Banks rarely set out to customize their core heavily. It happens one reasonable request at a time: a product that works slightly differently, a report the regulator asked for, a screen the branch prefers. Five years later, every upgrade becomes a project of its own.
What customization really costs
The build cost is the smallest part. Each change has to be re-tested, re-applied and sometimes re-written with every release from the vendor. It slows patches, delays new features and ties the bank to the few people who understand how the change was made. Over the life of a core, the upgrade tax is usually far larger than the original build.
Start with the product, not the process
Modern platforms such as Finastra Essence cover retail, SME, corporate and Islamic banking out of the box. The first question in any fit-gap should be whether the bank can adopt the standard process, not how to recreate the old one. Show business owners the product working with their own data early, so they compare real options rather than imagined ones.
Push differences to the edges
Where a bank genuinely needs something different, build it outside the core: in a workflow, an integration layer or an extension that uses supported APIs. The core stays standard and upgradeable, and the difference stays visible and owned.
Govern every exception
Treat each customization request as a decision with an owner, a cost and an upgrade impact. Visible trade-offs lead to better choices. A small design authority that meets weekly and records every decision is usually enough.
What it looks like in practice
ZB Holdings completed a full transformation in 12 months with zero core customizations and migrated 750,000 customers to the new platform. The bank now takes upgrades as routine releases rather than as programs.
Working on something similar?
Talk to our consultants about how this applies to your bank.