The insurer's spine. Underwriting on a governed property record
Why property insurance is the second natural home for the Property Spine, after specialist lending.
Abstract
Property insurance underwriting is a data problem in the same shape as specialist lending. Non standard construction, flood and subsidence exposure, EPC band, occupancy and tenancy state and claims history all sit in different systems with different confidence. A governed property record with provenance and confidence is the operating unlock. This paper sets out what changes for insurers when the join, not the data, becomes the unit of value, and where insurance can lead rather than follow lending.
Audience
Home and buy to let insurers. MGAs. Reinsurers. Brokers. Loss adjusters. Property claims teams.
Section one. The insurance data problem is a join problem
- Property attributes live across enrichment providers, geological and hydrological services and internal claims histories.
- Non standard construction, flood zone and subsidence exposure are the classic underwriting variables where confidence directly drives price.
- Occupancy and tenancy state are as material as physical attributes and yet are the least well joined today.
- The result is either broad rating with margin priced in, or referral to underwriter time.
Section two. Where the spine changes the underwriting economics
- Confidence as a rating input. Where the record is high confidence, price sharpens and referrals drop.
- Provenance as an audit trail. Regulator and reinsurer both consume the same record.
- Tenancy state as a rating input. Void, tenanted, HMO and short let are all differently exposed.
- EPC as an efficiency and moisture proxy. A to C bands correlate with fewer escape of water claims.
Section three. Green insurance products on shared data
- EPC A to C households can be offered a rate differential aligned with lower expected loss.
- Retrofit journeys can be underwritten in stages against the same record.
- Solar, heat pump and battery installations become underwriting inputs, not surprises.
- The shared record makes green pricing defensible to regulator and consumer.
Section four. Claims and the return loop
- Claims outcomes are the insurer's equivalent of the lender's default data.
- Aggregated claims signal, returned to the enrichment layer with consent, is a defensible asset for the insurer.
- The join lets a loss adjuster arrive at a property with the same record the underwriter used at inception.
- Fraud detection improves where the record is stable and provenance is preserved.
Section five. What insurers need from the spine
- Confidence and provenance as contracted fields on every attribute.
- Event based change notification. When occupancy or tenancy changes, the insurer knows.
- Clean room evaluation of price and loss models without publishing rating logic.
- Consent aligned data flow for shared use across insurer, lender and agent.
Section six. The operating model in practice
- Broker platforms and MGAs consume the record at quote.
- Panel providers deliver attributes with confidence, not just facts.
- Reinsurers evaluate portfolio exposure on the same record.
- Regulator gets a coherent audit trail for market conduct reviews.
At a glance
Outcome measures
Underwriter referral rate
Reduce where confidence supports auto rate
Combined operating ratio
Improve through sharper rating and fewer surprises
Fraud detection
Improve through stable record and provenance
Time to bind
Reduce for standard and near standard risks
Roundtable brief
Property insurance roundtable
A closed room of eight to ten property insurers and MGAs, one enrichment partner and one flood or subsidence specialist. Three hours, Chatham House. We test three propositions. One, that the join is the unit of value in insurance underwriting. Two, that confidence and provenance are rateable fields. Three, that a shared record is the route to defensible green pricing. Output. A joint statement on the property insurance operating layer.
Related reading