It is 2026. Artificial Intelligence is rewriting code, quantum computing is on the horizon, and we carry supercomputers in our pockets. Yet, if you try to access certain agent portals for some of the largest insurance carriers in the world—like Allstate’s EAXCIS portal—you are greeted with a message that feels like a time capsule from 2005: “This application is designed for IE browser”.
This live link is not just a quirk; it is a stark visual representation of a massive divide. While the global Property & Casualty (P&C) market has grown to an estimated $2.4 trillion, the backbone of the industry is still struggling with technological ossification.
Recent data reveals that 74% of insurance companies still rely on outdated legacy technology for core functions like pricing, rating, and underwriting. The reason is not ignorance; it is the immense complexity of “The Legacy Trap.”
Here is why the “Unkillable Ghost” of legacy tech persists, and why exorcising it is the most dangerous—and necessary—project a CIO can undertake.
1. The “COBOL Crunch” and the Talent Cliff
The reason a portal demands Internet Explorer isn’t just because the carrier hasn’t updated its website. It is often because the backend infrastructure relies on ActiveX controls or monolithic codebases written decades ago.
The industry is facing a “COBOL Crunch.” This 60-year-old programming language still supports 95% of ATM transactions and over 40% of banking and insurance systems. However, the experts who know how to maintain these systems—the so-called “COBOL Cowboys”—are retiring. With 25% of the insurance workforce currently over the age of 55, institutional knowledge is walking out the door.
Modern talent does not want to maintain green-screen mainframes. They want to work with Python, AI, and Cloud-native architectures. This skills gap forces carriers to rely on a shrinking, expensive pool of contractors, driving up the Total Cost of Ownership (TCO).
2. The “Open Heart Surgery” Risk
For giants like Allstate, State Farm, or Liberty Mutual, modernization isn’t a software update; it’s open-heart surgery. The failure rate is terrifying: 74% to 80% of legacy modernization projects fail to complete.
Why do they fail?
•Data Quality: Decades of data stored in silos or outdated formats is often “dirty.” Migrating this data is where projects go to die.
• The “Big Bang” Mistake: Attempting to replace the entire core system at once is a primary cause of the estimated $2.3 trillion wasted annually on failed digital transformations.
•Misalignment: There is often a disconnect between technical teams (focused on hardware dependency) and the C-Suite (focused on business competitiveness), leading to stalled funding.
3. The High Cost of “Keeping the Lights On”
Carriers are choosing operational stability over technological agility. However, this stability is an illusion that comes with a massive price tag. Research indicates that up to 70% of insurance IT budgets are consumed merely by maintenance, leaving a fraction for genuine innovation.
This technical debt is an anchor. While agile InsurTechs and MGAs can launch new products in weeks using low-code platforms likeBriteCore or Openkoda, legacy carriers relying on monolithic systems can take nine months or more to rate a new policy. In a market where speed-to-market is the new currency, this latency is an existential threat.
4. The Regulatory Vise: It’s No Longer Optional
The pressure to modernize is no longer just about profit; it’s about the law. New regulations, such as the Digital Operational Resilience Act (DORA), mandate strict tolerances for technical disruption and data recovery.
Legacy systems, which often lack automated failover and rely on batch processing rather than real-time data, make compliance nearly impossible. In 2025,94% of organizations reported that regulatory compliance strongly influenced their decision to modernize. The “Unkillable Ghost” is now a liability that attracts regulators’ attention.
5. The Path Forward: Think Big, Start Small
The giants are not standing still. They are finding ways to build a new ship while sailing the old one:
•Automated Refactoring: State Farm successfully migrated thousands of mainframe jobs to the cloud using an “automated refactoring” approach, converting legacy assets into modern Java applications without losing business logic.
•Value Stream Management: The Hartford overcame a stalled agile transformation by implementing Value Stream Management, aligning IT work with strategic business objectives to execute 350 projects in parallel.
•The Strangler Pattern: Smart CIOs are avoiding the “Big Bang” in favor of the “Strangler Fig” pattern—building new microservices around the edges of the old system and gradually migrating functionality until the legacy system can be safely decommissioned.
The Bottom Line
The industry is moving toward a hybrid future where “Platform Emperors” like Guidewire and Duck Creek dominate the core, but are increasingly challenged by composable, API-first architectures.
The message “Designed for IE” is more than a browser error; it is a warning sign of a business model running on borrowed time. The winners in 2026 will not be the insurers with the biggest history, but those who can dismantle the legacy trap before it dismantles them.
