Missed our webinar with Generali Deutschland AG and UBS HAINER last week? Watch it at your convenience and hear how Generali is modernising test data management. SIMPLY CLICK HERE.
A QA Financial webinar on March 25, hosted by Matthew Crabbe, CEO of QA Financial, brought together Melanie Beul, Team Lead Test & Release Management at Generali AG, Bastian Baudisch, Lead Sales Engineer US at UBS HAINER, and Jan Goebeler, Consultant at findic, to explore how financial institutions are rethinking test data management amid rising complexity, regulatory pressure and AI-driven development.
The discussion quickly converged on a familiar but often underestimated constraint: test data.
“In the last couple of months… AI-assisted coding [has been] driving a desperate need for test automation,” Baudisch said. “We’re creating code at a faster and faster pace. The question is, how can QA keep up?”
His answer was direct: “We automate as much as we can and remove bottlenecks – and the most prominent bottleneck is test data.”
For Generali, that bottleneck had already become operational. Beul described how test cycles can fail not because of defects, but because the right data is unavailable.
“Imagine a DevOps team preparing for an important test phase… and then the test fails. Not because of a software defect, but because the right test data isn’t available.”
The result is often manual intervention, delays, and uncertainty around root cause. As Crabbe noted during the session, this creates risk when teams misattribute failures.
Baudisch warned: “When you get a lot of false positives that are attributed to test-data, QA might grow numb to real errors and dismiss them.”
From legacy constraint to scalable platform
Generali’s response was to move away from a long-standing, in-house mainframe-based approach that had become increasingly difficult to maintain and scale.
“It served us well for many years,” Beul said. “But after 20 years… the end of life was approaching.”
Rather than taking an incremental approach, the firm chose its most complex domain, health insurance, as the starting point for a new platform.
“It’s our most complex domain… we have a lot of dependencies… and we have to anonymise a lot,” Beul explained.
Working in phased development, the programme focused first on enabling on-demand data provisioning with minimal maintenance overhead. Goebeler, supporting the implementation, described how the work was structured: “We divided the request into four development phases.”
Following a successful proof of concept, the platform was rolled out more widely. By the end of 2025, Generali had migrated around 80 applications, integrated more than 1,000 database tables, and enabled over 500 users running approximately 19,000 jobs per year.
“We successfully shut down our old legacy application and our team can focus now on implementing, integrating new systems instead of maintaining,” Beul said.
Beyond the technical implementation, the discussion highlighted a broader shift in how firms approach data itself. Baudisch described the challenge as balancing realism, compliance and usability.
“The game really here is balancing realism versus convenience,” he said, pointing to the trade-offs between production, masked and synthetic data.
“AI-generated data is a non-deterministic method… there are risks that have to be managed.”
Balancing act
As AI accelerates development, that balancing act is becoming more acute. Faster code generation increases the volume of testing required, placing greater pressure on data availability and quality.
Baudisch pointed to regulation as one of the drivers behind this growing focus. “I could talk about increasing regulatory pressure and data complexity – but the real disruptor in the past months has been AI generatd code.”
But as the Generali case shows, the issue goes beyond compliance. Test data is emerging as a core dependency in modern software delivery, directly shaping how quickly and safely financial institutions can release code.
“If I can’t get the data that I need today… I might miss something,” Baudisch concluded.
For banks and insurers, that risk is no longer theoretical. It is becoming embedded in the day-to-day reality of QA.
WATCH THE WEBINAR HERE
QA FINANCIAL EVENTS
Why not become a QA Financial subscriber?
It’s entirely FREE
* Receive our weekly newsletter every Wednesday * Get priority invitations to our Forum events *
REGULATION & COMPLIANCE
Looking for more news on regulations and compliance requirements driving developments in software quality engineering at financial firms? Visit our dedicated Regulation & Compliance page here.
READ MORE
- Inside Rabobank: Engineering resilience by design
- Can AI agents finally automate data testing?
- Continuous testing drives DORA compliance
- Why software testing may face a major rethink
- Buy or build? AI rewrites software testing for banks
WATCH NOW
QA FINANCIAL PODCASTS
CLICK HERE TO LISTEN TO OUR EXCLUSIVE CONVERSATIONS
