Balancing Agile and Waterfall in Banking Software Projects: A Pragmatic Approach

We are the software development department of a bank. Although the company has been pushing agile development from the top down, truly end‑to‑end agile projects are rare; business units still prefer traditional waterfall plans and commitments.

To C products and projects differ from our business projects – consumer‑facing products have uncertain requirements and users, allowing product managers to experiment and iterate quickly (MVP), which is unsuitable for long‑term planning.

In contrast, our banking projects require certainty: for each requirement we must state the time and budget needed, which agile methods cannot guarantee upfront, leading business stakeholders to view agile as an excuse for lack of planning.

Why business cannot accept MVP – While MVP works for market‑testing new consumer products, our business often follows existing competitors; offering only a minimal set of services would not win customers, so a full‑feature offering is required.

Waterfall and agile can coexist peacefully – Even when a project follows a waterfall schedule (WBS, dependency mapping, effort estimation, critical path, using tools like Microsoft Project), agile practices can still be applied where appropriate.

Our recent project involved setting up a fund‑outsourcing service for the business, which required extensive server architecture, environment configuration, security testing, and compliance with disaster‑recovery standards – a clearly waterfall‑type effort.

Nevertheless, we incorporated the following agile practices:

Using JIRA to record and track all tasks and changes, with daily stand‑ups and progress boards.

Applying Scrum with the external vendor, focusing each sprint on the most important deliverables to achieve faster, continuous delivery.

Splitting the overall plan to deliver the core system needed for licensing first, allowing parallel work on additional value‑added services, thus increasing flexibility.

Implementing deployment pipelines and automated regression testing for continuous deployment across environments.

In such business‑driven delivery processes, agile and waterfall are not mutually exclusive; a hybrid approach is often the most effective way to meet business goals quickly and reliably.