By James W. Moore
Key Takeaways
- Agents and brokers don’t complain about carrier inconsistency. They route around it — quietly, permanently, and without warning.
- Carriers have almost no visibility into the submissions that never arrive. They measure what they see; they’re blind to what doesn’t.
- The worst consequence of inconsistency isn’t regulatory exposure or internal dysfunction. It’s the profitable business that silently goes to a competitor.
- Agents don’t stop sending all their business to an inconsistent carrier — they stop sending the good business. The difficult risks keep coming. The loss ratios follow.
- AI offers carriers the first real tool to measure their own consistency from the outside in — before the damage becomes irreversible.
Two recent articles on this site explored related ground. The Governance Problem AI Didn’t Create (But Might Actually Fix) examined how governance failures in insurance predate AI by decades. The Reluctant Auditor: What AI Sees That We’d Rather It Didn’t explored how AI surfaces internal patterns that organizations have long preferred to leave undocumented. This article examines what happens outside the carrier when those patterns persist long enough.
The Relationship That Ended Without a Conversation
A commercial lines broker in the Southeast has a regional contractor account. Good loss history. Stable operations. The kind of account most carriers would be happy to write.
Carrier X is a logical market. Their published appetite says so. The broker has placed business with them before. But she stopped submitting contractor accounts to Carrier X about six months ago.
Not because of a dispute. Not because of a commission cut. Not because a relationship manager said something wrong. The last three submissions she sent produced pricing that varied by nearly 40% with no material change in the risk profile. One underwriter requested documentation she had already provided twice. A fourth submission sat in referral status for eleven days with no update.
She never called to complain. She never asked for a meeting. She simply stopped.
Carrier X’s regional vice president still believes the relationship is in good standing. He mentions her agency when asked about key producers in the market. He has no idea she’s gone.
That’s how distribution loss actually happens. Not with a conversation. With silence.
The Invisible Pipeline
Carriers measure what they can see: submitted accounts, quoted risks, bound policies, hit ratios. What they cannot see — and almost never measure — is the submission that was never sent.
When an agent routes a risk to a different market, nothing appears in the carrier’s system. There is no declined submission, no lost quote, no record of a missed opportunity. The carrier’s data simply shows less activity from that agency, which is easy to attribute to market conditions, seasonal patterns, or a slow quarter. By the time the volume decline is large enough to notice, the relationship has often been functionally over for a year or more.
This is the invisible pipeline problem. And it is not a marginal issue.
According to the J.D. Power 2025 U.S. Independent Agent Satisfaction Study, only 57% of commercial lines agents say their carriers are meeting their foundational needs. More pointedly, 22% of commercial lines agents say they do not feel valued as partners by their carriers — and those agents are seven times more likely to write less business with that carrier than agents who feel valued. Commercial lines agents report satisfaction scores 314 points higher when working with a carrier is described as “very easy” versus difficult.
Agents are not being polite when they stay quiet. They are making efficient economic decisions. Every submission represents time, reputation with the client, and political capital. A carrier that burns those resources through unpredictable behavior simply stops being worth the risk.
Three Ways Inconsistency Destroys Trust
Pricing volatility. The same risk class, quoted by two different underwriters at the same carrier, producing numbers 30 to 40 percent apart. Agents notice. Not because they run spreadsheets, but because they submit similar risks repeatedly and develop a feel for what a market should produce. When the pricing feels like a roulette wheel, they treat the carrier accordingly.
Appetite drift. What the underwriting guide says, what the underwriter said last quarter, and what actually gets written are often three different things. The First Connect 2025 State of the Industry Report found that 71% of agents struggle to understand carrier appetites, with 64% experiencing quote decline rates between 10% and 50% as a direct result. Despite that, only 19% of carriers offer real-time appetite indicators. Agents submit into a fog and bear the cost when the fog moves.
Underwriter-level variance. This is the version of inconsistency that is perhaps most damaging because it’s invisible on paper. The carrier’s appetite guide hasn’t changed. The pricing manual is the same. But Underwriter A writes restaurants and Underwriter B doesn’t, and no one has documented that distinction. The outcome of a submission depends less on the risk than on who touches the file. Agents learn this. They adapt their submission behavior accordingly — and some stop submitting to that carrier for that class entirely.
Taken together, these three failure modes don’t just frustrate agents. They teach agents that this carrier is unplaceable. And unplaceable carriers lose distribution quietly, one class of business at a time.
The Problem With IRPMs
Individual Risk Premium Modifications — IRPMs — are worth examining here because they represent a microcosm of the broader inconsistency problem.
The original logic was sound. Manual rates price the average risk, and no risk is perfectly average. A structured mechanism allowing underwriters to fine-tune pricing within a defined band — typically 25% in either direction — was a reasonable tool for precision pricing. Apply debits to risks with adverse characteristics. Apply credits to risks that outperform the average. Price the individual risk more accurately.
What happened in practice is something different. The debit side became a margin tool. The credit side became a negotiation. Brokers who know the market and know the account find themselves in the position of petitioning an underwriter for a credit that should have been the starting point — essentially begging to get the rate to where it probably should have been all along.
Two brokers submit essentially identical risks to the same carrier. One has a strong personal relationship with the underwriter. One doesn’t. They get different outcomes. The tool designed to price risk more accurately ends up pricing the relationship instead.
There may be a less obvious cost embedded in that dynamic. If carriers deliberately set base rates conservatively and treat IRPMs as part of the expected negotiation, the assumption is that average premiums land higher than a market-rate-first approach would produce. What that calculation ignores is the broker who won’t play the game — the time-constrained producer, the agency without the relationship capital to push back, the account where the timeline doesn’t allow for negotiation. Those submissions quietly go elsewhere. The opportunity cost never appears in any report.
The Adverse Selection Trap
Here is where carrier inconsistency becomes genuinely dangerous, and where most carrier executives are not looking.
Agents don’t stop sending all their business to a difficult carrier. They stop sending the profitable business.
The clean contractor account with a strong loss history and stable ownership goes to the carrier the broker trusts. The marginal contractor account — the one with a complicated mod, a coverage gap, a client who is harder to place – still goes to Carrier X, because at least Carrier X might write it when no one else will.
Over time, Carrier X’s book of business in that class skews toward adverse risks. Loss ratios deteriorate. Actuarial responds by tightening guidelines or raising rates. That creates more friction. More friction accelerates the silent defection from agents who were still on the fence. The carrier looks at the data and concludes the contractor class is getting worse. The more accurate diagnosis is that the carrier’s own behavior drove away the profitable risks while retaining the difficult ones.
This is a self-inflicted adverse selection spiral, and it looks exactly like a market problem until someone examines the submission data carefully enough to see what’s actually happening.
The Unspoken Contract
Experienced brokers operate with an unspoken contract in mind when they work with a carrier. The terms are simple: be consistent, and I will bring you business. Be unpredictable, and I will protect my clients and my time by going somewhere else.
A “no” is workable. A broker can explain a declination to a client. They can remarket the account. They can manage expectations when the carrier’s position is clear and stable. What a broker cannot work with is a “maybe” that resolves differently each time depending on factors they cannot see or influence.
That dynamic is confirmed by the data. The 2025 Ivans Insurance Agency-Carrier Connectivity Trends Report found that for the first time, real-time appetite information has become the single most important factor agents consider when choosing carrier partners – cited by 29% of respondents, more than doubling from 12% the prior year. Agents are not asking for lower rates or higher commissions at the top of their list. They are asking to know where the carrier actually stands.
Carriers that answer that question clearly and consistently tend to earn submission flow. Carriers that don’t lose it – usually without knowing why.
What AI Makes Possible
The first two articles in this series argued that AI exposes internal governance gaps and surfaces patterns that organizations have long preferred to leave invisible. This article is, in a sense, the consequence of failing to act on what AI reveals.
But AI also offers something that has never existed before: the ability for a carrier to measure its own consistency from the perspective of a distribution partner.
Submission pattern analysis can detect when an agency’s volume toward a specific carrier or class begins declining, often months before any human relationship manager notices. Quote consistency monitoring can flag when the same agent receives materially different pricing on similar risks, creating an opportunity to intervene before the agent gives up. Appetite transparency tools – real-time, queryable, reliable – can close the gap between what the carrier’s guide says and what agents actually experience.
None of this requires replacing underwriter judgment. The goal isn’t uniformity for its own sake. Legitimate underwriting expertise adds real value, and the best underwriters bring contextual insight that no system fully replicates. The goal is to make inconsistency visible and intentional – documenting when the deviation from standard is a considered decision, rather than leaving agents to guess whether this underwriter is having a good week.
Used well, AI gives carriers the observability to understand their own distribution behavior for the first time. Used poorly, or not at all, it changes nothing, and the silent defection continues.
What Carrier Executives Should Do Now
Measure submission velocity, not just hit ratio. A declining submission trend from an agency or within a class of business is a warning signal. Most carriers don’t track it. They should.
Audit quote consistency from the agent’s perspective. Pull the last 20 quotes sent to your top-producing agencies for the same risk class. If variance exceeds 20% with no corresponding difference in risk characteristics, you have a relationship problem, not a pricing problem.
Close the appetite transparency gap. Only 19% of carriers offer real-time appetite indicators, yet agents now rank real-time appetite information as their top factor in carrier selection. That gap is a competitive opening for carriers willing to fill it.
Examine your IRPM patterns. If credits are being granted inconsistently across underwriters for comparable risks, you’re not fine-tuning pricing — you’re introducing noise into the system and teaching agents that outcomes depend on who they know.
Add distribution consistency to your AI governance framework. If you are building AI governance dashboards to satisfy regulators, add the producer-facing view. How predictable are you to your distribution channel? If you don’t measure it, you won’t know until the submission flow has already moved on.
The Bottom Line
Two related articles on this site have argued that AI doesn’t create governance problems — it exposes ones that already existed, and that organizations often prefer not to see what AI reveals. This article makes the commercial case for paying attention.
Inconsistency used to be a governance problem that lived inside the carrier. AI made it visible, which made it a management problem. But the market was already responding long before AI arrived. Brokers have been running their own consistency audits for decades. They just called it “knowing who to send what to.”
The difference now is that data tools on both sides of the relationship are getting better. Carriers that use AI to understand and improve their own consistency will earn the submission flow that reflects it. Carriers that don’t will continue to wonder why their best distribution relationships seem to be slowly, quietly cooling — without a single angry phone call to explain why.
The penalty for inconsistency isn’t a fine or a failed audit. It’s the business you never saw, from the broker who stopped giving you shots, who never told you they left.
Sources
- J.D. Power 2025 U.S. Independent Agent Satisfaction Study
- First Connect 2025 State of the Industry Report — via Insurance Journal and Carrier Management
- Ivans 2025 Insurance Agency-Carrier Connectivity Trends Report
- InsuranceIndustry.AI — The Governance Problem AI Didn’t Create (But Might Actually Fix)
- InsuranceIndustry.AI — The Reluctant Auditor: What AI Sees That We’d Rather It Didn’t
AI Disclaimer: This blog post was created with assistance from artificial intelligence technology. While the content is based on factual information from the source material, readers should verify all details, pricing, and features directly with the respective AI tool providers before making business decisions. AI-generated content may not reflect the most current information, and individual results may vary. Always conduct your own research and due diligence before relying on information contained on this site.
