The insurance industry is highly competitive, heavily regulated, and subject to a number of legal restrictions. Staying on top of regulations is a constant battle, as is staying ahead of the competition. Companies must differentiate themselves from their rivals, and usually do so by offering lower prices, greater efficiency, and customer satisfaction, or offering niche products that target a certain market.
In order to achieve these objectives, insurers are turning to data analytics. According to the 2015 Bain & Company’s benchmarking database, which surveyed 70 insurers, annual spending growth on Big Data analytics is set to reach 24% in life insurance and 27% in Property and Casualty (P&C) on average over the next three to five years. However, despite this growth, the use of analytics in the insurance industry is far from mature, with the Bain survey also finding that around one in three life insurers and one in five P&C insurers do not apply analytics in any function.
The potential is vast, and given the intensity of competition in the industry, anyone who fails to realize the benefits will fall behind rapidly.
There are a number of areas around which insurers can apply data analytics, the most important of which is setting premiums. The core of insurance is evaluating the risks of insuring an individual and setting the appropriate premium. This has, traditionally, been done manually by an analyst, which is slow and often inaccurate because it relies on a relatively small pool of data, because this is what a human is capable of analyzing. With big data and machine learning techniques, algorithms can be applied to all of an individual’s data - financial, health, actuarial, claims, risk, and so forth. Premiums must be set at a level whereby the insurer can make a profit by covering their risk, but will also be affordable to the customer. Data analytics can help better reach this figure, and far quicker.
The key to modern insurance - as in so many areas - is personalization. People don’t want to be paying an insurance rate based on the behavior of others in their supposed demographic, they want to pay according to their usage and their own behavior. This is being better enabled particularly by technologies such as the Internet of Things (IoT) and wearables, which provide a wealth of data to insurers in real time that give them a better understanding of their customers’ behavior - something especially important in areas such as health. Many offer discounts to customers willing to install technology that monitors their behavior. In car insurance, for example, you can feed actual driving information back to their system to help construct a profile of an individual customer’s behavior to work out a more accurate assessment of that driver’s likelihood to be involved in an accident or be involved in a theft. In the case of home insurance, two US-based providers, Liberty Mutual Insurance and American Family Insurance, have partnered with smart-home IoT company Nest Labs, offering a discount to customers on their premiums in exchange for using the device as it lowers their risk of fire.
Another important application for analytics is fraud prevention. In 2014, insurers found 150,000 fraudulent claims which cost over £1.3 billion, and this is likely just the tip of the iceberg. One example of a company successfully using data analytics to combat fraud is South Africa insurer, Santam. They worked with IBM in an attempt to reduce fraud rates for their medical products, analyzing three years of customer data to pinpoint areas fraud was likely. As a result of the insights they garnered, they identified a huge fraud syndicate within a month. They were also able to put into place new processes that saw savings of $2.4 million in the first four months.
There are many such instances where insurance companies are successfully deploying data analytics to the betterment of their business. However, there are also many areas where they are still falling down, and insurers need to better use analytics if they are to stay ahead of the game.