Evidence guide
How to read insurance claims data
Insurance claims measures answer different questions. Reading them together is useful, but collapsing them into one verdict is not.
Acceptance needs a product boundary
Claims acceptance is the proportion of registered claims accepted. Higher can be favourable, but only when product definitions and reporting practices are comparable.
The FCA identifies inconsistent reporting for home-insurance acceptance. Behind Your Money shows that aggregate neutrally and suppresses directional firm comparison for it.
Frequency is not quality
Claims frequency measures registered claims relative to policies in force. It can reflect weather, customer mix, cover design and insured risk.
A higher or lower frequency does not by itself mean that an insurer serves customers better, so it remains a contextual measure.
Payouts reflect claim severity
Average payout can differ because one product covers expensive losses while another handles smaller, more frequent claims. It should not be ranked as if higher always means better.
The market aggregate is precise, while firm-level observations are published in ranges. The site preserves those ranges rather than using their midpoint as a reported fact.
Use history carefully
The 2025 FCA release includes a restated 2024 comparison. This is useful for direction, but it is not the same thing as preserving the originally published 2024 release.
Product entry and exit also change the firm population. The product pages disclose how many firms appear in both years.