Brokers need smarter data processing

Fraudulent claims continue to cost assurers and consumers millions a year. While settling claims quickly and accurately is a key customer satisfaction driver, inconsistent processes in terms of dealing with fraud and dishonesty affect profitability. The modus operandi of fraudsters and syndicates is adaptive and evolves over time. Fraud is shifting as assessors and SAPS investigating units clamp down on specific products and areas. It is therefore important for brokers and assurers to adapt faster than the fraudsters by employing tools that make fraud prevention integral at every point in the assurance transaction.

How to mitigate risk

There is a seemingly endless supply of statistics that shapes company strategies as they attempt to mitigate risk. Traditional models suggest that important elements of this strategy include:

  • Identification of potential problems during underwriting at time of application for the policy.
  • Using off-the-shelf or in-house developed models to properly segment claims and reduce workload.
  • Adapting tools and techniques to detect high-risk claims.

Most assurers have underwriting measures in place to mitigate risk at underwriting stage (this refers primarily to life products). Generally speaking, the lower the financial exposure, the less is being done to prevent fraud during the sales stage.

Assurers have measures in place to segment suspicious claims to manage their exposure. Approaches range from basic, where individuals work through claims to identify suspect claims, to the use of systemised predetermined decision rules when specific characteristics exists. Assurers also have tools and techniques in place to identify dishonest claims. Again, these measures vary, from human scrutinising of claims to the
use of predictive analytics.

The importance of data

Predictive analytics can be used to analyse the historical claim data and uncover hidden claim characteristics to avoid costs associated with claims.

While an approach with significant potential value, utilising predictive analyses presents assurers with a unique set of challenges. Institutions (assurers, brokers and assessing specialists) need to have the ability to compile the data necessary to build a good predictive segmentation model. For models to work effectively, the data needs to not only provide volume but span many characteristics, not all of which are easily extractable for analyses. While the industry’s custodians of data typically have a large quantity of data, it is often too narrow in scope. Models are consequently often unable to produce the desired results.

The broker’s dilemma

Brokers are largely dependent on assurers and underwriters to direct and manage data collation at sales stage – few have the capacity or scale to make it financially viable to develop their own prediction or segmentation models. The modus operandi of syndicates is also of such nature that the segmentation of individual broker models will hardly deliver the rewarded results. Effective models are dependent on volume and characteristics that make it challenging for the many individual brokers and smaller assurers.

Dealing with the challenge

There are ways that brokers, assurers and assessing experts can address this data challenge by bringing additional claims data to the table. This additional data may include not only personal but also demographic and other less obvious data that might require third-party data partners to develop effective prediction and segmentation models.

If the data quality and quantity is acceptable, current and is being provided in a way that supports refreshing the model, it can mean the difference between success and failure in the prevention and detection of fraud. Ideally, the model should be able to identify potential problems during underwriting.

The way forward

Effective segmentation coupled with actions, such as rule-based decision making, profiling and integration of data, will provide best practices that will directly contribute to brokers’ and assurers’ ability to control more of the fraud risk that they are exposed to.. With ever increasing claims costs, brokers and assurers should also improve their efforts to find new ways to reduce costs on all fronts. With properly implemented segmentation analyses, role-players now have a fighting chance to win the ongoing battle against fraud and dishonesty.