The loss ratio moves and nobody can say why
Rate, exposure mix, claim frequency and severity all sit behind the number, and reporting that shows only the ratio leaves the discussion nowhere to go.
Insurance
Loss ratios you can decompose, and reserves you can defend
Policy, claims and finance data reconciled, so loss ratio movements can be attributed to rate, mix, frequency or severity rather than noted and left.
Insurance reporting tends to arrive accurate and late, and aggregated to a level where nothing can be done about it. The policy system, the claims system and the ledger disagree on cohort, period and exposure basis. We reconcile them so the loss ratio can be broken into the parts that explain it, and so reserving, pricing and reinsurance decisions rest on the same figures.
Discuss your operationWhat we usually find
The problems that come up in almost every one of these businesses.
Reserving runs on a parallel set of data
The actuarial extract is built separately from the management pack, so the two disagree and reconciling them becomes a quarterly exercise in itself.
Expense allocation is arbitrary
Acquisition and servicing costs are spread by premium rather than by activity, which makes the combined ratio by line more of an accounting convention than a measurement.
Where we start
What the first engagement usually covers.
Scope depends on the state of your systems. These are the pieces that recur in this sector.
Loss ratio decomposition
Movement split into rate, mix, frequency and severity, at the level of the product and the distribution channel.
One reserving dataset
Actuarial and management reporting built from the same reconciled claims and exposure data, on agreed cohort and period definitions.
Expense by activity
Acquisition and servicing cost attributed to the work that caused it, so the combined ratio by line means something.
Reinsurance and accumulation
Ceded positions and exposure accumulation visible against the same data as the gross reporting.
What changes
What is different afterwards.
You own all of it: the code, the written definitions and documentation aimed at whoever maintains this after us.
- Loss ratio split into rate, mix, frequency and severity
- Actuarial and management reporting from one dataset
- Expenses attributed by activity rather than premium
- Ceded position and accumulation visible on demand
The builds behind it
Sector knowledge decides the order. The builds are the same four.
Single source of truth
Your finance, customer and operational records joined into one dataset, with each figure defined once and traceable back to the system it came from.
Analytics layer
Reporting built on top of the joined data, aimed at the few revenue and cost drivers that actually change the result.
Workflows
One defined task, automated inside a process that already exists, measured against whatever it replaced.
Financial modelling
Driver-based models built on the same definitions as your reporting, so the forecast and the actuals stop disagreeing.
Worth asking
Questions this tends to answer.
- What is actually driving the loss ratio?
- Which lines are profitable after real expense allocation?
- Are we reserving on the same data we manage on?
- Where is exposure accumulating faster than we priced for?