The Gate and the Compass
Why insurers who treat risk as a stop signal are leaving growth — and customers — behind
In January 2025, several of the largest property insurers in the United States announced they would not renew policies for tens of thousands of California homeowners. The reason was straightforward: the wildfire risk was too high. The actuarial logic was sound. The business outcome was a disaster — not just for the policyholders left without coverage, but for the insurers themselves.
Those companies did not fail because they lacked risk data. They had more risk data than any generation of underwriters in history. They failed because they used that data in only one direction: as a gate. The data told them when to say no. It never told them how to say yes differently.
This is the central problem with how the insurance industry currently treats risk management. And it is the problem this series sets out to solve.
The gate that stopped working
Most insurers today are organized as multidivisional structures — separate units for different product lines and geographic territories, each managing its own profit and loss. This architecture was designed for a world where risk concentration was the primary threat. Keep the silos clean, keep the exposures separate, and the portfolio stays safe.
Within this structure, risk management flows in one direction: downward. Actuaries assess exposure. Underwriters set limits. Sales teams receive territories and restrictions. Marketing campaigns are deployed within whatever space is left. Risk is a filter applied at the beginning and end of the product cycle — a gate that opens or closes, but never guides.
The consequence is a binary vocabulary: sell or do not sell. High-risk territory? Do not sell. Dangerous concentration in a region? Stop writing policies. Unfavorable demographic profile? Decline the segment. Each decision, viewed in isolation, is defensible. Viewed together, they represent an enormous amount of value being destroyed — customers unserved, premiums unwritten, brand trust eroded — in the name of a risk discipline that only knows how to say no.
Risk management currently has one setting: off. What the industry needs is a compass — a tool that says not just whether to go somewhere, but how to get there, what to carry, and when to move on.
From gate to compass: a different organizational logic
A compass does not stop movement. It orients it. The same risk intelligence that today tells a sales team “do not sell in this region” could, with different organizational wiring, tell a marketing team “advertise here, but only this product, to this segment, with these incentives, at this price point.” It could tell a product team “design a policy for this peril profile that we can profitably underwrite at scale.” It could tell a CFO “here is how much premium we need to retain from low-risk regions this year to fund our reinsurance position in the next scenario cycle.”
This is not a theoretical possibility. The data infrastructure already exists in most mid-sized and large insurers. What does not yet exist — in most organizations — is the lateral connection between the team that holds that data and the teams that execute campaigns, design products, and set prices. Risk intelligence reaches the underwriting desk. It almost never reaches the marketing calendar.
Repairing that connection requires more than a new software integration. It requires a structural shift in how insurers understand what risk management is for. Risk analysis cannot remain a one-way, top-down flow. It must become a continuous feedback loop — with marketing tactics, product design choices, and incentive structures feeding back into risk models, updating them in real time.
The organizational implication no one wants to discuss
This idea has a name in organizational theory: it is the move from a multidivisional structure to an integrated one. It requires risk professionals to sit alongside marketing strategists, not above them. It requires campaign decisions to be reviewed through a risk lens before they are approved, not after they produce claims. And it requires the risk function to accept that it is not the only intelligence informing portfolio decisions.
This is uncomfortable because it redistributes authority. The actuarial function, traditionally one of the most sequestered in financial services, would become a collaborator rather than a gatekeeper. Marketing teams, traditionally the most removed from technical risk analysis, would become responsible for portfolio implications they previously never considered.
But the alternative — maintaining the current silo structure while market volatility, climate peril, and competitive disruption increase — is not stability. It is slow fragility. The insurers most exposed to the next California-scale event are not necessarily those with the highest geographic concentration. They are those whose concentration was invisible to their own marketing and pricing teams until it was too late to rebalance.
Proof of concept: when risk data designed the product
The clearest existing evidence that risk intelligence can drive product design rather than merely restrict it comes from usage-based insurance. For decades, actuarial data showed that male drivers under twenty-five presented materially higher accident risk. The industry response was binary: charge them more. The higher premium neither fully compensated for the excess risk nor created any incentive for safer driving.
Usage-based and parametric insurance models changed the logic entirely. By making premium contingent on actual driving behavior — speed, braking patterns, time of day — insurers converted risk data from a pricing adjuster into a product architecture. The result was policies that attracted lower-risk drivers within previously high-risk demographics, improved loss ratios, and — critically — expanded the insurable market rather than shrinking it.
That same principle, applied to territory management, geographic clustering, advertising spend allocation, and scenario-based premium retention planning.
Is this ethical? The question every reader is already asking
The honest answer is yes — but the reasoning matters.
A skeptical reader might argue that risk-aware marketing is simply a more sophisticated version of redlining: using data to systematically underserve high-risk populations, maximizing profit while abandoning the communities most in need of risk protection. That concern deserves to be taken seriously, not dismissed.
Here is the answer: the worst outcome for a high-risk policyholder is not a higher premium. It is default. An insurer that writes more policies than its portfolio can support, concentrates risk beyond its reinsurance capacity, and eventually fails to pay claims has done more damage to those policyholders than one that never wrote the policy at all. The current system — which relies on blunt exit decisions rather than portfolio-intelligent engagement — produces exactly that outcome, and it produces it without warning.
Risk-aware marketing, properly executed, does the opposite. By managing concentration actively, insurers can justify writing more policies in aggregate — including in higher-risk regions, up to a balanced threshold — because the portfolio as a whole is more resilient.
- More people are covered.
- More claims are paid.
- More communities retain access to the risk transfer mechanism that insurance is designed to provide.
The goal is not to sell less. It is to sell smarter — so that when the event comes, the company is still standing to pay the claim.
Conclusion
The gate and the compass are not competing philosophies — they are the same instrument, read differently. Risk data has always told insurers where danger lies. The shift being proposed here is simply this: that knowing where danger lies is also, precisely, knowing where opportunity lies — if the organization is structured to act on that knowledge with creativity rather than caution alone.
The insurers who default are not always those who took on the most risk. They are often those who managed it so narrowly that they never built the portfolio depth to absorb the inevitable.
A risk-intelligent marketing strategy is not a loosening of discipline. It is discipline applied with a wider imagination — one that asks not just ‘should we be here?’ but ‘how do we belong here in a way that is good for the customer, the portfolio, and the long-term solvency of this company?’ That question, more than any single tool or tactic, is what separates the insurers who will still be paying claims in twenty years from those who will not.
The insurers who will lead the next decade are not those with the most conservative underwriting desks. They are those who recognize that their risk intelligence is, in fact, their most powerful and most underutilized marketing asset.
Author’s note: The ideas in this series draw on organizational theory, behavioral economics, actuarial science, and marketing strategy. They are intended for insurance executives, marketing leaders, product designers, and risk professionals who believe the boundaries between those disciplines are ready to dissolve.