The industry combined ratio looks like a scorecard, but it is not one for your underwriting team. U.S. property and casualty insurers posted a 92.9 combined ratio in 2025, according to AM Best. That number says nothing about which underwriters, segments, or brokers produced it.
Portfolio results arrive late and average everything together. A decision made today shows up in loss experience months or years from now. Meanwhile, quote turnaround, appetite fit, and decision consistency move every week, and each one is measurable today.
This article defines the underwriting KPIs that matter, groups them into a four-layer scorecard, shows how to calculate each one, and walks through building a scorecard your team will use.
Why Lagging Results Make Poor Underwriting Scorecards
Most underwriting leaders already track loss ratio and combined ratio. Both matter. Neither tells an underwriting manager what to change on Monday morning.
A useful scorecard pairs a small set of lagging results with leading indicators. The leading indicators show what is happening now. The lagging results confirm whether it worked.
What the 2025 Combined Ratio Hides
AM Best reported that catastrophe losses added 7.6 points to the 2025 industry combined ratio. Excluding $18.1 billion of favorable reserve development, the accident-year combined ratio was 94.9, per the same AM Best First Look coverage.
The two-point gap between 92.9 and 94.9 comes from reserve releases on earlier years of business. It does not measure how well 2025 risks were selected and priced.
Swiss Re Institute forecasts the industry ratio rising to 97 percent in 2026 and 99 percent in 2027, as competition grows and rates ease. Margin cushions shrink when rates soften, so individual underwriting decisions carry more weight.
The Delay Between a Decision and Its Result
An underwriter binds a policy today. Its premium earns over the policy term. Its claims report and develop over the months and years that follow.
Loss ratio therefore reports on decisions made long ago. A team that waits for it to move has already written the next book of business.
What a Useful Scorecard Does
A useful scorecard answers four questions. Is work moving? Are we quoting what we want to write? Do our decisions earn money? Do similar risks get similar answers?
Each question maps to a layer of KPIs. The next section lays them out.
The Four-Layer Underwriting Scorecard
The Four-Layer Underwriting Scorecard organizes underwriting KPIs by the question each one answers. Flow measures speed and cleanliness. Fit measures conversion and appetite. Quality measures results. Control measures consistency.
Pick two or three KPIs per layer. Eight to ten KPIs in total is enough for any underwriting team.
Layer | Question it answers | Core KPIs | Suggested cadence |
|---|---|---|---|
Flow | Is work moving, and where does it wait? | Cycle time by stage, queue time, rework rate, data completeness | Weekly |
Fit | Are we quoting the business we want to write? | Quote-to-bind ratio, out-of-appetite share, referral rate | Monthly |
Quality | Do our decisions earn money? | Loss ratio by segment, accident-year combined ratio | Quarterly, with monthly trend |
Control | Do similar risks get similar decisions? | Override rate, matched-pair variance, referral overturn rate | Monthly |
How the Layers Work Together
Flow and Fit are leading indicators. They move within weeks of a process change. Quality is the lagging confirmation. Control sits across all three, because inconsistent decisions distort every other number.
Read the layers together. A faster Flow result that comes with a falling Quality result means speed was bought with underwriting discipline.
Who Owns Each Layer
Assign every layer one accountable owner. Operations leaders own Flow. Underwriting managers own Fit and Control. The chief underwriting officer owns Quality.
A KPI without an owner becomes a number nobody acts on.
Flow KPIs: Measure Where Time Goes
Flow KPIs tell you where submissions wait and why. They are the fastest layer to improve, because most delay sits outside the underwriter's analysis.
Cycle Time by Stage
Cycle time is the elapsed time from submission received to decision issued. Break it into stages: acknowledgment, triage, underwriting, and quote issued.
Report the median and the 90th percentile for each stage. An average hides the submissions that wait longest, and those are the ones brokers remember.
Queue Time Versus Touch Time
Queue time is the hours a file waits for someone. Touch time is the hours someone works it. Measure them separately.
High queue time points to routing and staffing. High touch time points to complexity and data quality.
Rework Rate and Data Completeness
Rework rate is the share of submissions returned for missing or conflicting information. Data completeness is the share of submissions that arrive with every required field populated.
Both explain cycle time before an underwriter opens the file. A team that fixes data completeness shortens every stage that follows.
Fit KPIs: Are You Quoting What You Want to Write?
Fit KPIs connect underwriting activity to appetite and distribution. They show whether the submissions you work are the submissions you want.
Quote-to-Bind Ratio
Quote-to-bind ratio is bound policies divided by quotes issued, multiplied by 100. If a team issues 400 quotes and binds 100, the ratio is 25 percent. That figure is illustrative, not a benchmark.
Read it by product, producer, and region. A low ratio points to pricing, speed, or appetite problems. A very high ratio points to underpricing or an appetite set too wide.
Fix the measurement window before you report the number. For example, count binds within 60 days of the first quote, and exclude quotes still open. Otherwise recent quotes make the latest month look weak.
Out-of-Appetite Share and Referral Rate
Out-of-appetite share is the portion of incoming submissions declined because they fall outside appetite. Referral rate is referred submissions divided by total submissions reviewed.
A rising out-of-appetite share tells distribution partners what you stopped wanting. A high referral rate tells you the guidelines are unclear or the authority limits are set too low.
Pair Speed With Conversion
Plot time to quote against quote-to-bind ratio for each segment. Segments where slow quotes meet low conversion are the first place to fix intake and triage.
Segments where quotes are fast and conversion is still low point to pricing or appetite instead.
Quality and Control KPIs: Do Decisions Earn Money, and Do They Repeat?
Quality KPIs show whether decisions earned a return. Control KPIs show whether those decisions were consistent enough to trust.
Loss Ratio by Segment on an Accident-Year Basis
Loss ratio is incurred losses divided by earned premium. Calculate it by class, region, channel, and program instead of for the whole book.
Use an accident-year basis. It strips out reserve movement on earlier years, the same effect that separates 92.9 from 94.9 in the AM Best figures above.
Combined ratio adds the expense ratio to the loss ratio. A combined ratio below 100 percent means the book earns an underwriting profit before investment income.
Pair Speed Metrics With Quality Metrics
Time saved is a Flow result. Decision quality shows up in segment loss ratio and in Control KPIs. Measure both.
The gap is visible in the market. In the Underwriting Edge 2026 survey of 350 senior commercial P&C underwriters, 51 percent of those using AI said its greatest contribution was saving time on manual admin, and 21 percent said it improved decision quality. The survey was published by an underwriting software vendor, so treat it as one data point.
Consistency KPIs
Three Control KPIs expose variance. Override rate is the share of tool or guideline suggestions an underwriter changes. Referral overturn rate is the share of referrals decided differently from the referring underwriter's recommendation.
Matched-pair variance is the share of similar risks, quoted by different underwriters, that ended with unexplained gaps in terms. Our guide to why underwriters make inconsistent decisions walks through the matched-pair review step by step.
Building a Scorecard Your Team Will Use
A scorecard that sits in a slide deck changes nothing. Five steps turn the four layers into a working review.
Choose eight to ten KPIs, two or three per layer.
Write one definition per KPI: numerator, denominator, time window, and exclusions.
Segment every KPI by product, producer, and region before you average it.
Baseline each KPI across at least one full seasonal cycle.
Name an owner and a review cadence for each KPI.
Fix Definitions Before Dashboards
Two teams that calculate cycle time differently produce numbers that cannot be compared. Write the definition first. Build the dashboard second.
Record the definition next to the number, so every reader knows what the number counts.
Common Mistakes
The first mistake is tracking too many KPIs. A dashboard with thirty metrics gets ignored, and a short list gets reviewed.
The second is reporting averages. A healthy company-wide cycle time hides one slow line of business, and segmenting exposes it.
The third is measuring speed without outcome. Every Flow improvement needs a matching Quality and Control check.
The Data Foundation Behind Every KPI
KPIs are only as consistent as the data beneath them. Cycle time needs reliable timestamps. Segment loss ratio needs policy, claims, and billing data that agree with each other.
When those records sit in separate systems, KPIs disagree and trust drops. Our overview of data silos in P&C insurance explains how that happens and what to do about it.
How InsOps Helps
InsOps builds an insurance-trained AI that assists underwriting teams with the data behind their KPIs. 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 reaches Guidewire in consistent structures without custom engineering for each source. Consistent structures give every KPI a single definition of the same record.
InsOps migrates legacy data into Guidewire and keeps it flowing in real time.
If you are evaluating how to build underwriting KPIs on data your team trusts without a long engineering project, contact us to talk through what this could look like for your operation.
Frequently Asked Questions
What are underwriting KPIs?
Underwriting KPIs are measurable indicators of how well an underwriting team selects, prices, and processes risk. They cover speed, conversion, profitability, and consistency.
Why do underwriting KPIs matter?
Loss results arrive long after the decision that caused them. Leading KPIs such as cycle time and quote-to-bind ratio show what is happening now, while loss ratio confirms the outcome later.
Which underwriting KPIs should a team track first?
Start with cycle time by stage, quote-to-bind ratio, loss ratio by segment, and override rate. Together they cover one KPI from each layer of the scorecard. Add more only after each one has an owner and a written definition.
What is a good quote-to-bind ratio?
No single benchmark applies to every book. Ratios differ by line, segment, and channel, so compare each segment against its own history over a full seasonal cycle.
What is the most common mistake in measuring underwriting performance?
Reporting company-wide averages. An average hides a slow line of business or a weak segment, so segment every KPI by product, producer, and region before you review it.
How do you measure whether underwriting decisions are consistent?
Track override rate, referral overturn rate, and the share of matched pairs with unexplained gaps. Falling gaps and stable, explained overrides signal progress.
How does AI assist with underwriting KPIs?
InsOps builds an insurance-trained AI that assists with the data behind KPIs by mapping policy and underwriting data from source systems into consistent Guidewire structures. A person reviews and validates every mapping before it is deployed, and the underwriter keeps every underwriting decision.
How long does it take to put an underwriting scorecard in place?
Defining the KPIs and writing their definitions takes weeks. The baseline takes at least one full seasonal cycle. Plan to refine the scorecard over several review cycles, and do not build dashboards before the definitions are agreed.