Usually companies access claims and risk information in a fragmented and random manner.

1) Service providers like insurers and brokers keep their customers' claims data in different systems. This often leads to incomplete and inconsistent information.

2) Even standard statistics from completed claims files have limited value because data from brokers and insurers often ignores changes in the underlying risk. Analyses based on this data don't reflect the impact of mergers and acquisitions, or changes in the coverage.

3) Apart from internal corporate factors, insurance risks have to take into account external influences. It is only worth comparing claims from different financial years if inflation and currency fluctuations are discounted. It is also important to consider reported claims reserves (IBNER) as well as the run off within the portfolio (IBNR).

4) Usually the claims data of insurance companies is treated separately for each line of business and is often held in different formats. This ignores the potential to optimize from using multiline structures.

5) In practice, claims analyses are triggered by concrete demand and often built up from scratch. This is a costly and long-winded approach to the rapid adjustment of the insurance strategy necessary to match changing basic conditions.

To build an insurance program based on BEST DATA we have to consider many factors. However, the credibility and transparency of the underlying data is crucial, and any extra work is more than offset by the reduction in the cost of internal communication.