A perfect storm of technological, economic and societal forces is accelerating the insurance industry’s investment in the digital and data science agenda to drive better product structuring, price differentiation, claims management, customer engagement and risk profiling.
A new paper from Capco, ‘Foundations for Getting Ahead and Staying Ahead, An Insurance Industry View’, discusses the operational, legal and ethical risks associated with the latest data developments.
Insurers have been methodically analyzing data for decades, if not centuries, so that governments, businesses and individuals have the confidence to make investments and accumulate wealth by mitigating and pooling risks. Three factors have emerged that threaten the status quo:
- a tidal wave of structured and unstructured data generated through digitalization of the economy and wider society
- developments in techniques and technologies to scale the application of machine learning and artificial intelligence approaches to actuarial analysis
- accelerated globalization, simultaneously lowering international barriers to entry while rendering supply chains more susceptible to increased transmission of global systemic risk across borders – as seen with COVID-19.
Against a backdrop of claims inflation, increasing catastrophe loads, and market entrants from the insurtech sector unencumbered by legacy infrastructure, agility is key. The paper examines the role of data management in the industry and how a lean and agile approach reduces the burden of data curation and preparation while managing risks and maintaining stakeholder buy-in.
Organizations that build lean and agile data management foundations can perform data science and digital innovation at greater scale, effectiveness and with quicker returns on investment. This provides immediate competitive advantage through enhanced business insights and customer engagement, and the ability to sustain these advantages in the long run through uplifts in the data culture of the organization.
It is becoming increasingly critical that machine learning (ML) and artificial intelligence (AI) implementations are not only fair and beneficent, but also transparent and explainable so outcomes can be ethically audited. Insurers must implement controls to review ethical risks and mitigations related to intended outcomes before solutions are deployed, and mechanisms to enable the lid to be lifted on how ML/AI engines process data to generate the outcomes and the logic underpinning those outcomes.
Alvin Tan, principal consultant at Capco and author of the paper, comments:
“Data is the lifeblood of a modern insurance business. In the hunt for rate adequacy, product innovation, and market differentiation, insurers are rapidly increasing investment in data science and digital platforms to leverage competitive advantage from the vast quantities of data potentially accessible.
“Data is at the heart of an insurer’s ability to remain competitive in the long run. It must be managed as an asset to ensure that management of implementation, operational, legal and ethical risks do not overly ‘tax’ the organization, and to build the foundation of a strong data culture.
“To get it right requires a sustained investment not only in data science labs and consolidation of data architectures, but also in the organization, the people and the data processes involved.”