External data AKA data enrichment is becoming increasingly important for insurance pricing leaders, this is because the key power of external data is to avoid anti-selection while selecting against your competitors. That’s a double win for a pricing leader.
External data means any information that insurers obtain from outside sources other than the policyholder. Such as government records, public databases, and third-party providers. This data can be used to supplement internal data, providing a more complete picture and allowing insurers to make better-informed decisions about pricing. In the words of Sun Tzu, “He who knows the terrain, knows where to attack and where to defend”.
Currently, different countries are at different points with the adoption of external data to price insurance. For example, the USA is a leader in real-time motor telematics. While the UK is king of peril data. The future is telematics and IoT data for anything that changes rapidly and up to date lookups for anything that does not.
Accurate price setting absolutely leads to better loss ratios and bigger books. For example, verifying a person's address can improve loss ratios significantly. In one instance, approximately 1% of an insurance book did not have a verified address, and this group was responsible for about 10% of the claims cost for the entire book. This insurer stopped offering policies to customers if it could not verify them and profited by millions.
Source of External Data
There are several models that insurers can use to source external data during the insurance pricing process. Here are a few examples:
● Data Brokers: Data brokers specialize in collecting and selling data from a variety of sources.
● Telematics: Telematics involves using sensors and GPS technology to collect data.
● IoT devices: Internet of Things (IoT) devices such as detectors and cameras can collect data.
● Data providers: Specialists in particular types of data who provide this properly packages so it can be used for insurance pricing
By using these sources, insurers can gain a more comprehensive understanding of risk factors and make more informed pricing decisions. However, it's important to ensure that data is collected and used in compliance with applicable laws and regulations and that consumer privacy is protected.
What about modelling
In my view better prices almost always come principally from putting resources into finding more data rather than using more complicated modelling techniques. No amount of clever modelling will tell you what you need to know. Take an example from property flood. If you have a new property and you need to decide on its flood premium, no amount of modelling will beat knowing the property’s height above the water table.
Challenges Associated with External Data
While sourcing external data can provide valuable insights for insurers during the pricing process, there are also several challenges that insurers may face. Here are some of the main challenges associated with external data sourcing during insurance pricing:
Data quality: External data may not always be reliable, accurate, or up-to-date.
Data privacy: External data may contain personal information about individuals, which raises privacy concerns.
Data access: Some external data sources may be difficult to access or may require special permissions or agreements to use.
Data integration: External data may be challenging to integrate into the rating.
Cost: External data sources may come with a cost, although a surprisingly huge amount of data is available for free.
Conclusively, External data is becoming increasingly important for insurance pricing leaders; this is because it helps insurers make better and more informed decisions, improve loss ratios, and increase volume. None of us would be foolish enough to insure anything we have little or no knowledge about, therefore, it’s important that we find out everything we can about the things we want to insure. We may get general information such as the size, height, weight, length, power, number of seats, and safety rating of a vehicle. We could also enhance this with look-a-like information such as looking at our customers' neighborhoods for the average wealth, health, income, family size, education, and property types.
The we could find out our customers' credit scores, driving license information, bank balance, and previous claims if possible. Bottomline is getting as much information as possible that would enable us to make informed decisions.
● Write down what information you would like to know about your policies and policyholders but don't currently have; then seek out data providers who may have this information.
● Ask yourself some vital questions like; What would happen to you if your competitors knew key information about customers that you do not know?
● You can start with free and open-source data such as census records.
● Build up a repository of data that you can attach to your policies and test for significance.
● Ensure you build processes that mean you can easily update this data in the future.