Willis Flags Insurance Overbuying in the Data Centre Boom
Insurance broker Willis, part of WTW, is calling on data centre owners, developers, builders, operators and investors to fundamentally rethink how they buy insurance. The firm argues that as digital infrastructure expands rapidly to meet AI‑driven demand, many organisations have focused on securing ever‑larger insurance towers without first quantifying what they truly need — leading to coverage that may far exceed their actual risk profiles.
Willis notes that global markets could deliver up to US$15 billion in capacity for large data centre risks if needed, but the more pressing question is how much cover is actually required after a rigorous evaluation of exposures across the full lifecycle — from site selection and power infrastructure to climate vulnerabilities and cyber threats. The broker has introduced an eight‑point digital infrastructure risk framework designed to help stakeholders move from capacity‑led buying to a risk‑ and data‑led approach.
“Buying more insurance is not always the same as being better protected,” said Alastair Swift, Willis’ Head of Global Specialties & CEO. “When risks are properly modelled, understood and mitigated, clients can build more efficient, resilient insurance programs that reflect their actual exposures.”
Why Risk-Led Insurance Matters for Data Centres
Moving from Capacity to Exposure‑Based Buying
Willis’ intervention reflects a broader shift in commercial insurance away from conventional limit‑driven strategies. For data centres — assets whose values are enormous but whose risk profiles can vary wildly based on location, design and operational practices — a blanket approach to capacity often results in unnecessary premium spend. By first mapping out how much could actually be lost in realistic scenarios, owners can align cover tightly with exposure, potentially freeing up capital for other resilience investments.
Building Resilience into Design Lowers Insurance Needs
The broker specifically highlights early assessment of natural hazards and climate risk as a way to embed resilience from day one. Measures such as flood protection, seismic reinforcements and wildfire mitigation not only make an asset more robust but also demonstrably improve its risk profile — something that insurers, lenders and investors price in. This cost‑benefit lens gives developers a concrete way to justify upfront spending on mitigation.
What This Means for Lenders and Investors
The advice also targets the financing community. Lenders and equity partners typically demand high‑level protection, but Willis argues a tailored, data‑backed programme often satisfies those requirements more effectively — and at lower cost — than a generic large‑capacity tower. A sharper risk narrative can also smooth discussions with insurers and improve terms.
What Data Centre Stakeholders Should Do Differently
- Quantify exposures across design, construction and operations using a lifecycle framework, not just asset values.
- Model realistic loss scenarios instead of relying on generic market conventions; stress‑test for site‑specific perils like flood, wind, seismic activity and cyber.
- Assess natural hazard and climate risks early when selecting sites and incorporate resilience measures — from flood defenses to wildfire mitigation — to strengthen insurability.
- Use verifiable data on your infrastructure dependencies (energy, water, cooling) and continuity plans to back insurance negotiations and lender discussions.
- Review your insurance tower against your actual maximum probable loss to identify and cut capacity that offers little additional protection.
Risk & Opportunity Assessment
| Commercial Risk | Medium | Over‑insuring wastes capital and raises costs for data centre projects; however, a shift to risk‑led buying can correct this without leaving assets exposed. |
| Competitive Risk | Low | The issue is industry‑wide and not linked to any single player’s market share; adopting better practices could even become a differentiator. |
| Regulatory Risk | Low | Insurance buying itself is not a direct regulatory target, though lenders’ mandates for coverage levels must still be met. |
| Reputation Risk | Low | No reputational harm arises from over‑insuring; the concern is purely financial efficiency. |
| Technology Disruption | Medium | The rapid evolution of AI workloads and digital infrastructure is changing risk profiles, demanding new modelling tools that many organisations may lack. |
| Commercial Opportunity | High | Aligning insurance capacity with true exposure can significantly reduce costs, free up capital for resilience, and present a more compelling risk profile to investors and lenders. |
Comments 0