Your Catastrophe Model Is Only as Good as Your Exposure Data

Catastrophe models are extraordinary pieces of science. They simulate hurricanes, floods, earthquakes, and wildfires across millions of scenarios to tell an insurer how much it could lose in a bad year. Reinsurance treaties, capital requirements, and pricing all lean on their output.

And their output is only as good as one deeply unglamorous input: your exposure data — what you insure, where it is, and what it's made of. Get that wrong, and the most sophisticated cat model on earth confidently produces a precise, authoritative, wrong number.

Garbage in, catastrophe out

A cat model takes your portfolio of insured properties and asks, for each, "if this peril strikes here, how badly is this specific building damaged?" To answer, it needs to know, per property: exact location, construction type, year built, number of storeys, roof type, occupancy, and value.

Now consider what that data actually looks like at most insurers:

  • Location as a postcode, not a point. Flood and storm-surge risk can vary enormously within a single postcode — one side of a street floods, the other doesn't. Geocoding to postcode centroid instead of the actual building can misstate risk by an order of magnitude.
  • Construction type blank or defaulted. A frame house and a masonry house have very different hurricane vulnerability. When the field is empty, the model assumes an average — and the average is wrong for both.
  • Replacement values that are years stale. Underinsurance at the property level becomes a systematically understated portfolio loss.
  • Missing secondary characteristics. Roof age, roof-to-wall connection, whether there's a basement. These are exactly the modifiers that separate a survivable event from a total loss — and they're the fields most often absent.

Feed that into a cat model and you don't get a wrong answer you can see. You get a wrong answer that looks exactly as precise and credible as a right one.

Why this is getting more dangerous, not less

Two forces are converging. Climate change is making the historical record a worse guide to future losses, so the models lean harder on property-specific characteristics to compensate — which means exposure data quality matters more each year, not less. And regulators and rating agencies are scrutinising cat exposure more closely, so "our data is roughly right" is becoming an audit finding rather than an acceptable position.

The insurers who get hurt in the next bad season won't only be the ones with too much risk. They'll be the ones who didn't know how much risk they had, because their exposure data told them a comforting fiction.

What fixing exposure data actually involves

1. Geocode to the building, not the postcode. Precise coordinates for every insured location. This single change often shifts modelled losses materially, because peril gradients are steep.

2. Enrich the missing characteristics. The construction, roof, and age fields your systems never captured can be filled from property databases, aerial and satellite imagery, and permit records. This is exactly the kind of external-data join that's now practical and wasn't a decade ago.

3. Keep values current. Rebuild-cost indexing so replacement values don't drift into fiction.

4. Measure completeness before you model. Know what fraction of your portfolio has full, trusted characteristics — and treat the gap as a known uncertainty rather than a silent one. A model run on 40%-complete data should come with that caveat attached, loudly.

5. Make it a pipeline, not a project. Exposure data decays as the book changes. A one-off cleanup is stale by next quarter; the enrichment has to run continuously as policies bind.

The uncomfortable summary

Cat modelling gets discussed as a modelling problem — which vendor, which peril set, which climate assumptions. Those matter. But the largest source of error in most insurers' cat numbers isn't the model. It's the exposure data underneath it, and no model choice fixes bad inputs.

It's the same lesson as everywhere else in insurance AI and analytics: the sophisticated engine is rarely the constraint. The data feeding it is. In cat modelling the stakes just happen to be your solvency in a bad year — which is a good reason to fix the input before trusting the output.

We build the geocoding, enrichment, and exposure-data pipelines that catastrophe models depend on. More at IntelliBooks.

Comments

Post a Comment

Popular posts from this blog

Why Insurance Fraud Rings Are a Graph Problem (And Your SQL Database Can't See Them)

Your fraud team is good at catching individuals. A claimant whose story shifts, a receipt that looks altered, someone who files a bit too often. That's what the rules were built for, and they work. They're nearly blind to organised fraud — and organised fraud is where the money goes. The reason isn't that your team lacks skill or your rules lack sophistication. It's that you're asking a question your database structurally cannot answer. What a fraud ring actually looks like A ring isn't one bad claimant. It's a network: a handful of claimants, a couple of "witnesses", a repair shop, a medical clinic, a few bank accounts, some phone numbers, and a device or two — recombined across dozens of claims over months. Take any single claim from that ring and it looks fine. The claimant has no history. The damage is plausible. The repair estimate is in range. Nothing trips a rule, because the rule is examining one claim in isolation, and in isolati...
Read more

Claims Leakage: The Quiet Loss AI Can Close

Claims leakage is the money your company pays out that it didn't have to. Not fraud — leakage. The extra amount on legitimate claims that got overpaid because someone missed something, applied the wrong rule, or lacked the information to know better. Industry estimates put it at 3–10% of claims spend. For a mid-size insurer that's a rounding error you could retire. And almost nobody manages it directly, because it's nearly invisible — every individual leaked rupee looks like a normal payment. Why leakage is so hard to see Fraud has a villain. Leakage doesn't. It's the accumulation of small, defensible, individually-reasonable decisions: An adjuster settled at the reserve figure because they were busy, and the reserve was set high. A repair was authorised at shop rates when a network rate applied. A subrogation opportunity was never flagged, so recovery never happened. A policy limit or deductible was applied wrongly because the coverage detail was buried...
Read more

The Subrogation Money Insurers Leave on the Table

Subrogation is the least glamorous word in insurance and one of the most profitable. It's the process of recovering money from the party actually responsible for a loss, after you've paid your customer. Done well, it flows straight to the bottom line — recovered money is nearly pure margin. Most insurers leave a lot of it on the table. Not because they don't try, but because the opportunities are identified too late, by the wrong people, from incomplete information. It's a data problem, and it's an unusually clear one. Why subrogation is a timing game you're losing Recovery potential decays fast. The evidence that establishes another party's liability — the scene, the witnesses, the vehicle, the maintenance records, the CCTV — degrades or disappears within days. The other party's insurer gets less cooperative once they've closed their own file. So the single biggest lever in subrogation is identifying the opportunity early , ideally at first ...
Read more