Key points
- Westpac says AI-accelerated migration of the legacy St.George data warehouse cut costs from a budgeted $70 million to around $35 million.
- CEO Anthony Miller says AI tools sped up code conversion, testing and documentation, completing the migration more than two years early.
- The project produced a reusable AI data migration capability that Westpac expects to apply to other migrations for further benefits.
Westpac says that an “AI-accelerated” data warehouse migration for its St.George business enabled the bank to save $35 million in project costs, and produced a reusable model that should lead to further savings.
The bank briefly outlined how it used AI on the data warehouse migration at a digital, data and AI investor day in September. [pdf]
At the time, it was emphasized that AI had shortened the expected delivery time for the migration by two years.
But the bank went further in quantifying the financial benefits of the AI use case at its annual results. [pdf]
It’s fairly unusual for a technical project like a data warehouse migration to get its own slide in a bank’s financial results, let alone for the bank’s CEO to hold it up as an example of the organisation’s technical capability.
But CEO Anthony Miller said the work was a good example of how AI is helping the bank deliver Unite, the codename for a major technology simplification and transformation program.
“We are using AI within Unite,” Miller said.
“An example is the migration of the legacy St.George data warehouse. The team used AI tools to speed up code conversion, testing and documentation.
“This helped us complete the migration more than two years earlier at half the cost.
“We expect to start booking the benefits of this from FY27.”
On the slide, Westpac revealed that the data warehouse migration was initially budgeted to cost $70 million, but would now come in at around $35 million.
The migration is expected to deliver around $15 million a year in direct benefits.
The decommissioning of the old data warehouse is “planned for FY27,” Westpac wrote.
The destination for the data in that warehouse is what Westpac calls Adapt, which is essentially a cloud-based data and analytics platform built on Azure.
Miller indicated that additional financial and delivery benefits would now flow because, through the St.George data warehouse migration, the bank had been able to produce a repeatable, “reusable AI data migration capability”.
“The tools, methods and experience developed can be reused for other migrations,” Miller said.
“It is a good example of what can be achieved when AI capability is integrated [into programs of work].”
Reusable migration processes were a hallmark of banks’ previous embrace of cloud computing, where they were able to create and refine models for re-platforming applications from on-premises to cloud-based infrastructure.
Westpac reported a statutory net profit of $6.8 billion for the full year ended September 30 2026, which was down one percent compared to the previous year.
