The problem

Insurance claims considered as one of the most important factors for the profitability of insurance companies. Therefore, the estimation of the number of clients that are intend to procced to a claim announcement, contributes to the proactiveness and preparation of the company in order to keep claims in low levels. Additional advantages for the company may be considered as the prevention of “highly potential risk” clients from demanding an indemnity and a more reliable forecasting of future claims.

The desideratum

Predicting the percentage of customers that intend to announce a claim in a given period, typically monthly, quarterly or yearly and/or for a given sector (i.e. life, health, auto, property etc.).

The solution

With the use of machine learning and artificial intelligence models that will rely on the historical data of a given period (i.e. previous year, six months, etc.) a profile of the customers that intend to announce a claim can be recognized.

The benefit

The usefulness of the predictive model lies not only in identifying customers that are going to announce an insurance claim but also in identifying the main factors related to this decision for each individual case (e.g. specific lifestyle habits). This knowledge help insurance companies to anticipate the risk of factors that most related to a claim announcement and also help them to predict more accurately their future costs.