What if your pricing model doesn't know
what it doesn't know?
Every insurer prices risk with the data available at point of quote. But what if the data you're pricing on is incomplete, unvalidated, or, in a significant number of cases, deliberately wrong?
The fields that drive your premium are self-declared. The enrichment that validates them sees one quote at a time. And the gaps between what you know and what's actually happening are flowing straight into your loss ratio.
This page is about the blind spots. The ones you can't see from inside your own data.
Take 100 people quoting for motor insurance this month
This is what the UK quote market actually looks like. Not in theory. In the data.
Quote once and leave
They key one quote and either buy or walk away. No comparison, no cross-check, just a single declaration flowing into your pricing model.
You can't validate these against anything. There's nothing to compare. Whatever they declared is what you price on.
For some, that's fine, they're genuine one-and-done buyers. But you don't know which ones are fine and which ones aren't, because you only see the one quote.
Quote again within 30 days
This is where it gets interesting. Some are legitimately exploring. What if I put it in the garage? How much more for comprehensive? Those are fair questions.
But in that group, a significant number change things that can't legitimately change in 30 days. NCD years. Occupation. Claims history. These are the fields that materially shift your premium, and they're being changed between quotes to find the cheapest combination.
Your enrichment validates each quote independently. Each one looks fine. But across the market, the pattern tells a different story.
Multiple quotes, clear pattern
Multiple quotes. Significant changes. Systematic behaviour. That's not someone testing the excess. That's someone who understands which levers move the price and is pulling all of them at once.
Some are changing identity fields, driving licence, date of birth, email address, to create multiple identities and circumvent previous history.
These aren't edge cases. They're a persistent, measurable segment of the market. And unless you're looking across the whole market, you can't see them, because each quote, in isolation, looks perfectly normal.
The question isn't whether this is happening in your book. It's how much. And unless you're looking, you can't see it.
Five things your current data can't tell you
What happened between quotes
A customer who quotes with five insurers in the same week, adjusting their details each time, is invisible to any single insurer. Each sees one quote. The behaviour pattern, the manipulation, the identity variation, the field changes, only becomes visible when you have the full market view.
No individual insurer can build this. It requires a dataset that spans the entire UK quote market.
Which fields weren't validated
The fields that shift most between quotes are the ones nobody validates at point of quote: voluntary excess and occupation.
There's no CUE check for occupation. No validation on excess. No database that confirms annual mileage. These fields are self-declared. They flow straight into your pricing model and straight into your book.
Some of that is legitimate. Someone tests a higher excess to see what happens to the price. That's smart shopping. But when the same person changes their excess, their occupation, and their mileage across multiple quotes in the same sitting, that's a different pattern entirely.
What your enrichment actually matched
Enrichment is only useful when it matches. Every data source has a match rate, and it's rarely 100%. The quotes that don't match are priced without the data your model was supposed to have.
Worse, match rates vary by customer type, geography, and line of business. Your pricing model is learning from an unevenly enriched sample, and most actuaries don't know the extent of the gaps.
How much of your book was actually enriched the way you think it was? If you can't answer that precisely, you have a blind spot most insurers live with without measuring it.
The mispriced middle
Most insurers can identify the obvious bad risks and the obvious good ones. The expensive problem is the middle, customers who look standard but carry hidden risk from undisclosed claims, changed details, or connected identities.
These are being priced inaccurately in both directions: some underpriced (you're carrying risk you're not being paid for), some overpriced (you're losing good customers to competitors who price them more accurately).
The scale of this mispricing is rarely visible until an external dataset reveals it.
The household you can't see
A household with motor, home, and van policies across two insurers and three named drivers is treated as a collection of unrelated individuals. Each policy is priced in isolation.
The full picture, shared risk, shared behaviour, shared financial profile, is invisible. Pricing misses the patterns that only become visible when the household is viewed as a unit.
Cross-line visibility turns isolated policies into connected intelligence.
The intelligence that changes the equation
Each blind spot has a specific answer. Not a generic data product, a specific capability that closes a specific gap in your pricing model.
Quote-Stream Intelligence
See every quote across the UK market, not one at a time, but all of them. 25 million unique quotes per month across motor, home, van, bike, and travel.
Match every person who quotes with you against their full quote history. See what they declared elsewhere, what they changed, and whether the pattern is consistent or suspicious.
Closes Blind Spots 1 and 2
Powered by: Quote Intelligence
Learn more about Quote IntelligenceEnrichment Orchestration
Replace fragmented, single-source enrichment with a platform that aggregates data from multiple providers, monitors match rates, and fills gaps that individual sources miss.
When enrichment sources work together rather than in isolation, the combined match rate exceeds any individual source. Bureau discoveries feed into quote intelligence, which produces incremental discoveries for follow-up searches.
Closes Blind Spot 3
Powered by: Data Intelligence + Bureau Intelligence
Learn more about Data IntelligenceThe Mispriced Middle, Revealed
Cross-reference quote behaviour against bureau data, claims history, and market context. Separate the genuinely standard risks from the ones carrying hidden signals.
Green identification through consistent keys and clean history. Amber identification through marginal signals that need context. Red identification through systematic patterns that are invisible in isolation.
Closes Blind Spot 4
Powered by: Evaluate + Quote Intelligence
Learn more about EvaluateCross-Line, Cross-Market Visibility
See the full person, across every insurance line, every aggregator, every quote they've submitted. Motor customers quoting for home. Home customers with vehicle history. Connected households with shared risk profiles.
Five lines of business. Five years of history. 400+ behavioural attributes per person.
Closes Blind Spot 5
Powered by: Symphony
Learn more about SymphonyThe dataset behind the intelligence
per month
history
attributes per person
business
The Percayso Quote Lake captures every quote event across all major UK aggregators. This isn't sampled data. It's the full stream, from June 2021 to today, continuously growing.
Traditional data providers see policies and claims, what happened after you made a decision. They can't see what was declared at quote time, because they don't see the quotes.
We do.
Every organisation that prices risk faces these blind spots
Insurers
Your pricing model is only as good as the data it receives. If 40% of your enrichment doesn't match, if the fields that drive your premium aren't validated, and if you can't see what happens between quotes, your book is carrying risk you don't know about.
The Head of Pricing problem: the model doesn't know what it doesn't know. The book skews toward risks other insurers are declining, without anyone realising it until the loss ratio moves.
Solutions for InsurersBrokers
Your client book reflects the risks you've placed, but do you know which clients are being accurately priced and which are carrying hidden signals? When a client's renewal comes in higher than expected, is it because the risk genuinely changed, or because someone else saw something in the data that you didn't?
Visibility into quote behaviour across the market gives you the context to advocate for your clients, or to know when not to.
Solutions for BrokersMGAs
Scheme performance depends on the quality of the risks you attract. If your aggregator pricing is competitive for the wrong reasons, winning business that other schemes are deliberately avoiding, the loss ratio tells you too late.
An independent view of who's quoting into your scheme, what they declared elsewhere, and whether your pricing is attracting greens or accumulating reds.
Solutions for MGAsThe blind spots are measurable.
Evaluate measures them.
We take your data, map it against 25 million unique quotes per month, and show you exactly where your pricing model is flying blind. Red, amber, green, your book, segmented by the signals your current data can't see.
8 to 12 weeks. No long-term commitment. Every insurer who's done one has been surprised by what they found.
