By Ron Glozman1
1Founder and CEO, Chisel AI
Investment in InsurTech by commercial insurance companies has long lagged behind personal lines. According to Insurance Business Magazine,1 over the past decade, InsurTechs focused on personal lines saw $5.8 billion worth of investment versus just $1.3 billion for InsurTechs focused on commercial lines. For every dollar that personal lines spent on InsurTech Innovation, commercial insurers invested a mere 22 cents.
It is no surprise that commercial insurers have some catching up to do when it comes to technology innovation. What is surprising is the extent to which artificial intelligence (AI) has become a catalyst to accelerate their digital transformation. While commercial insurance has habitually been among the last industries to embrace new technologies, AI is proving the exception to the rule.
Kevin Kelly, founding editor of Wired, coined a famous phrase that captures the flux that commercial insurance finds itself in today: ‘The future happens very slowly and then all at once.’ A perfect storm of massive volumes of unstructured data, slow and cumbersome back-office processes and years of underinvestment in technology has created a commercial insurance industry that is ripe for disruption – and also finally ready to embrace it.
In the era of Amazon, commercial insurers are grappling with the on-demand expectations of a new breed of customer and a new business paradigm based on instant gratification. Customers expect the same high level of engagement and responsiveness in their working lives as in their personal lives. While purchasing commercial insurance may never be as easy as ordering dinner on Uber Eats, commercial insurers are under pressure to provide higher levels of customer engagement, a high degree of personalization and much faster response times – and to do it all digitally.
This is where AI comes in, offering insurers the opportunity to solve age-old challenges standing in the way of customer responsiveness and growth. Nowhere is this truer than in the back office, where cumbersome, high-volume, high-touch underwriting processes create significant bottlenecks.
Commercial insurance underwriting is data-intensive. Making sense of policy-level data is critical for risk mitigation, pricing, decision-making and growth. But getting at the data that underwriters need to thoroughly assess and price risk has become more difficult with the exponential growth of enterprise data stores and data lakes.
Much of this data is trapped in digital documents of various types and formats – policies, binders, quotes, submissions, loss run reports, statements of value, etc. – and most of it is unstructured. Older technologies and approaches can’t keep pace with the ever-growing volumes of data; nor is it sustainable for insurers to outsource or hire more people to manually process and re-key it.
Advances in artificial intelligence are providing new ways to address the industry’s data dilemma. AI can help brokers and carriers interpret massive amounts of data and identify correlations that would otherwise be invisible to the human eye.
The AI revolution is being driven by InsurTechs offering solutions designed specifically for insurance. These solutions use natural language processing (NLP) and machine learning to extract, read and interpret data in digital documents the same way humans do, only hundreds of times faster.
These purpose-built solutions typically employ named-entity recognition to identify and classify insurance-specific data types, such as premiums, limits, deductibles, types of coverage, exclusions and endorsements. Rather than simply recognizing page and layout patterns like optical character recognition (OCR), AI solutions are able to comprehend the context of the data they extract with a high degree of accuracy. Put another way, these solutions understand the language of commercial insurance.
As profit margins shrink and customer expectations rise, it’s more important than ever for insurers to find new ways to automate high volume, repetitive, error-prone underwriting processes. By extracting unstructured and structured data in real time and outputting the data in insurance industry standard formats like ACORD, AI solutions allow insurers to ingest actionable information and put it to use in ratings engines, core insurance systems and analytics platforms. By allowing ready access to policy-level data, AI powers intelligent automated workflows, enabling a range of transformative use cases, such as:
Insurers can no longer afford to ignore inefficient underwriting processes. As Senior Analyst Greg Donaldson of the Aite Group writes,2 “Carriers that do not explore ways to integrate AI into their underwriting processes will fall behind their competitors.”
AI is, first and foremost, an enabling technology. According to Entrepreneur Magazine,3 a report from Harvard Business Review found that firms achieve the most significant performance improvements when humans and machines work together: “The ideal AI-human arrangement is one in which AI technology drives the lower-level, repetitive processes associated with completing a task, while human oversight ensures the timely and accurate completion of that task.”
Human underwriters are unlikely to be replaced by AI. A more likely scenario is that AI will augment underwriters with superhuman powers. By digitally transforming high-touch workflows, AI will enable underwriters to focus on more accurate risk predictions, new distribution strategies, deepening customer relationships, offering bespoke services, and growing and shaping their book of business.
With all the hype, it’s worth remembering that we are still in the early days of AI. The initial pilots have come and gone, and we are now seeing production deployments, but there is still work to be done to scale up and scale out AI across geographies and lines of business. The types of results that commercial insurers have seen so far include:
In his book Behind the Cloud,4 Marc Benioff, founder of Salesforce, writes that, “In all industries, especially the technology industry, people overestimate what you can do in one year, and they underestimate what you can do in ten.”
It is safe to say that in ten years’ time commercial insurance is going to look a lot more like personal insurance. Brokers and their customers will no longer have to wait days or even weeks for an answer to their quote request, and carriers won’t lose business because back-office bottlenecks prevented them from responding fast enough. Underwriters – freed from mundane manual tasks – will spend more time listening to and responding to customers and focusing on high value work. And AI will have made this digital transformation a reality.