Tag Archives: Vodafone

Telematics: Moving Out of the Dark Ages?

While the number of usage-based insurance (UBI) policies reached 14 million at the end of September 2016, most insurance companies are still overwhelmed by the challenge of using collected data to rate their customers’ driving habits.

This conclusion is based on analyzing the world’s 27 largest UBI programs, including those of Admiral, Allianz, Allstate, AXA, Generali, Desjardins, Direct Line, State Farm, the Hartford, Unipol, Uniqa and Zurich.

See also: Why Exactly Does Big Data Matter?  

Progressive, the No. 1 telematics insurer globally, still uses a temporary device and does not collect GPS data. Unipol, the No. 2 player, still only collects mileage data from its customers.

We believe, however, that the prehistoric age of connected insurance analytics is ending. The era was based on the premise that all policyholders are reluctant to be “tracked.” But with most of us giving daily credit card, fingerprint, driving speed or location details to companies such as Apple, BMW or Vodafone, how to make sense of the self-censorship that insurers apply to their programs?

The truth is that more data benefits insurance companies… and the careful drivers! At the center of this change is advanced data analytics – the ability to extract insights from real-time data sources and discover risk-predictive patterns.

Our analysis, detailed in the Connected Insurance Analytics report, shows that the glaciation period’s ice is melting and that all the key insurers are now moving.

See also: Data Science: Methods Matter (Part 3)  

Progressive started a vast recruitment plan to attract data scientists. Generali also made a strong move by acquiring MyDrive, an analytics provider with early footsteps in smartphone UBI. Allstate just created Arity, which will collect data on drivers and sell analytics products to third parties. Simultaneously, Unipol created Alpha, a self-standing analytics and telematics operation.

The bulk of insurance companies is yet to act. To help them adapt to this new climate, Ptolemus published the Connected Insurance Analytics (CIA) report as a step-by-step guide to advanced analytics. It describes, analyzes and illustrates the process by which advanced analytics companies take raw driving data and transform it into real-time, individual risk profiles.

Screen Shot 2016-12-06 at 9.42.20 PM

The investigation shows that acceleration, braking and mileage are the most used — unsurprisingly — but also that the range of factors is much wider and illustrates the complexity involved in selecting the correct criteria.

To offer a predictive driving score, the report demonstrates that insurers must gain a deep understanding of driving conditions. Adding contextual data, such as road type or relative speed, is a necessary step to price customers fairly.

The full article from which this is extracted is available here.

How Analytics Can Prevent Fraud

What is common between Uber, Amazon, EE, Vodafone, Netflix and Progressive? All these companies have recently faced issues because of fraud.

In insurance, ghost brokers often target young drivers who want to cut the cost of their car insurance. Even though it is the ghost brokers who commit the fraud, the customers lose their cash and also risk a criminal record. And fraudulent claims from customers is a much bigger concern for insurers.

How big is insurance fraud?

The Coalition Against Insurance Fraud, America’s anti-fraud watchdog, estimates that nearly $80 billion in fraudulent claims are made in the U.S. annually. Fraud increases insurance premiums, raises the cost of goods and services and boosts spending on investigation and fraud-prevention programs by insurers.

Fraud is one among the many business challenges that insurance industry is facing, as I have outlined in my previous blog. IDC estimates that insurers spend approximately $100 billion on IT, of which $3.3 billion is spent on information security and to counter financial crimes. Four of the five biggest property & casualty (P&C) insurers have formal anti-fraud programs.

Analytics to detect insurance fraud

Though application fraud, underwriting fraud and premium fraud are also significant threats for insurance business, claims fraud has been the industry’s main focus. Major insurers started deploying new platforms to transform their claims management and minimize fraud. There is a spectrum of vendors from big IT players to niche analytics players that is providing claims fraud detection solutions. Zurich’s UK general insurance business recently deployed end-to-end claims management transformation in association with a major insurance vendor, which minimizes losses associated with fraud.

Manual detection of fraud is next to impossible in the insurance industry, as it is costly and the sheer volume of claims is too high to handle for any insurance company. Also, the velocity, the variety and the veracity of data generated in the claims handling process made the use of statistical models based on sampling methods obsolete.

Because analytics integrates data from diversified channels and combines internal data with third-party data, effective fraud detection can be made possible. Many insurers have started using analytics techniques such as reporting, descriptive analytics, predictive analytics and prescriptive analytics to detect fraud. For example, CNA, the 8th-largest commercial P&C insurer, implemented analytics and predictive modeling to identify claims fraud. In two years of implementation, CNA reportedly saved $6.4 million, attributed to recovered or prevented fraudulent claims.

In this post, we will see how insurers have started adopting innovative technologies along with analytics to detect insurance fraud, beyond traditional analytics techniques.

Social network analysis

A recent AM Best survey found that more than half of the companies surveyed use social media. Life insurance companies appear to be most likely to use social media (65%). Company size is also a driving factor for social media. The larger the company, the more likely it is to use social media. Insurers have also started using social media data of policyholders to investigate and detect claims fraud. For example, if a non-smoker applicant lights up even occasionally and if his social pages has the traces of that information, it can be detected via social network analysis (SNA) tools. SNA tools scan large amounts of data from business rules, statistical methods, pattern analysis and network linkage analyses to uncover possibilities for fraud.

Drones

Insuring drones and wearables is going to be difficult for insurers, as there are a multitude of insurance liability and coverage issues. However, insurers have started using drones for their benefit by adopting them in claims adjusting. USAA appointed its first drone pilot for claims handling. With drones in place for claims handling, insurers would no longer need to climb dangerous chimney or to visit catastrophe sites. Also, the data analysis from drones is used to detect insurance fraud. A British company, Air & Space Evidence, detected a fraud case after Hurricane Katrina with the help of drones. A couple who claimed that their home in New Orleans was severely damaged by wind and water was found to be committing fraud when aerial photos showed that the house was intact.

Wearables

One third of the insurers surveyed are already using wearables for customer engagement, according to Accenture’s 2015 technology vision report. We all know that disruptive technologies such as telematics and wearables (Oscar’s Misfit, Fitbit and the Apple Watch) have also begun to be used for calculating customized premiums. What’s new is, in Canada, data from a Fitbit wristband was used by a personal injury lawyer to support his client’s case. Soon, these technologies will also be used to detect insurance fraud as the data collected from these devices will become fodder for criminal and civil litigation. Insurers now have many reasons to turn to innovative technologies and analytics to protect themselves against fraud.

Please share your thoughts on how you have seen innovative technologies and analytics helping insurers to combat fraud and transforming the insurance industry. In the next few blogs, I will try to explore analytics’ role in the insurance industry in further detail.