Our work

Understanding intermediaries better

The Brief

More Metrics was asked by a leading protection insurer to help them understand their intermediary distributors better.
The insurer is a leading provider of term life, critical illness and income protection business. It distributes through a number of channels, the largest of which is the independent and restricted-panel intermediary channel. This is formed of many different firms, each of which has their own target markets, approach methods and sales processes. Each of these affects the business the intermediary provides to the insurer.
The insurer asked More Metrics to provide a mortality model that they could apply to business written by intermediaries so that it could better understand the quality of the business that the intermediary was providing to it.

The Output

More Metrics provided the insurer with our Geo Mortality model for them to use. This is a sophisticated data models using freely available "open source" data. Most of the data is provided by the Government's Office for National Statistics and is derived typically from Census data and Government collected data on local area characteristics.
The model is tagged by Output Area (OA), each of which contains about seven individual postcodes of similar characteristics. This model was accessed using the latest open-source Postcode to OA mapping file, that allows for any newly-formed postcodes.
The model was also tagged by age-band and gender, as information readily available to the insurer on new policies.
The insurer added the model results to their written business files, and then analysed the results by distributor. This was made easy by providing the model results as a percentage of a standard actuarial table.
The result was a simple expected average mortality percentage for business written by each distributor. The higher the mortality score the greater expected number of claims on the business and therefore lower profitability.
The results for the main distributors were generally in-line with the insurer's expected quality of business from subjective and other measures. However, the model was able to quantify the differences, which in some cases were greater than expected.
The model was also useful in quantifying the quality of smaller distributors with whom the insurer had little direct contact, and for whom other measures such as early lapse rates were too volatile.
The insurer was therefore able to refine its profitability analysis by distributor and so adjust its distribution strategy and terms of trade accordingly.