Wednesday, 11:00 a.m. A cyber renewal call. The client opens with good news — the premium is flat, capacity is everywhere, they’re not paying more this year.
Then they ask for the sublimit schedule.
There isn’t one in the file. There is a policy, a renewal declaration page, and a PDF of last year’s declarations. The previous broker’s schedule is in the file from 2024, and it does not match.
Nobody checked. The premium went up 4% last year, which is a conversation you have about the premium, and nobody opened the schedule to see what else changed while they were looking at the premium.
This is the least dramatic version of what a flat market did to cyber coverage.
What WTW Reported on Cyber, and What Marsh Called It
The fall 2026 edition of WTW’s Insurance Marketplace Realities report contains one number that requires reading twice: cyber rates are holding between a 5% decrease and a 5% increase. Effectively flat, with abundant capacity behind it.
Marsh’s Q2 2026 index measures the same line from the other side: global cyber rates declined 4%, the twelfth consecutive quarter of reductions, after a 5% decrease in Q1. US cyber came in at -2%, level with Q1, and Marsh notes that “US cyber rates have been declining since the second quarter of 2023.”
The same index carries the sentence that should worry you more than the rate: “while still available, capacity was increasingly selective, with a strong focus on risk quality and program structure.”
That is the coverage squeeze, stated by the market’s own index. Capacity did not leave. It got choosy.
Insurance Business reported the same trend on October 13, framing it as “lower premiums and abundant capacity may be obscuring exclusions and sublimits for businesses dependent on external technology.” Amwins’ Megan North describes the underwriting side: carriers are “putting guardrails on what they’re willing to cover” — specifically widespread-event exclusions and restrictions on dependent business interruption coverage.
Meanwhile, at the 1 January 2027 renewals, Marsh Re’s global co-head of Cyber reports buyers exploring higher cessions, with scrutiny on attachments, limits, reinstatements, and aggregation, and event definitions likely to receive closer attention given AI-accelerated clustering risk.
The price did not move. Everything around the price did.
Four Squeezes Hiding Behind a Flat Number
Stripped down, a cyber renewal is one of four things, and they are nowhere near equal in risk. The tiers are our suggested placements for your own scoring — no underwriter grades your book for you, so treat them as a starting position, not an official classification.
— The third-party gap. The client runs a distribution network, a multi-location retail operation, or a firm handling client funds. Their largest exposure is not their own servers — it is the payroll provider, the managed service provider, and the software vendor with authentication into their environment. Verizon’s 2026 Data Breach Investigations Report, as reported by Insurance Business, found third parties were involved in 48% of analyzed breaches, a 60% increase on the prior year. That is the population carriers are now writing exclusions around. Our suggested tier: High.
— The sublimit inside the limit. A $5 million limit with a $250,000 business interruption sublimit is a $250,000 policy in practice. In a capacity-rich market, sublimits are negotiable, and a client who never asked for them increased is a client who never knew they had them. Our suggested tier: High.
— The aggregation question. AI-accelerated clustering means one event can generate claims across multiple insureds, locations, or subsidiaries. Marsh Re describes event definitions and aggregation as a focus area at 1/1/2027. For a distributed client, how the policy groups claims decides whether one event erodes one limit or several. Our suggested tier: Moderate.
— The definitional drift. Autonomous AI agents are forcing carriers to revisit what qualifies as a cyberattack at all. A client who has deployed AI agents with system access may have a policy that does not respond the way they assume. A new question, and almost no client has asked it. Our suggested tier: Moderate.
The line worth memorizing: a flat cyber rate tells the client what they are paying. It does not tell them what they are covered for.
What the Market Data Actually Says Now
Last verified against primary sources October 23, 2026. Cyber is the line where the market is actively rewriting definitions, and the January 1 renewals are the next checkpoint. Re-verify before you rely on any of it.
— What changed: price, marginally, and for longer than most people expected. Twelve consecutive quarters of global rate decline is not a correction, it is a structural shift with a ceiling — as one underwriter put it to Insurance Business, “the easy money is gone.” Cyber premium growth had already slowed from 30%-plus annually between 2017 and 2022 to roughly 5% between 2022 and 2025.
— What did not change: loss activity, and the scrutiny attached to it. Cyber reinsurance demand remains steady despite lower pricing and rising attack activity. The first property/cyber combined cat protection, three new risk XL covers, and the first cyber hard retro at 1/1/2026 signal a market whose toolkit has broadened rather than settled. Hard retro opening to new entrants means capacity exists — but it also means structures are being tested rather than defaulted to.
— What a broker can fix: how the client presents its risk — the sublimit schedule they buy, the third-party wording they accept, the aggregation structure they want, the incident response relationships they have, and the evidence in the file. A broker cannot change what a cyber carrier considers a cyberattack. The evidence question is the one that bites. Multifactor authentication, password management, and employee training are common underwriting considerations, and documentation of them is what earns better terms. Insurers are increasingly examining compensating controls, including network segregation and restrictions on third-party access. Collecting that documentation takes weeks; assembling it during the renewal window does not happen.
— What to watch. The definition of the covered event, since AI agents force carriers to revisit what qualifies as a cyberattack. The panel firms, because a policy response often depends on the client satisfying notification conditions, and a missed condition is a foreseeable denial. And the margin discipline — cyber is the line where appetite is most likely to be withdrawn quietly, so watch submission guidelines, not just published materials. The wider cycle context is in our July piece on the cyber insurance market.
The Risk Nobody Has Absorbed
The risk most brokers have not absorbed is the claim scenario. Not the rate, not the sublimit schedule in isolation — the walk-through of an actual claim against the actual policy to find the point where the payment stops.
Most clients have never done it. They know their limits because they bought them three years ago in a hard market. They do not know that a $5 million limit with a $250,000 business interruption sublimit, a supplier exclusion, and a single-entity aggregation provision resolves to a materially different outcome than the one they picture when they say “we’re covered for $5 million.”
We covered what a soft market does to agency economics in the October commission math and how the Q4 workload stacks in the renewal capacity plan. Neither is this risk. The November broker Q&A has the short version of this and the other two.
The Numbers Your Client Will Google
Before the controls, a word about the arithmetic, because these numbers will reach a board deck whether you are ready or not.
The headline number in cyber right now is relief: flat pricing after years of climbing rates, abundant capacity, and reinsurers entering January flush. For pricing, that is largely true, and we wrote about what the soft market means for renewals.
The second number is certainty, and that is where the risk is. Some vendors still sell “comprehensive cyber coverage” without mapping it to the specific sublimits, exclusions, and aggregation terms in the client’s actual policy. Others quote a market price as though price were the only variable in the outcome. In cyber, price is the least informative number on the page.
Which question is being answered? A premium comparison is a comparison of premiums. Your client’s exposure is a stack of sublimits, exclusions, aggregation terms, and third-party conditions. What is the denominator? A policy limit is not a recovery. What is the measurement window? Ask when the comparison was last run against current wording, because where event definitions are actively rewritten, a comparison built on last year’s language is a liability. Keep the ROI column separate from the evidence column. A broker with a credible savings claim and a defensible coverage comparison walks into a renewal with an answer.
Five Checks to Put Around Your Cyber Program
None of this requires a data scientist. All of it is something a client can ask you for on the record.
— Run one claim scenario end to end. Take a live event from the client’s operations, walk it through the policy, and write down where the payment stops. It is the single highest-value hour in the renewal.
— Pull the sublimit schedule and read every line. If you do not have this year’s schedule, request it. Then test the sublimits against the loss the client actually fears, not the one you assume they fear.
— Reconstruct the payment chain. List every external technology dependency the client relies on, and trace what happens to coverage when that provider fails or is compromised. Some of this is now excluded or sublimited — and carriers are charging additional premium to restore dependent business interruption coverage to full limits.
— Confirm the notification conditions. Panel firms, forensics requirements, timeframes. A missed condition is a denial, and one of the cheapest to prevent.
— Get the control evidence in the file before the submission window. The evidence is what earns the terms. Request it in October.
Where to Start on Monday
You do not need a program. You need a week. Pull your ten largest cyber accounts. For each, get the current sublimit schedule, the current aggregation terms, and the third-party wording. For the three largest, run one claim scenario each. Any gap you find now is a roadmap.
The same gap, found by a client after a claim, is a coverage dispute. And if you would rather someone else pull the sample, our dedicated insurance operations and back-office support runs that review and hands you the comparison.
A flat rate is a chance to fix the cyber program at a known cost. The brokers who use it that way are the ones whose clients discover the sublimits before they need them.
Insurance market commentary for discussion, not coverage advice. Cyber terms, sublimits, and event definitions are changing faster than in any prior cycle, and every carrier’s wording is its own. Confirm current terms with the carrier before binding, and confirm the definition of the covered event for any client deploying AI agents with system access.
24×7 Synergy supports insurance agencies that have to answer for a cyber program in a file, not a slide deck — coverage schedule comparisons, application and questionnaire handling, evidence packet assembly, and COI processing. Book a 30-minute call and we will send our one-page cyber program review checklist for your January book.

