivery, along with Agile planning capabilities.
While the bank’s LOB teams still use Jira for continuous business planning, the banking group also deployed the IBM Engineering Workflow Management platform as the central tool for the IT developer community.
Agile planning occurs within the Engineering Workflow Management solution. Specific developer tasks are itemized and then linked back to the LOB-defined agile stories within Jira with IBM Engineering Lifecycle Integration Adaptors. All developer changes to the application code are tracked within the IBM Engineering Workforce Management and automatically linked against the detailed scrum tasks the developer team planned. By connecting projects, teams, and tools, the teams improve the speed of application delivery.
Minimizing chances of error for manual and repetitive tasks
The solution also provides continuous build and integration functionality. By using the Engineering Workflow Management platform’s capabilities, teams get automated integration builds on dedicated build hardware. The full audit trail generates a bill of material for every single build, outlining in detail which new tasks and pieces of source code went into the build and what software assets were created.
IBM UrbanCode Deploy was harnessed to provide continuous delivery across test and production environments. This further reduced the overall time from development to production significantly while implementing a fully automated and auditable delivery pipeline.
Reducing release cycles by 50%
IBM Engineering Lifecycle Management software provided the foundation for the banking team to continuously reduce release cycle times. The team has reduced release cycles by 50% and the executives say this has freed release engineers for other tasks and shortened the deployment window, allowing more deployments during the day while still keeping the financial institution in accordance with regulatory guidelines.