How AI and ML are changing insurance for good

By Alan O’Loughlin, Director of Analytics and Statistical Modelling, International and John Beal, Senior Vice President of Analytics at LexisNexis® Risk Solutions

The Insurance industry has been dealing with vast volumes of data for years, but analytics, Artificial Intelligence (AI) and Machine Learning (ML) techniques are increasingly being used to help insurance providers make faster data driven decisions.  Given the exponential level of data available today with AI/ML, insurance providers can now efficiently extract new insights into their customer’s needs and create stronger long-term value.

Personalising Insurance Pricing

Starting with how the market calculates premiums, the insurance sector now has access to thousands of data points to help them calculate premiums. Machine learning algorithms expedite the identification of the most predictive attributes driving claims losses – the most recent data points being historical cancellation data and gaps in cover.

This helps insurers become more competitive, match their risks to the most appropriate pricing strategies and write the risks that meet their underwriting appetite. In turn, customers get more personalised quotes based on their unique risk characteristics across any line of business

Achieving a single customer view

Personalisation within any sector works best of course when you really know who you are dealing with. Today, an explosive amount of data is collected, but it is vastly under-utilised as many organisations do not have the expertise to bring data together from different parts of the business to create a single customer view.  Add to this, the amount of mergers and acquisitions in the insurance market over the past few years and the challenge of managing multiple customer databases.  Linking and matching technology using policy history data to find common threads helps overcome this problem to create one consolidated view of the customer. Optimised matching algorithms are also the most accurate and relevant data is reviewed, reducing consumer friction during the quoting process.

Normalisation makes sense of masses of data

In the same vein, as organisations aggregate massive volumes of data, the value of cleansing and normalisation can’t be overlooked. One example, as usage-based insurance develops, whether through aftermarket telematics devices, smartphone apps, connected vehicles, even in the future from smart home data, all that data needs to be gathered, normalised, standardised. That way, any consumer can enjoy an improved shopping experience based on their needs and preferences, no matter the device brand and insurers have consistent quality standards and outcome decisions for all consumers.

Making Vehicle Data work for insurance

Data normalisation is already helping insurance providers understand the presence of Advanced Driver Assistance Systems (ADAS) on a vehicle at the quotation stage.  An ADAS classification system has been created using machine learning to scan millions of lines of car manufacturer vehicle data to logically sequence and classify vehicle safety features and component’s intended operation or purpose.  Extraction and proper classification of this type of data is extremely difficult, time consuming and error prone without the use of AI/ML

Thinking big, starting small in motor claims

At the claims stage in motor insurance, image recognition technology is being used to capture damage or invoices, run a system audit, and if the claim meets the approved criteria, it is automatically paid without human involvement.  This kind of virtual or ‘touchless’ claims handling is speeding up claim settlement times, cutting costs and improving the customer experience.  The ability to quickly analyse years of historical policy and quote history at the consumer level will add an additional level of security prior to a carrier releasing any claim payments.

Building context through AI and ML

Staying with motor insurance, telematics data can be used much more broadly than originally intended through AI and ML.  From the point of impact through to claim resolution, telematics data can allow insurance providers to get on the front foot at first notification of loss (FNOL), helping to deliver a better consumer experience post-accident, whilst providing invaluable insights regarding the circumstances of the collision.

AI/ML techniques communicate the conditions before, during and after the time of the accident.  Data points like air bag deployment impact sensor activation and g-force metrics can be analysed to understand claim severity and bodily injury potential.  In addition, by combining vehicle build data, carriers can understand the repair cost and potential impact to expensive ADAS features.  Insurance providers can instantly also help their customers with emergency services, vehicle rentals and repairs through instant analysis.

Taking the pain from home insurance applications

Moving into home insurance, we know that conversion rates of people shopping for home insurance is quite low due to a number of hard to answer questions along the customer journey. Rebuild costs is a classic example.  Prefill and data validation solutions are now helping to solve that problem but they are only possible through a huge amount of modelling, linking and AI-ML techniques to pull all the data together to return accurate and up-to-date information on the person and property.

Putting customers in the picture

AI is also at work in the commercial property insurance arena.  It can provide valuable insights regarding a potential location for a new branch or business relocation – footfall, crime rate, exposure to perils or other local circumstances that increase risk. This insight when provided to the customer enables them to take preventative measures if they do go ahead in that location, decreasing risk and loss costs, whilst helping to improve customer experience and retention.

AI and ML can help in the democratisation of data

Finally, AI and ML techniques are helping consumers take advantage of their individual data points which in turn provide the most accurate and updated view data to the insurance providers they choose to interact with on their own schedule.

A good example is the way driving behaviour data from aftermarket devices, or in the future, direct from the connected car gives a clearer picture of someone’s driving risk on the road.  Drivers then benefit from being judged based on their individual behaviours, rather than paying premiums based on average driving habits.

This requires transparency. Each time a consumer applies for insurance they consent to their data being used to provide the insurer with the best information possible, so they can set an appropriate premium based on the risk. Within insurance, we are focusing more than ever on educating consumers about how their data can be used and evaluated in a way they control and understand.  AI and ML automate and process the data consumers are happy to share – supporting greater choice, improved fairness and reduced friction with more personalised insurance protection.