Why the use of big data in the insurance industry "from 0 to 1"

It is well known that the insurance industry is in a stage of technology-driven change, with the Internet, mobile social networks, cloud computing and big data as the representative of the digital technology is accelerating the impact of the insurance industry's daily operations.?

"Of all the new technologies, big data has the most disruptive impact on the insurance industry." The Boston Consulting Group and the China Insurance Association recently released the "Internet + Era, Big Data Improvement and Reform of China's Insurance Industry," which points out that, on the one hand, big data analytics will "improve" the day-to-day operations of the traditional insurance industry, and that this impact will be reflected in all aspects of the value chain, including risk assessment and pricing, cross-selling, and customer churn management, This impact will be seen in all aspects of the value chain, focusing on risk assessment and pricing, cross-selling, customer loss management, claims fraud detection, and claims prevention and mitigation; on the other hand, big data and the Internet will also "disrupt" traditional insurance business boundaries and business models, such as usage-based insurance (UBI) and platform-based ecosystems, and bring about a large number of cross-border competition and disruption scenarios.

In fact, the impact of big data on the insurance value chain is reflected in all aspects, according to the Boston Consulting Group, the most important "improvement effect" occurs in five links, namely, risk assessment and pricing, cross-selling, customer churn management, claims fraud detection, claims prevention and mitigation.

As far as risk assessment and pricing are concerned, in the era of big data, risk characterization is greatly enriched, and access to data resources will become more and more convenient. In the field of auto insurance, in addition to obtaining data on car models, car zero-over-all ratios, and used car data, insurers are also using in-vehicle sensing devices to collect driver behavioral risks and develop UBI auto insurance; in the field of life insurance, insurers are able to monitor human health in real time (exercise, sleep, heartbeat, etc.) by using wearables, which makes up for the lack of the ability of the life table to provide insights into the health of human beings in niche groups and the probability of life and death.?

It is worth mentioning that the analysis of unstructured data from the Internet and social media can help identify potential risks to consumers. For example, ZestFinance in the United States analyzes more than 10,000 pieces of Internet data from loan applicants to provide high-quality guarantee assessments for bank loans, credit cards and insurance, resulting in a default rate about 60 percent lower than the industry average. And Ping An of China recently revealed that its Qianhai Credit will join hands with insurance organizations in the future to help identify the potential risks of policyholders for accurate pricing and fraud identification.

And for the most important means of maximizing customer value and promoting business synergy, cross-selling can also be enhanced in the era of big data. Given that only by segmenting and gaining insights into customers and precisely understanding their key needs can the accuracy of cross-selling be significantly improved. The Boston Consulting Group believes that insurers need to build an analytical customer relationship management platform to unify the management of customer data and establish customer analysis models, to take advantage of *** enjoyment and pooling, and to avoid specialized companies working separately. For groups with uneven business structure, it is more suitable for strong business to drive the development of weak business. If customer resources can be realized across legal entities***sharing, at least 10%~20% of the potential market value can be tapped.

In addition to the above, insurers need to build an analytical customer relationship management platform to unify customer data and establish a customer analysis model.

In addition, with the help of big data tools, insurers can also significantly improve the accuracy and timeliness of anti-fraud. Big data models can automatically identify possible fraud patterns in claims, potential fraudulent behavior of claimants, and possible fraud networks. At the same time, it is important to ensure that data resources, the more complete and diverse the data, the more likely it is to identify possible fraud through sophisticated algorithms and analytics, where the necessary data include claims history, policy information, other insurance company data, health insurance data, accident statistics, credit records, criminal records, social network data, etc.?

It's worth noting that while insurers are placing great importance on the application of big data, as Ping An Chairman Ma Mingzhe recently said at the company's half-yearly report communication meeting, "not everyone has big data, 99 percent of companies, including Internet companies, have just information, which can't be said to be big data yet."?

Ma Mingzhe frankly, to judge the whole picture of a person on the Internet, must master its 3600 different factors of data, although Ping An of China has more than 20 financial companies, with more than 700 million users of multi-dimensional information and data, including several hundred factors, but it is also the tip of the iceberg only. The so-called big data, there must be a large enough volume and frequency, to have diversity, the user's consumption data, social data, daily behavioral data, etc., and be able to intelligently interconnected, dynamic analysis, otherwise it is just a localized information.?

Among the insurers researched by Boston Consulting, 63% of auto insurance companies have carried out Telematics applications, and 16% have carried out platform ecosystem practices. Boston Consulting Group pointed out that compared with the European and American markets, promoting UBI auto insurance in China seems to be "a bit awkward". Considering that the overall profitability of auto insurance in China is worrying, using a more favorable value as an entry point is likely to result in a greater degree of industry losses. Unless insurers can use the car network to better select risks, identify claims fraud and provide value-added services, the pursuit of quality improvement in the overall cost rate and customer satisfaction.

And for the current application of wearable devices in health insurance, at present, domestic insurers generally take a wait-and-see attitude, although it is believed that wearable devices have great potential for future development, but the legal and ethical risks are huge, and it is urgent to further improve the relevant laws and regulations, so compared to human health data, insurers prefer to obtain electronic cases from medical and physical examination organizations for claims correlation and product pricing. The company's newest product is a new product that will be used in the next few years.

However, the current insurers are unanimously optimistic about the vertical platform ecosystem, believing that in the Internet era, insurers are bound to carry out a number of collaborations with various industries, provide a package of services,**** with the construction of a platform ecosystem of digital insurance. In this regard, the Boston Consulting Group suggests that the construction of the ecosystem is difficult and the cycle is long, which involves the improvement of the business model and resource integration and many other difficulties, and requires the insurers to be courageous to invest and patiently seek.

The above is what I shared with you about the use of big data in the insurance industry, why "from 0 to 1" of the relevant content, more information can be concerned about the Global Ivy to share more dry goods