The Margin Squeeze: Why Data Discipline Is Your New Competitive Edge

UK commercial insurance rates have now fallen for nine consecutive quarters. Marsh's Global Insurance Market Index put the Q1 2026 decline at 8% across the UK market, with property down 10% and financial and professional lines down 8%. For buyers, that's welcome news. For the insurers, MGAs and Lloyd's managing agents writing the business, it's the ninth quarter in a row they are trying to work out: where does profitable growth come from when rate isn't providing it?

The answer hasn't changed: risk selection, claims discipline and cost-to-serve. What's changed is how visible the difference has become, and how quickly it shows up. In a market this soft, the gap between a well-managed book and an averagely managed one now surfaces in the numbers within a year, not five.

That visibility is precisely what makes data discipline the edge worth building now, not later.

The squeeze is structural, not temporary

It's tempting to treat a soft market as a phase to wait out. The data doesn't support that comfort. Every broker surveyed by Ascend Insurance Holdings in a recent industry poll expects the market to plateau by Q4 2026, but "plateau" is not "harden," and most of those same brokers are still forecasting double-digit revenue growth over the next 12 months. Flat-to-soft pricing alongside real growth expectations only works if firms find margin somewhere other than rate.

International cyber insurance rates have fallen 43% since the fourth quarter of 2023, and combined ratios are now deteriorating across the US, Europe, the UK and Australia and New Zealand, with several markets potentially turning unprofitable by 2027 if current trends continue, according to specialist underwriter DUAL. That's not a rate problem. It's a data and claims-management problem, and it's already showing up in underwriting results.

What do we mean by Data Discipline?

It doesn’t mean more dashboards. We have identified three specific areas of data discipline that the best performing firms are utilising to deliver competitive advantage:

Risk selection you can trust. When capacity is abundant and everyone is competing for the same accounts, the firms winning in this soft market are the ones that can tell, immediately, which risks are genuinely well-managed, and which just look that way on the submission. That takes clean, structured, current data feeding underwriting judgement, not a portfolio review that surfaces the problem once a quarter, after the damage is done.

Claims data you can see through, not just around. Rising claims costs are hard enough to manage even when the underlying data is reliable. They become far harder when legacy systems and fragmented data slow down how quickly a claims team can spot a deteriorating loss ratio or a delegated authority arrangement drifting off-track. The FCA's continued focus on claims handling and oversight of outsourced arrangements makes this doubly urgent: good data discipline and good regulatory standing are rapidly becoming the same thing.

Portfolio visibility in real time. The firms furthest ahead aren't necessarily the ones with the most advanced AI. They're the ones whose underlying data foundation is solid enough that AI and automation can actually be trusted to act on it. That distinction matters more than it sounds. It's the difference between a faster underwriting cycle and a faster way of making the same undetected mistakes.

The infrastructure conversation has changed too

There's a second, quieter opportunity for data discipline to drive competitive advantage, and it lies in the cancellation of Blueprint Two. Whatever your view of that decision, it removes an excuse many firms have quietly leaned on: waiting for market-wide infrastructure to solve the data problem for them.

That excuse no longer holds. Data standards work continues; the Lloyd's Market Association, for instance, is still pushing for ACORD-aligned, Core Data Record-compliant data, but responsibility for getting there now sits with individual firms, not a central programme.

That shift arrives alongside a genuinely fast AI adoption curve. The LMA's April 2026 survey found AI adoption across the Lloyd's market has more than doubled in twelve months, with most firms now building formal AI governance frameworks. But the same survey found only a small minority of managing agents have deployed agentic or generative AI into live underwriting decisions. The rest aren't behind because the technology isn't ready. They're behind because their underlying data isn't ready for it yet.

The edge is in the foundations, not the feature

None of this is an argument against AI, automation or innovation; quite the opposite. It's an argument for sequencing. In a market where rate isn't delivering margin, the firms that win are the ones treating data quality, claims discipline and portfolio visibility as the underwriting edge itself, not as the unglamorous prerequisite to one.

The soft market didn't create this problem. It just removed the one thing that used to paper over it: rising rates.

Over the next few months, we'll explore what that looks like in practice: how strong data foundations support more value-creating uses of AI, how firms move beyond defending margin into launching new products and revenue lines, and why M&A and consolidation are turning data debt into the next real deal risk.

If any of this reflects a conversation you're already having internally, we'd like to be part of it. We're bringing together a small group of senior data, claims and underwriting leaders next month for a candid, Chatham House Rule discussion on exactly this theme. if you'd like to join us and other industry peers.