When leadership asks how your underwriting operation compares to peers, you need numbers from two places. The public figures come from surveys, and the internal figures sit in separate policy, claims, and billing systems.
Survey data gives you a reference point. Your own timestamps and file samples give you the answer that matters.
This article gives you a four-measure scorecard, the public figures to hold it against, and the order to work through any gaps.
Where Do Peers Stand on Analytics and AI?
Insurers with more sophisticated analytics posted combined ratios six percentage points lower than slower adopters between 2022 and 2024. That finding comes from WTW's 2026 survey of 59 property and casualty insurers in the United States and Canada.
The survey measures a pattern across carriers. It does not prove that analytics alone caused the gap, so treat it as a direction, not a target.
Analytics Leaders Run Lower Combined Ratios
Combined ratio is a public yardstick for underwriting results. Pull yours by line and class, then place it next to the WTW gap before drawing any conclusion about your own desks.
AI Assistance Is Still a Minority Practice
Only 16% of the surveyed insurers currently use AI to augment human underwriting. Looking ahead, 60% plan to prioritize it between now and 2028.
If your team already uses AI-assisted tools, you sit with the early group. If not, you sit with the other 84%.
Where Does Underwriting Time Actually Go?
Among underwriters who use AI, 51% say its greatest contribution has been saving time on manual administration. Only 21% say it has led to better decisions, according to the Underwriting Edge 2026 survey.
Admin Time Is the Biggest Gain So Far
Coleman Parkes fielded the survey in June 2026 with 350 senior commercial P&C underwriters across the US and UK. Underwriting software vendor hyperexponential published it, so read the figures as one data point, not an industry standard.
Manual Entry Tops the Blocker List
Manual data entry across systems ranked first among workflow blockers, named by 42% of those underwriters. That makes re-keying the first place to look when you size your own admin load.
What Should You Measure to Benchmark Your Own Operation?
Four measures make up what we call the Underwriting Benchmark Scorecard. Each pairs a number from your own systems with the closest public reference point.
Measure | Pull from your own systems | Public reference point |
|---|---|---|
Admin load | Share of underwriter hours spent on data entry and re-keying, from a time log on 20 recent files per line | The top workflow blocker in the Underwriting Edge survey |
Quote speed | Hours from submission receipt to quote, split into working time and waiting time | None used here. Compare desks and lines inside your own book |
Decision consistency | Share of matched risk pairs with unexplained differences in terms | None used here. Compare desks inside your own book |
Portfolio results | Combined ratio by line and class | The WTW gap between analytics leaders and slower adopters |
Pull the Numbers From Your Own Systems
Start with a sample, not a full study. Pick twenty recent files per line of business and log where each hour went.
Run a matched-pair review for decision consistency, the method described in our guide to inconsistent underwriting decisions.
Count how many screens an underwriter opens per file. Our overview of data silos in P&C insurance explains where those screens come from.
Read the Gaps in Order
Start with admin load, because it sits upstream of the other three measures. Re-keyed data slows quotes and gives two underwriters different facts.
Then move to quote speed, decision consistency, and portfolio results, in that order.
How Do You Act on the Scorecard?
Work through the scorecard one line of business at a time:
Score one line of business on all four measures.
Mark the measure with the widest gap between desks.
Trace that gap to the system or handoff where it starts.
Fix that handoff, then re-score a sample of the same size next quarter.
Start With One Line of Business
Pick the line with the most re-keyed submissions. A narrow scope shows whether a change moved a number before you widen it.
Keep a Person in the Review
Every score and every proposed fix needs an underwriter's sign-off. In the Underwriting Edge survey, underwriters described their ideal setup three years out the same way: AI takes on data, triage, and drafting, while the underwriter keeps complex risks and the final call.
How InsOps Helps
InsOps builds an insurance-trained AI that assists underwriting teams with the data behind each decision. LiLa, our insurance-trained LLM, runs inside your own environment, so PII and PHI never leave controlled infrastructure. A person reviews and validates every mapping before it is deployed.
Our Integration Gateway connects to Guidewire PolicyCenter and UnderwritingCenter, so policy and underwriting data from your source systems arrives in consistent Guidewire structures without custom engineering. That addresses one source of re-keying, the data side of your admin load measure.
InsOps migrates legacy data into Guidewire and keeps it flowing in real time.
If you are evaluating how to benchmark underwriting operations without adding manual reporting work, contact us to talk through what this looks like for your operation.
Frequently Asked Questions
What is underwriting benchmarking?
Underwriting benchmarking compares your operation's speed, consistency, and results against internal baselines and external reference points. External figures come from surveys and public filings. Internal figures come from your own system timestamps and file samples.
Why does underwriting benchmarking matter?
It shows which part of the operation to fix first. WTW's survey ties stronger analytics capabilities to lower combined ratios, though it measures association, not cause.
How do you benchmark underwriting performance against peers?
Score your operation on admin load, quote speed, decision consistency, and portfolio results. Then hold each score against the closest public figure. Where no public figure exists, compare desks and lines inside your own book.
What makes underwriting benchmarking hard?
Data sits in separate systems, so timestamps and file histories take effort to pull. In WTW's survey, data concerns and IT bottlenecks were the main challenges in adopting analytics.
Which metrics show whether you are making progress?
Track the share of underwriter hours spent on re-keying, hours from receipt to quote, the share of matched pairs with unexplained differences, and combined ratio by class. Re-score the same sample size each quarter.
How does AI assist with underwriting benchmarking?
InsOps assists by mapping the data underwriters work from into consistent Guidewire structures, so scorecard inputs arrive in one shape. A person reviews and validates every mapping, and the underwriter keeps the decision.
What should you expect from a first benchmark?
Plan for the first scorecard to cover one line of business and one quarter of data. The common pitfall is widening scope before that first line's numbers settle.