How the Hard Market Broke Commercial Insurance Operations

In 2019, a large commercial property placement might involve one carrier writing $400 million in limits. Today, the same risk is commonly split across 40 carriers writing $10 million each. The administrative workflow never adapted, and the consequences are now structural.

The shift

Commercial property insurance underwent a structural change between roughly 2019 and 2024. Carriers that had once been comfortable deploying large capacity on single risks pulled back. Limits that had been written in single blocks started appearing in layered, shared, and quota-share structures. The drivers were not mysterious: climate-driven catastrophe losses, reinsurance treaty hardening, and capital reallocation away from concentrated property exposure.

The operational consequence has been under-examined. The industry has discussed the hard market extensively in terms of rates, capacity, and coverage restrictions. What has received less attention is the transformation of the placement workflow itself.

A commercial property program in 2019 might have looked like this:

1 primary carrier writing $400M in total limits
1 set of policy documents
1 set of invoices
1 relationship to maintain mid-term

The same program in 2025 might look like this:

40 carriers writing $10M each across primary, excess, and shared layers
40 sets of policy documents
40 sets of invoices
40 mid-term relationships to coordinate

Every exposure change, every endorsement request, every mid-term update now has to reach 40 counterparties instead of one. The administrative work has not scaled linearly with the number of carriers. It has scaled super-linearly, because coordinating among 40 parties through email and attachments is harder than doing the same thing 40 times independently.

Where the complexity comes from

The fragmentation of capacity is only the first layer. The multi-hop structure of the global commercial property market adds more.

A large commercial property placement does not just involve the insured, the retail broker, and a list of carriers. It involves:

The retail broker holding the direct client relationship.
One or more wholesale brokers accessing markets the retail broker cannot reach directly, particularly for E&S capacity.
London-market brokers accessing Lloyd's syndicates and UK-based capacity.
Bermuda-market brokers accessing Bermuda-based reinsurance and specialty capacity.
Retail admitted carriers in US jurisdictions.
E&S carriers in the US non-admitted market.
Lloyd's syndicates and other UK-based carriers.
Bermuda carriers often providing high excess or catastrophe capacity.
Reinsurers sitting behind the direct markets.

A single exposure change, such as an insured acquiring a new property, has to propagate through this entire graph. The insured tells the retail broker. The retail broker notifies direct carriers on the retail side. The retail broker also notifies the wholesale broker, who notifies the E&S carriers. If London or Bermuda capacity is involved, the retail broker notifies the specialist brokers in those markets, who notify their respective carriers.

Each of these nodes has its own data format requirements, its own communication preferences, its own endorsement process, and its own timeline. A mid-term change that in 2019 involved one email and one endorsement now involves dozens of parallel workstreams, each of which must complete before the change is fully bound.

The multi-hop view

If you represent the commercial property placement workflow as a graph, with parties as nodes and data exchanges as edges, the structure begins to look less like a typical business workflow and more like a complex distributed system.

Each placement is, effectively, a multi-hop asset transaction. An exposure change initiated at the insured propagates through a broker network to reach carriers in different jurisdictions operating under different regulatory regimes. Unlike software distributed systems, which have transport layers and protocols designed for multi-hop coordination, the insurance equivalent runs on email and PDF attachments. Every hop is a manual rekeying. Every node maintains its own copy of the data. Every synchronization is an opportunity for drift.

The comparison to other multi-party industries is unflattering. A payment transaction touching a merchant, an acquirer, a card network, an issuing bank, and a consumer completes in a fraction of a second. An insurance transaction touching roughly the same number of parties takes weeks of email volleys to complete. The fact that the payment industry built infrastructure to make those handoffs invisible, while insurance did not, is the structural difference that matters most.

It is the reason Polysea was built to address this exact coordination problem. The hard market created a multi-party transaction requirement that no single participant can solve from inside their own systems. We are building the neutral infrastructure layer that makes the multi-hop coordination work automatically rather than manually.

By the numbers

A step-by-step analysis of a typical mid-term exposure adjustment in a multi-carrier commercial property program identifies approximately 101 discrete steps from the insured's change notification through to final policy update and invoice reconciliation across all carriers. Of those 101 steps, roughly 5% are currently automated in a typical workflow. The remaining 95% involve manual formatting, manual sending, manual acknowledgment tracking, manual reconciliation, and manual follow-up.

A 2025 survey of commercial insurance professionals found that more than half of respondents spend the majority of their time on day-to-day servicing rather than on renewal preparation or new business development. Among respondents, 63.3% spend more than four hours per week on exposure data handling alone, and 43.3% have experienced a lost or delayed placement because of data quality or version control issues. 100% still exchange exposure data through email and spreadsheets.

Even the CEO of Willis Towers Watson, one of the world's three largest commercial insurance brokerages, acknowledged in early 2025 that providing quality service is the firm's top challenge. This is a notable admission from a firm whose core function is service. It is also the kind of admission that only becomes speakable when the underlying operational problem has reached a point where denial is no longer viable.

These numbers are not symptoms of a few poorly-run operations. They describe the industry-wide baseline. The hard market has pushed the workflow into a regime where administrative overhead is consuming a meaningful fraction of the industry's operating capacity, and the existing tooling is not designed to keep up.

Why this happened

The hard market is a response to forces that have been intensifying through the 2020s.

Climate-driven catastrophe losses have risen sharply. NOAA tracks billion-dollar weather and climate disasters in the United States, CPI-adjusted, and the annual count has moved from a long-term average of approximately 9 events per year to 27 events in 2024 and 28 in 2023. Property insurance, which bears a large share of the economic cost of these events, has re-priced accordingly. Carriers have limited per-risk capacity to manage aggregation, which forces programs to layer across more carriers to reach the same total limits.

Reinsurance has hardened in parallel. Treaty pricing and retention levels have moved in ways that constrain how much direct capacity carriers can deploy on single risks. Fragmentation pushes further down the chain.

Capital allocation at the carrier level has reinforced the trend. Property catastrophe exposure now competes against other uses of capital on less favorable terms than it did a decade ago. Carriers that still write property are writing smaller pieces of it.

These forces are not transient. Even if the commercial property market softens in rate terms over the coming years, the underlying capacity fragmentation is likely to persist. The industry has crossed into a structural regime in which programs involve many parties per placement, and the administrative infrastructure needs to accommodate that permanently, not temporarily.

What should have happened

In most industries, a 20 or 40-fold increase in coordination complexity would have triggered infrastructure investment. New tooling would have emerged to absorb the complexity, standards would have tightened, and the workflow would have re-equilibrated at a higher level of automation.

In commercial insurance, this has not happened, or has happened only in scattered pockets. Individual brokers have built internal automation. Individual carriers have upgraded policy administration systems. Agency management system vendors have added features. None of these adaptations have addressed the cross-party coordination layer, because no single participant owns that layer or can unilaterally fix it.

The result is that the hard market's administrative burden has landed on individual human coordinators. Underwriting assistants, broker account managers, and carrier processors absorb the increased load by working harder, longer, and more carefully. This is not a sustainable equilibrium. The error rates, the burnout, the turnover, and the lost placements are all symptoms of a workflow being held together by human attention in a regime where the complexity has outgrown what human attention can reliably manage.

The operational consequences

Several things follow when coordination complexity outgrows the tooling.

Response times lengthen. A mid-term exposure change that once took a few days now routinely takes weeks or months. Not because any single party is working slowly, but because propagation across 40 carriers, through multiple broker layers, in multiple jurisdictions, cannot be compressed without infrastructure.

Error rates rise. Each additional party is another potential point of miscommunication, version drift, or data loss. Errors that were rare when a placement involved three parties become statistically predictable when it involves thirty.

Mid-term changes get deferred. If the administrative cost of a mid-term change is high enough, parties rationally defer changes to renewal. This accumulates risk misalignment: the policy's stated exposure diverges from the actual exposure, and the divergence is only corrected annually, if at all.

Coverage disputes become more common. With 40 carriers on a program, the question of which carrier is responsible for a specific loss, under which endorsement version, on which date, becomes a non-trivial forensic exercise. Disputes that would have been simple between two parties become complex negotiations across a layered tower.

Top talent is consumed by administration. Brokers and underwriters capable of senior-level risk work spend meaningful fractions of their time on data coordination. The opportunity cost of this allocation is the strategic work that does not happen: the risk engineering insights that go unidentified, the program optimizations that do not get designed, the client advisory work that does not take place.

None of these consequences are speculative. They are the current operating reality of the commercial property market. Every senior broker and carrier leader can recite examples from their own practice.

Where this ends

Industries do not stay in operationally broken equilibria indefinitely. Either the coordination complexity comes down, which would require the hard market to reverse more completely than current trends suggest, or the coordination infrastructure catches up with the complexity.

The second path is the more likely one. It is also the path the industry has not yet taken, because coordination infrastructure in a multi-party regulated market is hard to build, hard to fund, and hard to get adopted. These are solvable problems, but they have not yet been solved.

The conditions for a solution are clearer now than they have been in the past. The complexity has reached a level where no participant is defending the existing workflow as adequate. The technology, particularly AI-assisted document processing and structured data exchange, has become accessible in ways it was not even five years ago. And the economic incentive is large enough that early adopters stand to gain meaningfully from moving first.

What the industry needs is infrastructure built for the multi-hop, multi-carrier reality that the hard market has produced. A shared data layer that every party in the transaction can trust and build on without routing through a competitor. Authorization records that make the chain of who-did-what-when auditable across parties. Document infrastructure that treats invoices, endorsements, and binders as structured data rather than email attachments to be manually reconciled.

The hard market broke the existing workflow. The next chapter of commercial insurance operations will be defined by what gets built in its place.

Polysea is building neutral infrastructure for the commercial insurance ecosystem, including shared exposure data management, authorization chain tooling, and automated loss run extraction. If the problems described in this article are relevant to your work, we would like to hear from you at hello@polysea.ai.