The insurance industry's biggest transformation challenge is no longer access to technology.
Most insurers already have access to AI tools, cloud infrastructure and increasingly sophisticated data capabilities. Yet many continue to struggle to convert experimentation into operational change, according to Clara Silvestri (pictured), chief transformation officer at Allianz Partners.
Silvestri joined Allianz Partners in March 2025 after serving as Chief Financial Officer for Western Europe at Microsoft. She sees technology as only part of the transformation equation.
"Transformation is often seen as technology rollout," she said. "Many insurers believe implementing AI automation alone delivers transformation. The harder part is really redesigning the processes, the operating models, the customer journeys around it."
That distinction, she argued, is becoming a defining fault line across the industry. The organisations making progress are not necessarily those with access to better technology. They are the ones finding ways to embed it into everyday operations.
For Silvestri, one of the industry's biggest obstacles is its tendency to remain in perpetual experimentation.
"Change fatigue comes because the speed at which technology is evolving brings constant disruption," she said. "It's important not to slow down. Try to stay away from proof of concept, from pilots, and go to live operations, otherwise it's tiring."
The observation goes beyond project management. Pilot programmes can demonstrate technical capability, but they rarely deliver transformation on their own. Employees are asked to adapt to new initiatives, new systems and new ways of working, often without seeing lasting operational change.
Insurers have legitimate reasons for caution, particularly when introducing technologies that affect customer outcomes or operate within tightly regulated environments. The challenge, Silvestri suggested, is ensuring that caution does not become inertia.
As a result, organisations can find themselves caught between innovation and execution, continuously testing what is possible without fully committing to what works.The insurers making the most progress are those that have moved beyond experimentation and embedded change into day-to-day operations.
"The project is not over when it is delivered," she said. "It is continuous improvement. You learn, change, improve, deploy, and then you continue."
The distinction has significant implications because a pilot measures whether technology functions. Transformation requires an organisation to change around it.
The difference becomes clearer when technology reaches the customer. Silvestri pointed to roadside assistance, where AI may handle the initial stages of an interaction by gathering information, verifying coverage and coordinating support. But she described situations where efficiency alone is not enough.
"The range of assistance is very broad," she said. "You can go from claiming a flight cancellation to asking for help while you're stuck on a motorway with your baby in the car and cars speeding [past] you at night. The emotions are different. The system you need to give is different."
The significance of the example is not the technology itself. AI-powered customer service is becoming increasingly common across the industry. What matters is that decisions about where automation ends and human intervention begins have been built directly into operational processes.
The approach has also produced measurable results. Since introducing AI voice agents into its roadside assistance operation, Allianz Partners has recorded a six basis point improvement in voice of customer scores, a metric Silvestri described as notoriously difficult to shift.
Customer satisfaction measures tend to be relatively stable, making incremental improvements difficult to achieve. For Silvestri, however, the broader lesson lies less in the technology itself than in the operational decisions surrounding its deployment.
"Trust is not just a value," she said. "It's actually a competitive advantage that underpins everything we do. Transparency is critical, customers need to understand when they're interacting with AI and be able to access human support when they need it."
Asked how the UK compares with other European markets, Silvestri described a sector that is embracing AI while maintaining a strong focus on regulatory oversight and customer outcomes.
"The UK is perhaps slightly more cautious, focusing heavily on regulatory compliance and ethical AI usage," she said. "But that focus on the customer, which the FCA is essentially enforcing, aligns with where transformation should be heading anyway."
She pointed to growing adoption of embedded insurance and AI-enabled customer interactions across the market. Yet technology itself was not the issue she returned to most often. Instead, she repeatedly highlighted the organisational challenge of change.
As AI reshapes workflows, insurers face growing questions about how roles will evolve. Programmes that fail to engage employees risk generating resistance at precisely the point where organisations need support.
"The roles are evolving significantly," she said. "People are part of the change and can drive the change. But you need to bring them along, not impose it."
That may ultimately be the lesson many transformation programmes are still learning. Technology is becoming easier to access, easier to implement and easier to replicate. Competitive advantage is increasingly determined elsewhere.
The insurers pulling ahead are not necessarily the ones experimenting with the most technology. They are the ones finding ways to make it part of everyday operations.
