UK: Algorithmic Underwriting: How To Balance Automation And The Human Touch

Algorithmic underwriting is a concept fast becoming the frontlines of cutting-edge technology for insuring risks in the London market. But what real-world implications will this have for an industry already undergoing significant change?

Algorithmic underwriting, bionic underwriting and unicorn underwriting are all emerging concepts fast becoming the frontlines of cutting edge technology for insuring risks in the London market. Given the industry's need for a reduced cost of writing business, compounded by the drive to become more digital, the growth in interest is understandable. But what real-world implications do these concepts have for an industry already undergoing significant change?

Algorithmic underwriting encompasses the use of algorithms to support digital trading, case underwriting and decision making. To focus on just one of these elements would be to leave out an integral part of the picture. An insurer implementing these techniques needs to be thinking about how to deploy them across distribution channels, different lines of business and different technical functions.

When it comes to data, the immediate consideration for insurers is often to ask how much data they can collect. Instead, the priority should be to collect the right data at the right time, and then to close the gap between data collection and decision making.

Dynamic portfolio underwriting

Algorithmic underwriting is expected to have an especially profound impact upon the way that case underwriters work. By automating much of the traditional, time-consuming and manual underwriting processes usually undertaken by underwriters, they will find themselves interacting with fewer cases to generate the same amount of business. They will also see a fundamental change in how they interact with non-automated cases and the systems they use to underwrite them.

A simple example would be an underwriter deciding which case they're going to underwrite next from a collection of new business. Through a dashboard showing 15 cases allocated to them by an algorithm, the case flagged as the top priority to work on is a company manufacturing widgets, having been matched to a portfolio segment of high strategic importance in line with the growth strategy set by the portfolio head. The case has had its submission data ingested and a triage process has passed the case through a series of sanctions, duplications and other checks. It's ready to work on.

Removing the need for an underwriter to spend hours poring over a risk submission, validating data and manually comparing this to existing policies and wording before deciding if it is one for that insurer, saves the underwriter valuable time.

While algorithmic underwriting cuts out many time-consuming parts of the process, human decisions will remain key, with only policies that meet the carrier's specific criteria presented to the experienced and knowledgeable underwriter. They will be able to focus on more complex tasks, such as pricing, managing mid-risk applications and increasing personalisation options for customers.

This transformation of portfolio underwriting will also fundamentally change how insurers work, pushing down walls that have been built around products and business lines. Instead, the portfolio concept will become more dynamic and creative, and ultimately enable more profitable portfolio decisions to drive growth.

Future implications

The London market is a unique ecosystem of face-to-face relationships, renowned for its personalised approach to business. Until recently, however, the market negotiated and agreed deals the 'old-fashioned' way, exchanging a wealth of data on physical documents. With modernisation in mind, the market has come a long way in transforming itself in order to create the world's most advanced insurance marketplace.

However, the growth of digital across the London Market can hardly be said to be meteoric. There is still a quantum leap that needs to take place to transition to a truly data-first marketplace that will make the day-to-day work of brokers and underwriters quicker and easier, and realise its potential to revolutionise the London market.

The transformation envisaged by a digital marketplace is only possible if complete, accurate and timely data is available. Greater transparency will then drive trust and open up new markets, segments and classes more quickly.

Already helping to improve the performance of today's fragmented market, WTW's 'many-to-many' digital platform, Neuron, offers a fast-track route to trade more easily and at lower cost. An end-to-end trading solution, Neuron has the capability to connect multiple broker and insurer systems together, enabling risks to flow at scale and with common data standards, across many lines of business.

Neuron offers unprecedented levels of automation, governance and transparency, giving the user complete control, flexibility and confidence in how they use Neuron alongside their own underwriting and pricing tools when conducting business from start to finish.

As the digitisation of the London market continues apace, WTW continues to create complementary and transformative technologies through our broking and consulting businesses, which we believe will be the standard for the connected ecosystems required to unlock future innovative risk solutions, trading and underwriting.

Neuron is integral to these wider investments, creating a new benchmark in speed, agility, accessibility and ease of use that will help brokers and underwriters adapt to competitive and quickly evolving insurance markets worldwide.

Today the London market is world leading, but what happens next is critical to its future. Relying solely on personal experience is no longer enough, especially in a market that boasts a treasure trove of data and insights to support decisions. Driving insight from data analysis to better plan, price and manage portfolios is fundamental to delivering sustainable competitive advantage.

To be successful, insurers will need to overcome the tensions between traditional and newer data-driven techniques. And the ability of different teams within the business to collaborate will be key to meeting this challenge. Working out how to train, retain and recruit the talent within those teams will be equally critical if insurers are to harness the full potential of algorithmic underwriting to drive future growth.

Even in the distant future, the integral role of expert underwriting is not in doubt. It is only the balance between which products can be delivered algorithmically and which ones will be delivered by human underwriting that will shift over time.

This article originally appeared in Insurance Day on January 8, 2023.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.