Overview

The BFSI industry across the globe has been embarking on cloud migration initiatives over the last decade or so, like a lot of other enterprise scale industries. However, a lot of them also struggle with this move, and have had mixed results. Let's look at a few factors to consider, and look at how institutions can put together a coherent Cloud Migration strategy that allows them to unlock tangible, incremental gains on this journey. This post is the first in a series that will look at how BFSI enterprises could approach Cloud Migration alongside a wider tech modernisation journey, and covers the intent and motivating factors to migrate. We look at some of the potential factors that deliver underwhelming results with the migration as well. In future posts, we'll dive deeper into each of the high level topics covered here.

Why migrate to the cloud?

‍

To start with, there are a lot of potential benefits to be unlocked by Cloud Migration. There are also some pre-conceived mental models at play when people think about Cloud adoption in Enterprise scale organisations like BFSI.

So in order to clarify the thought process and decision making, let's start with what the high level benefits could be (if done effectively) from cloud migration

Time-to-Market and Competitive Advantage

Agility: Cloud adoption allows BFSI enterprises to quickly adapt to changing market conditions, and operate with a reduced time-to-market. This allows them to stay competitive. Tech innovation: Using cloud services also unlocks the opportunity to leverage technological advancements and managed services much easier than if it were to be done in a company-managed data center. This allows the in-house development efforts to focus on delivering business value faster.

Architectural Flexibility: Being on the cloud makes it much easier to enable offerings like Banking-as-a-Service, Platform and Ecosystem integrations as well as Omnichannel Experiences, further enhancing competitive advantage

Cost

Capital Expenditure: Cloud services eliminate the need for substantial upfront investments in hardware and data centers. Operational Cost Savings: Pay-as-you-go models and resource optimisation tools can help reduce ongoing operational costs. It also allows the institution to focus on delivering business value instead of having a lot of engineering capacity dedicated towards data center operations.

Reliability, Performance, and Flexibility

Automation and Operational Efficiency: Practices like Infrastructure-as-Code (IaC), Everything-as-a-Service (XaaS), CICD, and DevSecOps etc allow teams to have consistency, repeatability and speed of provisioning. We can also ensure aspects like Security and Governance are baked in to the software delivery cycle.

On-Demand Scaling: Cloud platforms allow BFSI enterprises to scale resources up or down based on demand, ensuring optimal performance during peak times. There are also fault-tolerance mechanisms and rapid spin-up of new resources possible with the right investments in observability and automation. Global Availability: Most cloud providers have globally available Data Centers (or Regions/Zones) as well as Edge Computing resources. This has implications on speed, latency etc, as well as in allowing geographical location based regulatory compliance. Resilience and Business Continuity: Cloud platforms provide robust backup and recovery solutions that ensure data integrity and availability in case of disasters. They also unlock the potential of multi-region or multi-data-center High Availability and Disaster Recovery mechanisms.

Security and Compliance

Security: Leading cloud providers offer robust security features, including encryption, access controls, and continuous monitoring, which can be more advanced than on-premises solutions. Compliance Automation: Cloud platforms offer tools and services to automate compliance monitoring and reporting, reducing the burden on BFSI enterprises. They also come pre-certified in a lot of regulatory Data Residency: Cloud providers offer data residency options to help enterprises comply with local data sovereignty regulations.

Developer Experience

Flexibility and Speed: Developers can focus on delivering business value quickly, and at scale, without being bogged down by the wait times, implementation details and inefficiencies in self-managed data center approaches

Technical Innovation: The availability of latest technology stacks in the cloud makes it easy for engineering teams to adopt new technology and techniques in the way they build solutions for customers.

‍

Antipatterns in Cloud Migrations

‍

‍

Given all these benefits, Cloud Migration should be a no-brainer, right? However, in our experience consulting across various organisations across geographies and at various scales, we noticed some patterns that lead to sub-optimal results. Among these are:

1. Lack of coherent strategy: Cloud migration is not a simple task even for smaller institutions. An enterprise scale BFSI institution needs to consider a lot of factors in adopting migrations. This strategy needs inputs from across the organisation, and a clear evaluation of current state, and a staged roadmap for the migration. Failing to thoroughly assess the current infrastructure and applications can lead to compatibility issues and performance problems in the cloud. Overlooking the complexity of legacy systems and their dependencies can result in significant challenges during migration. Misjudging the total cost of migration, including hidden costs such as data transfer, training, and post-migration optimisation, can lead to budget overruns.
2. Boiling the Ocean: There is also an all-or-nothing approach organisations tend to take in approaching overall “Digital Transformation” initiatives, as well as Cloud Migration initiatives. It might be more feasible to approach Cloud Migration incrementally, updating and migrating services that are ready for the cloud incrementally.
3. Lack of Cloud Readiness: Legacy applications may face compatibility issues with cloud platforms, requiring significant modifications or re-architecting. The applications and technology stack should be architected and set up in a way that enables them to reap the benefits of being on the cloud. However, a lot of times organisations tend to try to force-fit the existing stack into the cloud, often with a lot of wasted effort and cost. A lot of time this also needs to organisation to listen to inputs from the ground and not drive the migrations purely as a top-down exercise
4. Lack of Competencies and Skills: A shortage of skilled personnel with experience in cloud technologies can hinder the migration process. Insufficient training and support for employees adapting to new cloud environments can lead to operational disruptions.
5. Lack of Data Management : The institution should have a clear awareness of what data is at rest and in flow across their various systems. Then they need to have a coherent data migration and management strategy. Maintaining data integrity during migration is critical. Data loss or corruption can occur if proper protocols are not followed. Poorly planned data migration can lead to extended downtime and disrupt business operations. They also need to ensure the right security and privacy controls for data.
6. Governance: Poor governance, oversight and management can lead to multiple issues. Inefficient management of cloud resources can lead to resource sprawl and inefficiencies. Without a robust governance framework, managing cloud resources, policies, and compliance can become challenging. Lack of proper cost management and monitoring can result in escalating expenses during and after migration.

In order to avoid the above, institutions need to adopt a more structured and flexible approach to cloud migration.

‍

Structured approach to Cloud Migration

At Greyamp, we have what we call the 4 Levers of transformation and agility.

Too often, organisation wide Cloud Migration initiatives tend to focus on Technology and some of the Governance aspects, without devoting enough bandwidth to the wider scope it entails.

Keeping the above levers in mind, let’s look at some aspects that allow institutions to migrate to the cloud in a more pragmatic manner. We’re just touching upon these at a high-level, but will dive into details in subsequent posts.

Aligned Strategy, Decentralised Initiatives

A lot of times, a big-bang, transform-everything approach seems to be adopted in organisations, with Cloud Migration as a part of a wider set of “Digital Transformation” initiatives. However, managing an All-or-Nothing approach to change of this magnitude is often not feasible.

‍

What is needed is coherence and alignment across the organisation in terms of what needs to be achieved, but this should also be accompanied by decoupling and independence in how individual units are able to implement the transformation. This way, the Cloud Migration initiatives can focus on independently deliverable stages of work, allowing the organisation to unlock incremental gains without having tightly coupled initiatives that act as logjams.

Org structures and Governance

Team topologies and the right team alignments to unlock efficiencies of having teams operating at the right levels. Identifying what teams need to be stream aligned, what teams build domain-aligned services, and what teams provide platform and foundational capabilities is necessary. Then strategic and tactical work can be planned factoring in the interfaces and ways that requirements and information flow between these teams.

Platform Engineering and a Platform-as-a-Product approach

‍

The platform engineering function provides the infrastructure, tools, and processes that enable efficient and reliable software delivery. By abstracting away the complexities of infrastructure management and operations, platform engineering empowers development teams to focus on building and deploying high-quality applications. A "Platform as a Product" approach ensures the internal platform used by development teams is treated with the same principles and care as an external customer-facing product. This means applying product management practices to the internal platform to ensure it meets the needs of its users (i.e., developers, operations teams) effectively and continuously evolves based on feedback and changing requirements

Staged, incremental migration

‍

One major advantage of cloud based systems is that you can provision what you need, and incrementally provision more resources as the need arises, without up-front capital investment in infrastructure.

First, a platform engineering team can set up the foundational capabilities that allow application development teams to be onboarded to the cloud. Then there can be a staged onboarding process with early adopters, and suitable prioritisation, risk assessment etc can be run on them.

Then the services of the teams are made cloud ready, and migrated to the cloud. This will allow the Platform teams to validate their infrastructure setup as well as delivery automation with the early adopters, before scaling out to the rest of the organisation as and when their systems become cloud ready. It also allows the Platform team the ability to fix any gaps in their systems as well as in the developer experience before a wider roll-out.

Architectural Alignment and Cloud Readiness

‍

Legacy applications and systems will need to be assessed and evaluated to figure out the migration strategy for them.

It could be a combination of applying strangler-pattern approaches, re-architecting, refactoring or even a rebuild/replacement wherever needed. However, it is ideal to migrate the appropriate applications to the cloud. They should meet characteristics described in approaches like those described in 12-Factor applications etc.

Staffing, Competency and Skill Building

‍

A lot of this is not achievable without the teams having the appropriate skills across the board to achieve success in the migration initiatives. This includes:

1. The leadership having the right context, skills and knowledge to effectively manage and govern the project
2. The platform teams having the right skills to maintain a platform with reliability, speed, cost effectiveness and ease of use
3. The development teams having the right skills to architect applications that leverage cloud-native approaches and tooling
4. The business and finance teams having the knowledge and planning for the right investment and understanding the timeframes for returns on the investment in cloud migration

Conclusion

For cloud migration to be effective and to get all the benefits we can from it, we need to approach the migration keeping the above factors in mind. This post serves to set the context. In the next post, we’ll look through things to consider around cloud costs and FinOps. Subsequent posts will delve into details around each of the aspects mentioned above, and details on how they can be implemented.

‍