Financial trading analytics and technology platforms
Finance & FinTech

Core Banking Modernisation for a Regional Commercial Bank

Client
Regional Commercial Bank, Singapore
Services
Cloud Migration · Core Systems · API Integration
Timeline
14 Months
42%
Faster transaction processing times post-migration
0
Unplanned downtime incidents during the entire migration
3x
Increase in concurrent digital account openings supported
31%
Reduction in annual infrastructure operating costs
The Client

A 60-Year-Old Bank, Running on 30-Year-Old Infrastructure

Our client is a regional commercial bank serving over 400,000 retail and SME customers across Southeast Asia. Founded over six decades ago, the bank had built a strong regional reputation but its core banking platform was a heavily customised mainframe system implemented in the early 1990s, maintained by a shrinking pool of specialists familiar with its codebase.

The system worked, in the sense that it processed transactions reliably. But it could not support the products customers now expected: real-time payments, instant digital account opening, or integration with modern payment networks. Every new feature request took months and carried significant risk of disrupting core operations.

The bank had previously attempted a full platform replacement with a different vendor a project that was abandoned after 18 months and significant cost overruns, having never reached production. When they approached DashMindsIQ, the primary requirement was clear: this needed to work, on a realistic timeline, without repeating that experience.

The Challenge

Three Constraints That Shaped Everything

CONSTRAINT 01

Zero Tolerance for Downtime

The bank's branches and ATM network needed to remain fully operational throughout the migration. Any extended outage would directly affect customer trust and regulatory standing.

CONSTRAINT 02

Regulatory Audit Trail

Every change to core ledger systems required a documented audit trail acceptable to the Monetary Authority's examination process no "move fast" shortcuts were viable.

CONSTRAINT 03

Institutional Trust Deficit

After the previous vendor's failed attempt, internal stakeholders were skeptical of any "full replacement" framing. The approach had to demonstrate value incrementally.

Our Approach

A Phased Migration Strategy Strangler Pattern at Bank Scale

Rather than proposing a "big bang" replacement, we recommended a strangler-fig migration: build a modern core banking layer alongside the existing mainframe, progressively route transaction types to the new system, and decommission mainframe components only once their replacements had proven stable under full production load.

01

Discovery & Mainframe Mapping (Months 1–2)

We spent the first two months mapping every transaction type processed by the mainframe, working alongside the bank's existing operations team to document business logic that had never been formally specified much of it existed only as undocumented behaviour in 30-year-old code.

02

Parallel Core Build on AWS (Months 2–8)

We built a new core ledger system on AWS, using event sourcing to maintain a complete audit trail of every account transaction satisfying regulatory requirements while enabling real-time processing. This ran in parallel with the mainframe, processing shadow transactions for validation without affecting live operations.

03

Phased Cutover by Product Line (Months 8–13)

Starting with the lowest-risk product (savings accounts), we progressively cut over transaction types to the new core, each phase running for 4–6 weeks in shadow mode before going live, with instant rollback capability maintained throughout.

04

Digital Channel Integration & Mainframe Decommission (Months 13–14)

With all transaction types running on the new core, we connected the bank's mobile app and online banking platform via new APIs, enabling instant account opening and real-time payments for the first time and decommissioned the remaining mainframe components.

Technology Used

The Stack Behind the Migration

AWSKafka KubernetesPostgreSQLEvent SourcingGoTerraformGraphQLISO 20022
"DashMindsIQ delivered our core banking modernisation programme on time and within budget something our previous vendors had repeatedly failed to achieve. Their domain knowledge of financial systems was immediately apparent, and the phased approach meant our branches never felt a thing."
Chief Technology Officer Regional Commercial Bank Singapore
← Previous Case Study HIPAA-Compliant Telemedicine Platform Next Case Study → Headless Commerce & AI Recommendations