Separating Substance From Hype

Tech startups and data wizards are claiming to make insurance modern, agile, and friendly. Will they succeed where other tech waves have failed?
By: | August 29, 2017

Silicon Valley, the land of startups and innovation, is invading the insurance industry under the banner of “Insurtech.” Data scientists and technologists, financed by venture capitalists, are claiming to be able to dramatically transform the industry, from underwriting and claims to customer service, sales and administration.

The insurance industry has heard similar claims of a technology revolution before. From mainframes to the web, each successive technology wave claimed to be able to transform the business. But the lasting impacts of these efforts has been mixed, with many enterprise level IT projects costing millions of dollars with little to show. Add to that history the industry’s heavy regulatory burden and it’s easy to understand why many firms remain skeptical.

“Insurance is highly regulated. It has so many legal and compliance hurdles and is fraught with political risk, so the culture is naturally risk-averse,” said Ellen Carney, a Forrester analyst.

“There are systemic issues in the industry that prevent new tech-driven products and services from being widely commercialized,” she said.

Martha Notaras, partner, XL Innovate

While the term Insurtech refers to a wide range of technologies and approaches, almost all these efforts focus on the ‘connective tissue’ of insurance — data, distribution and customer service. They are not looking to actually take on risk — the very heart of the business.

And there is a lot of opportunity to improve these functions.

According to research by the InsureTech Connect conference, there was a 156-percent increase in venture capitalists investing in the insurance technology space over the past year, contributing to a total of about $5 billion in investment dollars.

Some traditional carriers have created their own investment arms specifically to back the Insurtech firms they see as promising — XL Catlin, AIG, Travelers and Munich Re among them. Many others have created in-house innovation labs to develop their own technologies.

The industry appears to be at an inflection point. Beyond the hype and buzzwords, the traditional insurance sector does see value in Insurtech’s promise to materially impact the industry across its entire value chain. Here are some ways change is beginning to take hold.

Policy Administration

The insurance industry is hundreds of years old and the business was literally built on huge stacks of paper. But paper can’t keep up with the pace of business in the modern world. On its administrative back end, insurers grapple with overhead and inefficiencies that could be minimized if processes were digitized and streamlined.

Referred to as “digitization” or “digital transformation,” this is an area where Insurtech shows great promise.

Ilya Bodner, CEO, Bold Penguin

Gathering information from clients, updating policies, incorporating industry standards, sending and receiving contracts, soliciting feedback from stakeholders, and other tasks take valuable time and resources when done manually.

Legacy policy administration systems are often siloed and do not always communicate well with other systems within an organization. Layering new technology solutions on top of them may create more IT problems than they’re worth.

Digital systems are slowly beginning to replace these dinosaurs. Delphi Technology, for example, has digitized and automated policy administration with its Accelerator workbench and Delphi Policy admin system. Ultimately, this allows insurers to get products to market faster — a key competitive advantage.

Even simpler technologies can make a big difference in helping carriers increase efficiency while cutting costs. DocuSign, for example, saves time and money by eliminating paper. This speeds up communication between insurers, brokers and clients, boosts productivity and increases efficiency by streamlining transactions.

These technologies aren’t necessarily flashy or disruptive. They simply take existing processes and give them a modern makeover.

But there are other innovations on the horizon. Blockchain, though still not well-understood by many in the industry, shows potential as a communication and compliance tool.

AIG, through a partnership with IBM and international bank Standard Chartered, recently completed a pilot phase of a multinational “smart contract” policy, managed using blockchain.

The program consisted of a master global policy supplemented by three local policies.

“People think they can throw a lot of money at technology to build a better customer experience, but that’s just the tip of the iceberg.” — Ilya Bodner, CEO, Bold Penguin

“As with any multinational coverage, there are multiple stakeholders involved, and a regulatory minefield to navigate,” said Carol Barton, president, AIG Multinational.

“There are lots of moving parts, and communication is the pain point.”

Blockchain technology allows every party to see the policy and view updates in real time. Because every party in the blockchain needs to vote on a change in order for it to be approved, there are no surprises. Using blockchain for this type of policy makes communication easier, eases friction and improves transparency.

“Transparency is something insurance buyers want more of. Blockchain has the potential to create a new level of communication, trust and transparency, which is key in this business,” Barton said.

Underwriting & Pricing

Smart homes and cars, wearables, fleet telematics, and a host of other sensors and systems connected to the Internet of Things provide the insurance industry with a stream of valuable data. Insurtech firms, untethered by legacy systems, are better positioned to capture and leverage data for incumbent carriers, or so they say.

But for the most part, data collected via the IoT, social media, and other new sources cannot directly influence commercial underwriting or pricing because regulators do not allow it. As promising as it sounds, the truth is that regulatory constraints will make it difficult for underwriters to incorporate data from new streams into their underwriting and pricing process for some time.

“The level of government regulation across the U.S. is complex. It’s 50 states and 50 rules based on line of business and coverages. As you begin to add distinctively unique coverages it becomes a challenge with individual line-of-state filings. Consequently, much of these creative ideas must be non-admitted coverages. I think it’s an area that Insurtech investors could be underestimating,” said Jamie Miller, head of property & special lines North America at Swiss Re Corporate Solutions.

Currently, the clearest way to incorporate new data streams into the underwriting process is through the development of parametric coverage, which is triggered by characteristics of an event, rather than characteristics of a loss. Swiss Re has pioneered parametric policies to respond to unpredictable events with large losses like natural catastrophes.

Big Data can help to identify new triggers, which ultimately helps to solve coverage gaps and enables faster claim payment.

Machine learning can help mine the vast expanses of data for those nuggets that will be most useful for insurers, and fine-tune the underwriting process over time so it becomes more automated.

In July of this year, Milliman completed a study with DRTS Ltd., examining “multi-criteria decision-making using an iterative process of advanced computing and human input.” It asked, in other words, can we make better decisions if human knowledge and intuition is combined with machine learning?

Milliman used DRTS Ltd.’s DACORD platform to study a complex dataset.

“DACORD contains a range of tools to look for non-linear relationships between variables and study the dynamic relationships as they vary over time,” said Neil Cantle, principal, consulting actuary, Milliman.

“The study was designed to show that a combination of humans and computing can deliver a superior result for complex problem solving than using either on their own.”

Claims and Risk Management

Where Big Data and machine learning can make a profound impact is in claims, and more specifically, claim prevention.

According to Willis Towers Watson and CB Insights’ Quarterly Insurtech Briefing Q2 2017, about 90 percent of claims management is controlled by incumbents. But Insurtech is helping incumbents leverage data and technology to streamline and automate the process.

WeGoLook, which dubs itself the “Uber of inspections providing on-demand field services,” taps into its network of 30,000 “Lookers” or field agents who can inspect a loss and upload photos through a mobile platform within hours of notification.

Backed by the resources of claims management provider Crawford & Company, it uses the crowd sourced data to speed up claim resolution and ease communication between claims managers and clients, again improving the transparency that insurance buyers increasingly demand.

Allstate similarly is developing a platform to automatically process smaller and more straightforward claims, so that notification, determination of coverage, predictive damage estimates, fraud detection, and electronic payment all happen automatically — and quickly. At the Bank of America Merrill Lynch Insurance Conference in February 2017, and as cited by the WTW and CBI report, Allstate’s president Matt Winter said payment could be delivered to clients within moments of them uploading photos.

“In 2010, we were the first domestic insurer to enable small businesses to get quotes and buy insurance online, although in recent months, established players are now stepping off the curb and coming into the same space.” — Kevin Kerridge, executive vice president of small business insurance, Hiscox

The platform will also refine its process and rules over time using machine learning.

When it comes to risk mitigation and claim prevention, machine learning can be a boon to both insurers and risk managers looking for more targeted strategies to reduce their risk.

“We aren’t always looking for companies that are creating new sources of data, but that are mining data with machine learning so they can deliver the data to insurer clients and help them understand the risk better,” said Martha Notaras, partner, XL Innovate.

Cape Analytics, an Insurtech firm backed by XL Innovate — the Insurtech arm of XL Catlin — uses machine learning to derive more accurate property data from aerial geospatial imagery, gathered by drones and satellites.

“We have previously deployed systems mining techniques on insurance portfolios to reveal the interactions between the risks being underwritten and how these vary over time. This insight can help to ensure that the risk profile of the portfolio is consistent with your expectations and that there are no underlying relationships between the risks which you need to take into account. You can also extend the dataset being studied to look for risk drivers which might help to identify new behaviors or shifts in risk profile,” Neil Cantle of Milliman said.

Examples of this include property sensors that detect moisture, so home or business owners can detect and fix a leak before water damage occurs.

Fleet telematics systems show safety managers which drivers or specific behaviors are causing losses, so they can tailor their driver training to prevent accidents. AIG’s internal innovation and technology branch partnered with the City of Atlanta and used such a system to examine a “compilation of traffic and weather patterns” and learn how to “leverage that data to drive down accident rates,” Barton said.

Customer Experience

Front-facing, mobile claims management systems have as much to do with customer experience as they do with streamlining operations for insurers and claims managers, and indeed the most visible aspect of Insurtech is its quest to revamp the front-end customer experience through streamlined websites, mobile apps, automation and self-service capabilities.

The on-demand economy has elevated expectations regarding ease of use and a more pleasant buying experience across the board. Customers want to find what they’re looking for in just a few clicks, see all of their insurance options and buy directly through a portal, making the transaction easy and fast.

Kevin Kerridge, executive vice president of small business insurance, Hiscox

“Insurtech is driven by consumer preferences. There is a demand for a different experience and a new way of buying insurance. Technology is developing to address these unmet needs,” said Puneet Kakar, partner, Monitor Deloitte.

Creating an improved experience is the bread and butter of Insurtech startups. Their websites and apps are clean and simple – stripped of industry jargon or too much detail. That’s by design.

“Part of it is not wanting to give away our secrets to competitors; part of it is because we want customers to have questions, and reach out to us with those questions,” said Ilya Bodner, CEO of Bold Penguin, whose website consists of a single page separated into vibrantly-colored blocks of information. A request to enter an email address to learn more sits at the very top.

“They are trying to make the insurance experience delightful, which is a high bar to set,” said Martha Notaras of XL Innovate.

While this is a big factor in modernizing the insurance industry, some say it’s a surface level fix.

“People think they can throw a lot of money at technology to build a better customer experience, but that’s just the tip of the iceberg,” Bodner said.

Sales & Distribution

Insurtech was born in personal lines, where the peer-to-peer sales and distribution model works best.

In this space, Lemonade leads the way.  Simplicity has proved a strong marketing and customer retention tool.

In commercial insurance, that model is more difficult because of greater risk complexity… but it may be adaptable for small businesses which bear greater resemblance to an individual buying personal insurance than to a large commercial account.

“In 2010, we were the first domestic insurer to enable small businesses to get quotes and buy insurance online, although in recent months, established players are now stepping off the curb and coming into the same space,” said Kevin Kerridge, EVP of small business insurance, Hiscox.

Hiscox delved further into that space through its partnership with Bold Penguin, which caters to brokers of small business accounts.

And while the direct-to-consumer nature of Insurtech threatens to take the broker out of the insurance transaction altogether, it seems unlikely that this will happen in the commercial space, especially for larger businesses.

“In commercial lines, there is more complicated risk assessment, risk placement, appetite, claims, adjustments, payouts. It’s more fragmented and there is less consistency,” Bodner said.

“At some point, a human being is involved. Bold Penguin’s goal is to empower the broker, not cut them out. The end goal is to use AI and machine learning to better predict the day-to-day workflow of the broker and streamline processes for them.”

That means brokers’ jobs are secure… but they can’t rest on their laurels.  Insurtech will shift the broker’s role from that of a buyer to that of a consultant or trusted advisor, guiding clients through their risk and coverage options and determining which carrier policy fits best.

“We don’t buy into the idea that agents are dead,” Kerridge said. “Their role will change, but not disappear. They’ll become digitally enabled. We’re leveraging the $250 million we’ve invested in our direct-to-consumer infrastructure to help them with that.”

Some Insurtech firms are in fact not just catering to brokers — they are the brokers.

Julie Zimmer, COO of Embroker, said Insurtech companies like hers should be treated as a separate distribution channel. Embroker collects data from insureds’ existing policies to find areas where coverage or contract terms and conditions could be improved, and finds products on the market that could fit the bill.

The Challenges Ahead

The biggest thing preventing incumbent insurers from jumping into the Insurtech pool with both feet is corporate culture. Tech startups are used to “flying by the seat of their pants” and fixing glitches as they go, Sam Friedman of Deloitte said. Insurers don’t operate that way.

It is also true that the Insurtech market remains highly fragmented, with lots of small players doing lots of different things. Several sources said it’s likely that many startups here today will be gone in 5-7 years as more consolidation and cohesion takes hold.

“There’s nothing broken about insurance,” said Jay Weintraub, Founder and CEO of NextCustomer, which operates InsureTech Connect. “It doesn’t need to be upended or replaced, but there is a desire to invest in innovation and transformation.” &

Katie Dwyer is a freelance editor and writer based out of Philadelphia. She can be reached at [email protected].

More from Risk & Insurance