Big Data debates due for Insurtech in 2021

By Joanna England

March 11, 2021

undefined mins

The management of Big Data looks set to create better insights but tighter regulations for insurtechs over the next 12 months...

With the growth of the Internet of Things (IoT), Big Bata has got,well, a lot bigger. The sheer volume of collected information from vast numbers of online devices is the cause of much discussion in the world of regtech.

Currently, for many companies, oceans of data gravity are pooling on servers with no particular place to go. Without adequate aggregation and analytics, such information remains a financial storage drain, and largely useless. Furthermore, without proper regulation in place and the right analytics, the use of the data also becomes a legal hot potato.

Security is a major concern. Facebook, Microsoft, Equifax and Yahoo! have all experienced massive data breaches in recent years. While companies have collected more and more data, they have not always taken the proper precautions to secure it.

The rollout of 5G is generating ever-faster data collection and transfer too. The problem of data storage, security and aggregation, therefore, may well get worse before it gets better.

But, the situation has also led to massive strides in artificial intelligence (AI) and machine learning (ML) which, when employed correctly, are the perfect antidote to Big Data gravity problems.

AI insights and Big Data

Big Data becomes valuable when its insights are extracted. However, only certain aspects of the data are used to build insights, which means much collected by companies, is unnecessary. Callum Rimmer, founder and CEO of By Bits, a new technology platform for the motor insurance industry, believes a more selective approach to data harvesting is required.

He says, “Data will beget data. It is not about collecting everything and anything on the off chance that something might be useful. That just creates noise. It is disingenuous to the supplier of the data and fundamentally creates problems maintaining that data within robust security processes and procedures.”

Rimmer says that while data volume inevitability increases over time, proper segmentation and archiving means insurtech businesses can scale in line with volume while maintaining order, accessibility and portability.

He explains, “Big Data needs structure and a clear application to be of benefit to anyone. As Big Data, both as data itself and the ecosystem around it, grows - it can be used to innovate industry and benefit end consumers.”

However, this is easier said than done, and Rimmer fully admits that the biggest challenge faced by insurtechs is the security of the information they collect. "Protecting sensitive data by managing its access internally and externally" is essential, he says.

“It’s critical to have robust data security policies and processes to maintain ISO27001 accreditation. Processes and infrastructure to be GDPR compliant from the ground up which is an advantage that insuretchs have over most incumbent insurers with legacy systems.”

Privacy rights and Big Data

But this raises other questions regarding how that data is used, and to whom it ultimately belongs. GDPR has partly addressed this problem in Europe. The regulation was a game-changer for its emphasis on citizens’ rights. It was created to give EU citizens control over their personal data, as demonstrated by the rights of individuals within the regulation.

Robert Bond, a data protection expert from law firm Bristows, agrees that data collection and retention in a legal minefield that many enterprises fail to appreciate. He explains, "While the use of Big Data has many valuable commercial facets, there are risks over the use of such data breaking the law.

“If the data is taken from various sources such as websites and apps and third-party databases, care needs to be taken that such use does not infringe the contractual nor intellectual property nor data protection rights of others,” says Bond.

He points out that the use of personal details by insurance companies must be only be utilised with consent to avoid privacy breaches. “Where personal data is processed as part of Big Data and analytics there must be a lawful ground for such processing, such as consent of the data subjects or legitimate interests of the business. Insurance businesses need to ensure that its data processing using Big Data is open and transparent and compliant with the law.”

Useful data collection

Successfully data analytics can streamline companies from the inside out, and provide far better services to customers. “The growing use of digital services, combined with the collection and logging of previous interactions and preferences means there is no excuse for financial service companies to have an incomplete picture of their customers wants and needs,” says Chris Vinnicombe, VP Financial Services at Acxiom

“Firms can take a step closer to their digital future by bringing together the data they have on everyday interactions, requests and comments in order to offer the level of personalisation that benefits customers and keeps them coming back.”

Insurtech and ML

Once consent to data usage is provided, operators are tasked with extracting information from it. With advances in AI, ML and HPC (high-performance computing) this task becomes much easier – and the resulting insights become far more valuable.

“Big Data provides insurers with a better way to understand and support their customers and make more accurate decisions when it comes to risk and fraud,” says Nicolai Baldin, CEO and Founder of Synthesized an artificial intelligence service that enables businesses to research with sensitive data.

He explains, “However, this brings attention to a primary concern: security and data quality. How can you get fast, secure access to high-quality data and the ability to collaborate with your partners, be able to test against all possible scenarios, without putting customer data at risk?”

It's a tricky situation, says Baldin, that is made more complicated by the world of insurtech where regulations are particularly complex. “On top of data privacy, organisations in insurtech also have to take into account other elements such as intellectual property rights and meeting marketability standards. Beyond meeting compliance standards, they need to ensure that the data is bias-free, to avoid skewed and unfair models and possible reputational risk,” he points out.

Insurtech industry changes

Baldin believes the entire insurance industry is experiencing a seismic shift – with data at the heart of it. “We use data to drive innovation, create competitive advantage, reduce costs, and make decisions critical to the future of any business,” he says, pointing out that the solutions managing data are the key to making it valuable.

“In fact, insurtech is based on the very premise of enabling innovation and disruption. But to do this, data needs to be readily available, in a safe, compliant, privacy-preserving manner. It needs to be easily available when needed, safe to share, and also flexible so one can shape it for the specific task at hand, whether that’s testing new risk models or collaborating with external partners. Without robust solutions, this can be an impossible task.”

Rimmer comments, "Fundamentally, other than the insurtechs' regulatory responsibility to store policyholder data, a customer can request deletion of collected data under GDPR. The customer is the ultimate arbiter of that data.

“Businesses that will extract the most value from data are those that use the data to create more customer centric insurance products that help keep our customers for longer. Big Data's value is derived from its application, not the collecting.”

He adds, “It is a corporate responsibility to be clear about what data is collected and how you use that data to benefit the supplier of that data. If you don't have a clear use for the data, then don't collect.”

3 trending data compliance issues for 2021

1) Use of data: There will be more supervisory activity, early intervention, and involvement across sectors. There is also likely to be a better focus on customer behaviour and the use of data to create better consumer outcomes.

2) More AI: Upgrades systems and AL solutions will be used to manage real-time data both for analytics, transmission and storage purposes.

3) Better data protection: Companies will build better data protection rights for customers and use Data Protection Impact Assessments to manage rights.