By Peter L. Miller, CPCU
As insurers rush to find competitive advantages in big data, many are devoting considerable resources to predictive and prescriptive analytics and systems that aid organizations in more accurately evaluating and pricing risk.
These firms are betting big on complex algorithms and measured process updates as they scrub and analyze massive amounts of static data they’ve collected. With the right predictive analysis, this historic data can serve as a powerful road map for predicting market trends and customer behavior.
But as organizations seek to carve out their individual approach to big data, some are choosing to focus less on complexity and historical datasets and instead are prioritizing speed and immediate insights with real-time data analytics. Real-time data analytics is more like a GPS system than a map, providing fast updates and course corrections for the most effective route forward.
Insurers’ ability to monitor risk and customer behavior, and to take actionable steps in real time or near real time, is leading to improved underwriting and pricing of current products, as well as new models of selling insurance based on monitored use and a steady stream of information.
Across all industries, including insurance, the push for real-time data analytics has been driven primarily by two significant technological developments:
- The rapidly evolving Internet of Things (IoT) and the growing network of connected devices
- Advancing capabilities that allow firms to analyze and process high-velocity data
Telematics and connected devices
Devices that share information over the cloud have implications for pretty much every corner of the insurance industry. Here’s just a sampling of how the IoT can be applied to coverage:
- Vehicle telematics
Telematics and sensors that send real-time driving data to insurers or fleet managers can provide immediate safety interventions and help establish accurate risk profiles and pricing.
In fact, SAS estimates that by 2020, more than a quarter of auto insurance premium revenue in the United States will be based on data generated from telematics. That’s more than $30 billion.
Drivers who are speeding, tailgating or taking other risks behind the wheel can receive alerts and warnings in real time. For commercial vehicles and rental cars, this information can be sent directly to fleet operators and vehicle managers.
If a sensor detects an accident, it can immediately initiate a claim or even contact emergency responders.
- Wearable devices
Health insurance companies are beginning to experiment with wearable devices, providing financial incentives for customers who demonstrate healthy lifestyle choices. So far, this has mainly been limited to the use of pedometers, which track daily walking distances, but look for this to expand in the future.
In the workplace, these wearables can provide immediate feedback on unsafe actions, such as dangerous lifting, and can track fatigue, body temperature and heart rate. In some cases, they can prevent a worker from engaging in an unsafe behavior altogether.
Wearables can also help improve security through GPS updates and automatic alerts when individuals enter restricted areas. With the ability to take corrective action in near real time, all of these wearable devices can help prevent a claim or more accurately predict the likelihood of a claim over time.
The connected devices we carry with us can provide insurers with a wealth of actionable data.
From GPS information to social media updates, insurers can pick up on a customer’s current activity and send updates, from weather and safety warnings to targeted marketing materials.
All of this data is leading to a whole new market of insurance, with disrupters and established players getting involved and creating new products and other offerings. One area that’s gaining particular prominence is usage-based insurance, in which customers pay an insurer based on the time they engaged in the insured activity. Of course, accurate, real-time updates about usage are the foundation of this model.
According to one Allied Market Research study, usage-based insurance will be a $123 billion market globally by 2022. Much of the current investment and adoption is centered around auto insurance and pricing policies based on how much people drive.
However, it’s easy to see applications in other insurance markets based on how often people need coverage and the behaviors that influence risk and the chances of a claim being filed.
Clearly, this increase in the amount of data collected about individual customers is a huge opportunity for the industry.
In addition to new products and pricing models, real-time data and historical data can be combined to significantly reduce fraud, create more targeted marketing materials and streamline the claims process.
Another significant benefit is that premiums can be based on more relevant, individualized customer data pertaining to the covered risk, reducing or eliminating the industry’s reliance on other traditional, less precise data such as education or gender.
However, the industry needs to take careful, measured steps to implement these emerging datasets, with a keen eye toward risk modeling and public perception. These new products and pricing schemes should prioritize transparency and informing customers of how their behaviors and personal data are affecting their coverage. Although there is great opportunity to improve business processes, significant individual privacy concerns also need to be considered.
As insurers work to collect and act on data in real time, it’s important to note that real-time data does not go away after it’s collected. It becomes historical data, ripe for predictive and prescriptive analytics, which will be further augmented by machine learning and artificial intelligence capabilities.
Organizations that get a head start on collecting and analyzing this data will have a clear competitive advantage as others struggle to catch up. Insurers that can capitalize on opportunities in real-time data analytics now will drive the next era of big data applications in the industry.