It’s clear developments in artificial intelligence (AI), machine learning and other innovative technologies will have an impact on nearly every industry — including insurance and financial services. But what areas will be affected most in the near future, and how will they be affected? These are questions we explored with nearly 100 industry leaders during Denim Summit 2017 in Des Moines. When we posed the question, “What insurance process will be most affected by AI?” in a live poll, “underwriting” won with 39% of the responses. Following closely were “marketing and distribution” with 32% and “customer service” with 26%. Clearly, it’s not just one area that will be affected. Perhaps a better question is, “How will each area be affected?” Let’s take a look. Underwriting As author, speaker and futurist Blake Morgan writes in Forbes, AI has the potential to automate the entire underwriting process. Imagine a world in which bots scan a consumer’s social and digital profile to gather information and find trends and patterns. Someone who has a healthy lifestyle and steady job may be less likely to get into car accidents or rack up medical bills, which could lower insurance premiums. “AI can analyze data better than humans to more accurately predict each customer’s risk, thereby providing customers with the right amount of insurance and companies with protection from risky customers,” Morgan writes. See also: Strategist’s Guide to Artificial Intelligence   Marketing and distribution Hyper-personalization is the new norm in marketing and distribution. Brands are becoming dramatically more attuned to the needs and priorities of consumers and increasingly shaping their product offerings around rising lifestyle trends. Traditional blanket methods like cold calling no longer cut it in today’s uber-connected, digital age. AI can pull in consumer data to create a full profile that can be used to offer only relevant insurance products and remember a consumer’s preferences. Customer service According to a study by Oracle, nearly eight out of 10 businesses have already implemented or are planning to adopt AI as a customer service solution by 2020. There are two primary ways organizations are augmenting their customer service experiences with AI:

• Front-end, AI-powered bots, or conversational computer programs that interact directly with a customer without human interaction.
• AI-assisted human agents, or human customer service representatives who are supported by AI technology.

For at least the foreseeable future, chatbots won’t replace humans in customer service centers. They will, however, replace some of the tasks traditionally handled by people and, ultimately, enhance the experience for consumers. Customer service and experience expert Shep Hyken shares four reasons AI and chatbots are improving customer service in big ways:

• Chatbots never sleep.
• Chatbots won’t make you wait.
• Chatbots personalize the customer experience.
• Chatbots make friends and build relationships.

See also: Group Insurance: No Longer Overlooked   While AI’s value proposition may be clearer in some areas than others, it’s not hard to imagine a future in which nearly everything we do — in both business and in life — is somehow affected by AI.

Insurers move cautiously when embracing automation and other tech tools, and for good reason. As technology changes the insurance industry, challenges arise that appear in few other verticals.

Nonetheless, business automation is set to change nearly every industry, Deloitte argues, and insurance companies can certainly benefit from the efficiencies that automation will introduce. Data collection, analysis and decision-making, once the sole domain of humans, can now be improved by automation and AI. These tools range from simple time-savers, like auto-completion during data entry, to complex pattern recognition and data mining, which is transforming the way we analyze risk.

Further, automation offers numerous opportunities to improve efficiency, retain customers and reduce errors — but only when automation is enacted thoughtfully, as Corrine Jones notes at Property Casualty 360. Here, we look at five obstacles to automating agency operations, plus ways to overcome them.

1. Departments Aren’t on the Same (Digital) Page

Most insurers’ systems have difficulty talking to one another. Property and casualty insurance companies tend to operate like federations: departments that fall under the same umbrella, but that operate independently most of the time. This means their customer data gets hidden away in silos, and data-driven intel therefore cannot be shared among departments.

Insurers that continue to federalize this way miss key connections that can lead to improved coverage and more satisfied customers, says Dan Reynolds, editor in chief of Risk & Insurance. Breaking down silos can be a daunting task, but the reward can be well worth the effort — and automation can help.

When data can flow across departments, machine-learning algorithms can perform analyses across departmental lines. This lets the technology spot patterns and recommend solutions more easily, says Forbes contributor Bernard Marr. With access to a single cohesive system and its data, machine learning algorithms can handle a wide range of tasks, from spotting potential fraud to providing an interactive FAQ for customers.

2. Current IT Infrastructure Might Not Support an Ambitious Implementation

McKinsey partners Tanguy Catlin, Johannes-Tobias Lorenz, and Shannon Varney stress that one of the big lessons the insurance industry learned in 2017 was that tech-driven strategies aren’t a goal in and of themselves. Rather, executives need to think about what strategies make sense in a tech-driven world.

As such, organizations must ensure the capabilities of their IT teams are keeping pace with plans to implement automation. IT cannot take on a supporting role when implementing automation technology. IT must help lead the implementation, the McKinsey partners argue, and it’s up to the organization to position IT in a leadership role.

See also: How to Solve the Data Problem  

Here is how company leaders can position IT teams to assume that role:

• Hire tech leaders. IT teams leading changes need project leaders, agility coaches and scrum masters to guide their work.
• Promote a tech-friendly environment. Demand for tech talent is quickly outstripping supply in many industries, and insurance companies today must establish “an environment that attracts talent, promotes personal growth and offers a desirable and interconnected work environment and flexibility.”

3. Existing Processes Might Not Scale Quickly Enough

Many tech professionals who focus on insurance solutions, like EZLynx project manager Derek Armentrout, caution insurers to “start small” when considering the switch to automation. Starting small can benefit some companies.

But a small start can derail an entire automation project when “small” isn’t combined with “scalable.” Not only must the system be able to grow into the existing insurance company structure, but it must also be able to grow with that structure as the company expands. It must handle not only additional users but also more intensive calculations, recommends Richard Seroter.

Prasad Jogalekar and Murray Woodside in the IEEE Transactions on Parallel and Distributed Systems, provide a definition of scalability that is particularly apt for insurers: “Scalability means not just the ability to operate, but to operate efficiently and with adequate quality of service, over the given range of configurations.” A system that fails customers when overloaded is not scaling adequately to meet either the insurer’s or the customers’ needs.

As Seroter notes, working with a Software as a Service (SaaS) provider is one way to ensure scalability that meets both insurer and customer demand. That means choosing a provider that understands the connection among scalable platforms, automated activities and customer experience to maximize the value of automation in customer retention.

4. Workforce Obsolescence

McKinsey principal Sylvain Johansson and senior expert Ulrike Vogelgesang predict that automation will render 25% of all insurance industry jobs obsolete by 2025. Operations were hardest hit, with a 13% predicted drop in human employees, caused largely by automating everything from report generation to answering customer queries.

“That’s neither a negligible amount of job loss nor an unimaginably distant time frame,” Johansson and Vogelgesang wrote. “On the contrary, given the magnitude of these changes and the looming future, it’s important that insurers begin to rethink their priorities right now.”

Among the rethinking steps the McKinsey report recommends are:

• Retraining existing staff,
• Identifying imminent skills gaps and hiring to fill them, and
• Crafting employment value propositions that reflect a tech-heavy world.

Despite McKinsey’s predictions, the insurance industry will need to retain human workers for a number of key positions, Sabah Karimi writes at Great Insurance Jobs. Digital analysts, online marketers and other tech-minded positions will still demand the human touch. McKinsey explains that some insurance jobs are relatively safe from automation for the time being: Actuaries, for instance, are unlikely to see their jobs automated in the near future.

5. Existing Interfaces That Fail to Attract, Inspire and Retain Customers

Customer loyalty to their property and casualty insurer is a unique relationship. Because customers rarely interact with their insurers except in a crisis, building a relationship over time poses particular challenges.

See also: 3 Keys to Success for Automation  

Raising the difficulty level is the fact that today’s customers expect their product and service purchases to be easier than ever before. Web-based business has created an expectation of a seamless omni-channel experience and instantaneous results.

How can automation help?

• Improving self-service. Increasingly, customers who use the Internet to contact businesses do so with the expectation of self-service, Steve Wiser writes in an article at P&C 360. Automated systems streamline the collection of customer data. When incorporated with machine learning, they can automatically recommend the best additional coverage or next steps for the user.
• Better analytics. In the age of big data, Wiser notes, insurance companies that don’t gather and analyze customer information are missing an extraordinary opportunity — not only to manage their own risk, but to better connect with customers, as well. A personalized customer experience boosts customer ownership, and it’s a process that can be automated with the right tools.
• Improved ownership by packaging product lines. When Allstate first tried to switch to a commercial offering, the company found itself stalled by agents who needed to search out information before offering recommendations to customers — and a system that turned this process into a major stall, Kumba Senaar says. An automated system responded to these information requests more quickly, intuited what agents would need next and recommended additional coverages based on available data.

The result? Happier customers, larger purchases and more efficient agents. A win-win(-win) for Allstate.

The insurance industry has a long history of reclassifying “obstacles” as “opportunities.” When insurers partner with SaaS providers, they gain an ally that understands the connections between these major challenges and that can implement systems that address multiple challenges simultaneously.

Can you imagine what the insurance industry would be like without the internet? Many of us remember a time before email and online price comparisons, but we can no longer picture doing our jobs without the web. Hands down, the internet created a fundamental shift in how business gets done in the insurance industry. The Big Data Revolution There’s another revolution on the way in the form of big data. In fact, nine out of 10 companies told Accenture that big data will change how they conduct business on a scale comparable to how the internet changed the world from the 1990s on. Think about that—less than 20 years after the internet upended business, another movement is coming along that enterprises say will have a similar impact. What a revolution, right? Big data is already shaking things up in a big way. As a result, insurance organizations are racing competitors to employ analytical techniques and develop predictive models that will give them a competitive edge in everything from pricing risk and detecting claims fraud to developing products. But as companies develop these tools, many overlook the significant role that frontline employees will play in optimizing and adopting new big data techniques. Companies put as much as 95% of their big data budgets toward employing analytic techniques and developing models, ignoring employee training and knowledge building in the process. See also: What Does ‘Data-Driven’ Really Mean?   A Crucial Role for the Front Lines Frontline employees and managers working in traditional insurance fields like claims and underwriting will play a crucial role in working with data scientists to develop and implement effective solutions based on big data. Data scientists are not insurance experts and don’t necessarily have a firm grasp on how the industry operates. Without key intelligence from employees who know the insurance world, even the most elegant predictive modeling tools won’t have sufficient business impact. If data analytics is going to have a positive impact on insurers’ bottom lines, it stands to reason that a basic understanding of it will be good for your career, too. Insurance organizations across the country are forming teams to figure out the best way to integrate big data into their daily operations. Employees with big data know-how who can act as connectors between data scientists and traditional insurance operations will quickly find themselves in high demand. So just how can frontline employees and managers become data-driven decision makers? By following these three steps: 1. Focus on data literacy Data literacy means getting a handle on the terms and concepts behind data science and how they’re being used in the insurance world. So even if you can’t segment data into a classification tree, you should at least understand what a classification tree is and the basics of how it works. Big data is a fast-growing field with a lot of lingo and jargon. Check in with data scientists at your company to get a better idea of how your organization talks about data and where it is prioritizing using new data collection and analytical techniques. Focus your big data learning in these areas. 2. Sharpen your data mindset Once you understand what’s possible with data science and where your company wants to go, start thinking in terms of big data. When a problem comes up, think about what data you have at your disposal and how it could be analyzed to solve that problem. For example, can the data be analyzed in a new way to create different insights? Should additional data be collected from government agencies, customers, agents or adjusters? If some data point isn’t at your disposal but would help you do your job faster or better, chances are your boss wants to hear about it and your company’s data experts can help find a way to pull it. 3. Hone your data skillset After you’ve got the lingo down and think of data first when trying to solve problems, the third step is to position yourself as a data steward who can bridge the gap between data scientists and the insurance processes they’re working to optimize. You can still leave most of the modeling (and the crazy math!) to the data scientists, but you should understand big data to the point where you can advocate for specific analytical strategies and point out when processes can be improved. If you can fill this role, you’ll be an indispensable resource to your department and your organization as a whole. See also: Digital Playbooks for Insurers (Part 4)   This progression of developing a data mindset can be found on an organizational level as well. As big data expands and touches on more aspects of operations, entire organizations will have to think in terms of data analytics. Individuals who can develop their skills to satisfy their organization's growing appetite for data-based solutions will help those organizations find and implement strategic improvements while also growing their own expertise in the process. Interested in becoming a data-driven decision maker? Learn more about the Associate in Insurance in Data Analytics.

In many quarters, the above title could be fighting words! Because I can’t even watch an Olympic boxing match, much less an all-out fight, let me explain. For as long as I have been in the insurance industry, life and annuity insurers have been thought of as being significantly behind P&C insurers in terms of technology adoption and innovation. L&A insurers hung on to “build, not buy” strategies long after P&C insurers were advancing “buy” strategies. Technology providers with potentially cross-segment capabilities frequently did not even have a road map for selling to L&A insurers because, in their view, the front door was nailed shut! When STP (straight-through processing) started to become table stakes in personal lines operations, many life insurers seemed to feel that STP was simply a good motor oil. Every application for life insurance needed manual review. Let me enthusiastically state that those days are gone. L&A has most certainly caught up – and sometimes surpassed P&C – on many fronts. One of the measures of being “smart” is learning from prior mistakes. L&A insurers have had opportunities to learn, and not necessarily from their own mistakes. They have learned from P&C insurer mistakes, as well. See also: How to Insure the Gig Economy   SMA recently issued two research documents based on L&A insurer surveys:

• 2018 L&A BUSINESS AND TECHNOLOGY INVESTMENTS: Balancing Producer and Customer Requirements
• 2018 STRATEGIC INITIATIVES: Life & Annuities

Many exciting insights are revealed through the survey data. Not the least of which is that, in 2018, 43% of L&A insurers indicate they are transforming. Eight short years ago, only 13% indicated this to be the case. While this blog cannot hope to recap all the findings contained in the two SMA research papers, what did jump out quickly was the differences between how L&A has approached some things versus P&C – lessons learned:

• Digital isn’t all about fancy front ends and apps. When it became apparent that insurers needed to respond to the reality of a digital world, many P&C insurers ran headlong into introducing apps – most usually first notice of loss (FNOL) apps. Click to pay with a credit card on websites was another common feature. More examples could be cited. But to cut to the chase, the problem was that these digital capabilities stopped right at corporate walls, dropping into legacy technology and manual processes. They ceased to be digital. P&C insurers learned the hard way – through disconnected customer experience – that core modernization was necessary! L&A insurers have seemingly learned that lesson, with 55% having policy admin projects in 2018 – the No. 1 project overall.
• Love the hand that feeds you. The mantra across most all insurer segments, and strongly for P&C insurers, is customer experience. “We Love Our Customers” T-shirts are on every desk. This is absolutely critical, but for many P&C insurers this focus went to the exclusion of distributors. Agent and broker technology fell to the bottom of the top priority lists at many insurers. Given that agents and brokers have not disappeared, and, in fact, are critical as advisers for many consumers, this created a gap. L&A insurers do want to show some “love” to the distributors who play a critical role in customer acquisition and service. 55% of L&A insurers are executing distributor portal projects for both sales/submissions and service.
• It’s not all about BI. SMA research shows a historical trend among P&C insurers to invest in BI technology. In fact, in relation to other components of data and analytics such as dashboards, data and text mining and predictive analytics, 71% of P&C insurers indicate they are advanced users of BI tools. This is certainly good, but for many years P&C insurers have invested in BI and have not invested to the same degree – or at all – in other capabilities, which stalls advanced execution in this area. L&A insurers are investing in BI, as well, but, in 2018, 22% are investing in behavioral analytics, big data and AI. Getting into the game in these advanced areas is imperative, and L&A insurers understand that.

And there is one lesson that L&A insurers have learned from themselves:

• Building it yourself is a long and painful road. Over time, L&A insurers have attempted to tweak internally developed technology to step up to new market requirements. Given the rapidly shifting technology landscape, most insurers are not positioned to keep up, both from an IT capacity perspective and in terms of general skills levels. When it comes to emerging technology, 43% of L&A insurers are partnering with others that have emerging technology solutions. Only 29% continue to leverage their own capabilities.

See also: 3 Ways to Keep Training Fresh   Many exciting things are happening at L&A insurers in 2018. Both of the SMA research reports, which can be found here and here, provide insight into strategic initiatives and projects. Clearly, there are opportunity areas that are challenges, but there is little evidence that L&A insurers are content to support the status quo. Smart L&A insurers are looking over the fence to see what they can learn from P&C insurers. Over time, the opposite may be the trend!

We are seeing a flurry of advances in the insurtech space, be it product innovations, reimagined service experience or reduced premiums for customers. A question I often get asked is why personal lines in insurance is blazing ahead of commercial lines when it comes to innovation. The easy answer is to just follow the money, specifically the funding trail. Venture capitalists whose metric for early-stage startups is growth have rushed to personal lines as it is easier to show the volumes. Personal lines insurtech startups have focused on the distribution side of the problem – lowering the premium to increase the volume of transactions. Their lever for this rapid growth is a slick UI and a digital broker; betting on increasing throughput, consequently the adoption. In the subsequent rounds of funding, when the motive of the investors shifts from growth to profit, insurtech companies will realize that distribution is only one part of the equation, and not the core of the problem. Insurtech companies are amazing, and they all solve a part of the problem. However, to solve systemic problems, companies need to attack improvements in the loss ratio (i.e. the product problem, not the distribution problem). More than the profitability of the insurer, the ripple effect across the insurance industry and other adjacent industries is massive (for example, think of the impact on workplace safety as opposed to underwriting workers’ compensation). So, for systemic industrial change, I think the commercial industry is better-placed than personal, even though it will take longer. What Does the Anatomy of a Commercial Insurtech Unicorn Look Like? Like all quick analyses, I look at this in two dimensions:

• The opportunity
• The execution needed  to deliver on the opportunity

Both of these point to commercial as a better option. Opportunity Driven by Sharing Economy The sharing economy is drastically reducing asset ownership, with car ownership in urban areas the most-cited example. This is a loss for personal auto and a gain for commercial auto (the car is going to be a computer, and cyber risk from the manufacturer will likely become the highest coverage). This trend exists in other areas, including home ownership, renting of equipment, physical storage, cloud computing etc., but it is not talked about as much. The second shift I see happening is a fundamental change in the product structure from static to dynamic. Across all lines, the change in what you need to know upfront and what you will know throughout the life of the policy will change. The usage-based policy (sort of pioneered in parts in personal auto) will start to become the norm in commercial, despite having only a minuscule footprint currently (remember, these are exponential changes, and the initial doublings are not noticeable – think of the 0.01 megapixel camera becoming a 0.02 megapixel camera). See also: 3 C’s for Commercial Brokers in 2018   Executed With IoE and Machine Learning Let’s have a look at how you can execute on these trends: First, the current 1.5 to two touch points a year with the carrier become 365 touch points at least. The key touches in this sense are not human touches but data-driven touches. Both the upfront and post-bind data, the certainty and access of data on commercial is better, with access to personal lines data prone to consent due to privacy reasons (at least until DNA sequencers take privacy out of the equation). Meanwhile, in commercial, even if you were to replicate the existing forms (which you should NOT!) you can probably find 50% to 60% of the data -- general company, financials, locations and parts of workers’ compensation, commercial auto, general liability and the directors and officers -- to be as little as their names and addresses. However, under a usage-based policy, even knowing 100% of the upfront static data is not enough; it is the dynamic IoE (Internet of Everything) data that shifts the paradigm. These IoE solutions that I talk about have already reached a level of maturity in industries such as mining, manufacturing and construction. They have been deployed in cutting machines, heating/cooling equipment, cranes, thermal cameras, traditional cameras, forklifts, trains and guided vehicles for years. This has enabled a level of sophistication in IoE solutions, which has data from running mission-critical systems (PLCs, data loggers, historians, etc.) But why would a manufacturer or a construction company give a carrier this data? Come to think of it, the true financial incentives to increase safety and decrease risk have never existed before! This has to come in to create a win-win scenario between the insured, its employees and the carrier. Despite the commercial insurtech not taking as much premium upfront, it will get to unlock many other opportunities, simply due to the data and touch points it has. As you may have realized by now, other than driving loss prevention, what a commercial insurtech really does is switch the insurance from someone/something like the insured (“broad risk pools”) to someone exactly you (i.e. “pool of one”). One can argue this can be achieved on the wellness side with device data, but the industrial automation data has been collected and proven across many industries for 20 years now. The wellness data is just starting. Disintermediation – Stating The Obvious So far, we have got to the shape of this active, personal commercial insurtech unicorn. However, it would be remiss of me to not briefly talk about its distribution structure. A traditional carrier spends around 30%-plus of direct written premium (DWP) between expenses and commissions to “touch” an insured 1.5 to two times a year. Now, if you want to be able to continuously “touch” an insured, both the acquisition, retention and renewal structure has to be re-imagined bottom up for it to scale. One thing is for sure that in a world of IoE and machines, “human” intervention is minimal; people simply will not be able to handle the volumes and variety of data. So, there is no chance a commercial insurtech unicorn will be intermediated. None of this is just gleeful optimism; I will admit to there being regulatory hurdles. Despite having regulatory “sandboxes” setup, it is a massive step up for traditional regulators who are grounded in easy-to-regulate forms and structured data to switch to on-the-fly decisions, price adjustments made by machine learning algorithms and data flowing from the IoE devices. I see the legal and regulatory skills needed to maneuver the commercial insurtech company to being a unicorn to be as big, if not bigger, than the technology and algorithmic skills. This cannot be underestimated. My hope here is that ultimately any regulatory body remembers who they are regulating for: the insured. See also: New Era of Commercial Insurance   To Sum It Up You can see an outline of what a potential commercial insurtech unicorn would look like. Instead of being reactive, impersonal and intermediated, the successful company will likely target loss ratio improvement with active, personal service, powered by a large network of data partners, commercial IoE partners and machine learning partners. To operate at a global scale, this unicorn will have to have low cost per digital touch, and hence it will likely be disintermediated. There is already a large (and growing) opportunity for an insurtech to target major commercial segments in commercial packages, commercial auto and workers’ compensation. The solution options are massive, but the problem space is even bigger. As a word of caution, it isn’t just about technology here; the ability to carefully guide the company through the many regulatory hurdles is also essential. I look forward to seeing the first commercial insurtech unicorn. I wonder who it will be?

Around two years ago, the term "chatbot" shot into our vocabulary and onto the agendas of CIOs and CMOs everywhere. The idea that a customer could simply "chat" with a robot any time, anywhere made so much sense — or did it? Any technology solution or product implemented without a clear problem in mind is just wasteful. And it is this lack of planning that put chatbots on a fast path to nowhere in many companies. Two and a half years ago, there were only a handful of chatbot providers. A year ago, there were thousands. Any remotely adept coder could whip a bot together in a few hours and, surprise, surprise, VCs went in hot pursuit of companies to fund. Fast forward to today, and we’re constantly hearing the phrases “our chatbot proof of concept was not what we hoped,” or “we tried chatbots, and they didn’t work” rolling off the tongues of those same CIOs and CMOs. But we’re not surprised. In fact, we welcome the demise of pointless technology. When we last checked, Facebook Messenger had more than 100,000 chatbots. Many of them are failing to impress, leaving users underwhelmed and frustrated. See also: Chatbots and the Future of Interaction   Automation needs a purpose So, is this the end of chatbots? It certainly is the end of companies creating chatbots for the sake of having a chatbot. But it is the beginning of a major technology shift, a quasi-revolution called AI-based automation, and chatbots certainly have an important role to play. Companies waste resources when they implement new technologies without first establishing an actual problem to solve. The same theory applies to automation, AI and chatbots. For chatbots to survive, they have to solve a business problem. Period. Executives must clearly define this problem and distill it into real use cases that have true ROI or Net Promoter Score implications — meaningful implications. As soon as a team clearly maps out the use cases, the case for automation comes next. Can the company solve this problem by removing the human element in the back end? If so, there will undoubtedly be a cost benefit to the company. A smart design here will allow for escalation to human agent in the (let's hope) shrinking contact center. Once the higher-ups give automation the green light, the company must spin up myriad other technologies to create an effective system that solves the problem in the long term. As an example, if the business problem were around customer service and the use case were automating bill pay, then payment gateways, an asynchronous messaging channel, an authentication system, encryption and privacy layer, feedback loop, API bridge into the billing system and others would need to work in unison to provide a complete solution. Rethinking the word ‘chatbot’ You’re now probably wondering where the chatbot comes in. Well, therein lies the point of this article: A chatbot only has a role to play if it delivers utility to the customer. In the case of bill pay, the visual experience the bot presents to a consumer is in the form of a chat. Developers program this conversation inside the chatbot using either decision trees or natural language understanding. See also: How Chatbots Change Open Enrollment   If I had one wish for this industry, it would be that we get rid of the term “chatbot” and instead call this user interface built around conversations a CI, or conversational interface. CIs done properly, with a true business problem in mind, will reach deep into the back end through a persistent and secure messaging channel, allowing the customer to do business — any time, anywhere and, most importantly, happily.

Our survey of leading carriers shows that insurers are increasingly looking to integrate data scientists into their organizations. This is one of the most compelling and natural opportunities within the analytics function. This document provides a summary of our observations on what insurers’ analytics function will look like in the future, the challenges carriers are currently facing to make this transition and how they can address them. We base our observations on our experience serving a large portion of U.S. carriers. We supplemented our findings through conversations with executives at a representative sample of these carriers, including life, commercial P&C, health and specialty risk. We also specifically address the issue of recruitment and retention of data scientists within the confines of the traditional insurance company structure. The roles of actuaries and data scientists will be very different in 2030 than they are today Actuaries have traditionally been responsible for defining risk classes and setting premiums. Recently, data scientists have started getting involved in building predictive analytics models for underwriting, in place of traditional intrusive procedures such as blood tests. By 2030, automated underwriting will become the norm, and new sources of data may be incorporated into underwriting. Mortality prediction will become ever more accurate, leading to more granular (possibly at individual level) premium setting. Data scientists will likely be in charge of assessing mortality risks, while actuaries will be the ones setting premiums, or “putting a price tag on risk” – the very definition of what actuaries do. Risk and capital management requires extensive knowledge of the insurance business and risks, and the ability to model the company’s products and balance sheet under various economic scenarios and policyholder assumptions. Actuaries’ deep understanding and skills in these areas will make them indispensable. We do not expect this to change in the future, but by 2030, data scientists will likely play an increased role in setting assumptions underlying the risk and capital models. These assumptions will likely become more granular, based more on real-time data, and more plausible. Actuaries have traditionally been responsible for performing experience studies and updating assumptions for in-force business. The data used for the experience studies are based on structured data in the admin system. Assumptions are typically set at a high level, varying by a few variables. By 2030, we expect data scientists to play a leading role, and incorporate non-traditional data source such as call center or wearable devices to analyze and manage the business. Assumptions will be set at a more granular level – instead of a 2% overall lapse rate, new assumptions will identify which 2% of the policies are most likely to lapse. See also: Wave of Change About to Hit Life Insurers Actuaries are currently entirely responsible for development and certification of reserves per regulatory and accounting guidelines, and we expect signing off on reserves to remain the remit of actuaries. Data scientists will likely have an increased role in certain aspects of the reserving process, such as assumptions setting. Some factor-based reserves such as IBNR may also increasingly be established based on data-driven and sophisticated techniques, which data scientists will likely play a role in. Comparing actuarial and data science skills Although actuaries and data scientists share many skills, there are distinct differences between their competencies and working approaches. PwC sees three main ways to accelerate integration and improve combined value 1. Define and implement a combined operating model. Clearly defining where data scientists fit within your organizational structure and how they will interact with actuaries and other key functions will reduce friction with traditional roles, enhance change management and enable clearer delineation of duties. In our view, developing a combined analytics center of excellence is the most effective structure to maximize analytics’ value. 2. Develop a career path and hiring strategy for data scientists. The demand for advanced analytical capabilities currently far eclipses the supply of available data scientists. Having a clearly defined career path is the only way for carriers to attract and retain top data science (and actuarial) talent in an industry that is considered less cutting-edge than many others. Carriers should consider the potential structure of their future workforce, where to locate the analytics function to ensure adequate talent is locally available and how to establish remote working arrangements. 3. Encourage cross-training and cross-pollination of skills. As big data continues to drive change in the industry, actuaries and data scientists will need to step into each others’ shoes to keep pace with analytical demands. Enabling knowledge sharing will reduce dependency on certain key individuals and allow insurers to better pivot toward analytical needs. It is essential that senior leadership make appropriate training and knowledge-sharing resources available to the analytics function. Options for integrating data scientists Depending on the type of carrier, there are three main approaches for integrating data scientists into the operating model. Talent acquisition: Growing data science acumen Data science talent acquisition strategies are top of mind at the carriers with whom we spoke. See also: Digital Playbooks for Insurers (Part 3)   Data science career path challenges The following can help carriers overcome common data science career path challenges. Case study: Integration of data science and actuarial skills PwC integrated data science skills into actuarial in-force analytics for a leading life insurer so the company could gain significant analytical value and generate meaningful insights. Issue This insurer had a relatively new variable annuity line without much long-term experience gauging its risk. Uncertainty about excess withdrawals and rise in future surrender rates had major implications for the company’s reserve requirements and strategic product decisions. Traditional actuarial modeling approaches were limited to six to 12 months of confidence at a high level, with only a few variables. They were not adequate for major changes in the economy or policyholder behavior at a more granular level. Solution After engaging PwC’s support, in-force analytics expanded to use data science skills such as statistical and simulation modeling to explore possible outcomes across a wide range of economic, strategic and behavioral scenarios at the individual household-level. Examples of data science solutions include:

• Applying various machine learning algorithms to 10 years of policyholder data to better identify most predictive variables.
• Using statistical matching techniques to enrich the client data with various external datasets and thereby create an accurate household-level view.
• Developing a simulation model to simulate policyholder behavior in a competitive environment as a sandbox to run scenario analysis over a 30-year period.

Benefit The enriched data factored in non-traditional information, such as household employment status, expenses, health status and assets. The integrated model that simulated policyholder behavior allowed for more informed estimates of withdrawals, surrenders and annuitizations. Modeling “what if” scenarios helped in reducing the liquidity risk stemming from uncertainty regarding excess withdrawals and increase in surrender rates. All of these allowed the client to better manage its in-force, reserve requirements and strategic product decisions. This report was written by Anand Rao, Pia Ramchandani, Shaio-Tien Pan, Rich de Haan, Mark Jones and Graham Hall. You can download the full report here.

I don’t think any of us would dispute that the insurance industry is facing an inflection point, with changing market conditions, emerging technologies and startup insurtech companies. In an environment in which both personal and commercial lines of business are affected by a digitally empowered consumer, many insurers are rethinking their IT infrastructures, distribution networks and communication strategies. Yet recent coverage in the media paints a different picture for smaller insurers trying to achieve transformative change.

I was struck by a comment made by Tom Benton, VP of research and consulting at Novarica, in an article called “Smaller insurers lean on partners to navigate disruption.” Benton argues that, while all insurers continue to struggle with limited IT resources, capabilities and access to specialized skills while facing increased demand for operating efficiency, smaller insurers are at a disadvantage.

“Most insurers are focused on three things,” he said, “running IT for the organization, projects that help the organization grow and transformational projects. Most small carriers don’t have the budget or resources (including talent) to apply to transformative projects.”

See also: How Small Insurers Can Grow  

While it’s true that small property and casualty and commercial workers’ comp carriers, municipal risk pools, captives and self-insured groups may be vulnerable to more rigid budgetary concerns than their larger Tier 1 and Tier 2 counterparts, I’m not convinced that transformation is unattainable to them.

Although smaller carriers may not have the desired resources, they can move faster than larger organizations on certain initiatives. Smaller carriers don't have to jump through all the organizational hoops usually present in a larger company. Plus, smaller carriers usually have a culture that embodies taking risks, getting faster approvals and moving into a pilot much quicker than larger insurers.

Let’s look at the agility of a smaller carrier and add the notion that these employees tend to “wear many hats” (often running IT operations while functioning in another capacity within the organization). Here, choosing the right technology partner is critical, and long-term issues must be considered when making decisions on platform, systems and applications.

For example, Maine School Management Association (MSMA), a state-wide non-profit federation that administers various insurance programs to the state’s school systems, replaced a decades-old process that involved spreadsheets and manual entry, with cloud-based insurance management software. The decision, made to provide secure and efficient online renewals of property and casualty (P&C) coverage for its 98 member school districts, is transforming the entire renewal process, reducing renewal process time and streamlining members’ ability to respond. With just 23 employees, MSMA is an example of an insurance organization that has achieved transformational change due to its commitment to successful risk-taking, a calculated plan to work exclusively with “best in class” vendors that specialize in serving public entities and a culture that is committed to innovation-fueled growth.

Even less successful experiences serve to inform future operations. We aren’t perfect, and firms that proclaim, “not us,” or “we won’t have those issues,” are either disingenuous or naïve.

See also: Have Insurers Lost Track of Purpose?  

The call for transformational improvements is upon us, with pressure to innovate using technologies such as cognitive computing tools, machine learning, predictive analytics, robotics processing automation, chatbots and natural language processing. Rather than be at a disadvantage, smaller insurers are embracing a new level of confidence that maintains that transformation is not only possible, it’s realistically attainable.

When insurtech sprang to prominence in 2015, most startups focused on personal lines disruption. Our August 2016 infographic showed that 75% of insurtechs were targeting personal lines and that 56% were focusing on distribution. Most corporate and specialty insurers concluded that insurtech presented no threat and only limited opportunity and continued with business as usual. That was then, and now is now. Insurtech now matters for corporate and specialty insurers. (Incidentally, we agree with the point Adrian Jones, head of strategy and development at SCOR, makes in this excellent article: it’s a myth that insurtech has been around only since 2015. We do, however, believe that there has been a new thrust since then, harnessing the pace and power of new technologies.) 2015-2017: The first wave of insurtech It is not surprising that insurtech started as a personal lines disruption play. Entrepreneurs, buoyed by what was happening in fintech and other industries, saw huge opportunities to make insurance more customer-centric based on their own experiences. Entrepreneurs wanted to simplify insurance (e.g. Sherpa), offer more tailored propositions (e.g. Bought By Many) or change the whole insurance paradigm (e.g. Guevara). But the truth is that insurance has not been disrupted over the last three years, and it’s hard to see that this is about to change. As Adrian illustrates in another article, even the most prominent disruptors in the U.S. (Lemonade, Metromile and Root) are finding the going tough and burning through a lot of capital, whether directly or via  reinsurance. See also: Digital Playbooks for Insurers (Part 1)   We argue in our insurtech Impact 25 paper (February 2018, page 7) that many distribution insurtechs are not scratching sufficiently major customer itches to be worth the switching cost for those consumers. As a result, the perceived potential is worrying incumbents far more than their actual performance to date. 2018: The second wave of insurtech If we were to update our insurtech landscape infographic, supplier insurtechs would feature much more prominently. These companies are developing technology (or, as in the case of German insurtech Kasko, have repurposed consumer propositions) to help incumbent insurers, reinsurers and brokers operate more effectively. Supplier insurtechs have found getting traction in consumer markets tough and are developing technologies or techniques that they can sell to the established insurers. Many of these companies are targeting corporate and speciality underwriters. This is perhaps not surprising – at least not from the U.K. perspective. U.K. personal lines insurers have been investing in pricing capabilities, efficiency and fraud analytics for years as competition has become cutthroat. They are mostly advanced in many areas. This is in strong contrast to corporate and specialty classes, where much underwriting is still judgment-based, processes are manual and underwriters and risk managers are resigned to poor data quality. As such, we believe that many of the Impact 25 Members can be valuable for corporate and specialty underwriters in 2018. Some examples are below:

• Insurdata was set up by ex-RMS executive Jason Futers and helps (re)insurers obtain more accurate building location information. This is helpful for underwriting (e.g. commercial property, reinsurance portfolios), risk management and portfolio reviews.(website, Impact 25 two-pager)
• Risk Genius uses AI to read policies and understand coverage. Founder Chris Cheatham noted recently. “[My trip to] London was amazing. It took two days for one very big learning to sink in: Underwriters in Europe are empowered to manuscript with little or no formal approval process.” His business allows corporate insurers to get a better understanding of their exposures.(website, two-pager)
• Flock is an analytics platform currently used to price drone flights dynamically, for example taking into account hyper-local weather conditions and locale of flight. The technology’s ability to process big data quickly could be helpful for commercial IoT propositions, for example. (website, two-pager)
• Cape Analytics and Geospatial Insight generate underwriting or claims insight from aerial imagery. This is useful, for example, in natcat losses when (re)insurers need to assess their exposures quickly. (Cape Analytics: website, 2-pager; Geospatial Insight: website, two-pager)

See also: Have Insurers Lost Track of Purpose?   What it means for corporate and specialty insurers Technology is not, of course, a new phenomenon in corporate and speciality insurance. However, the speed of proliferation of new vendors (of both technology solutions and data sources) is arguably unprecedented. It challenges the corporate clock speed of most incumbents and will present opportunities to successful adopters to tilt industry profits in their direction. But identifying the correct response is challenging for incumbents and, as we argue in our Impact 25 paper, there is no single, correct course of action. Choices that need to be made broadly fit into three categories:

• Strategy: Should we focus on customer experience/proposition or efficiency?
• Technology: Do we build or partner or buy? If we partner, how do we create and protect differentiating IP?
• Execution: Should we innovate within the business or in dedicated teams? What structures and processes do we need?

These questions – among others – need to be answered to ensure an effective corporate response.

Another week, another controversy on autonomous vehicles. The latest involves the crash that killed a Tesla driver on March 23 while the car was in Autopilot mode. After being silent for several days, Tesla on Friday wrote a blog post assigning much of the blame, if not all, to the driver. In a blog post, the company said he had received several visual warnings and one audible warning that he needed to retake the wheel but had not done so. The post said the driver had five seconds to react and 150 meters of unobstructed view of the concrete divider that he crashed into.

In the wake of these recent crashes, I think there will be a flight to quality, which doesn't bode well for either Uber or Tesla in the short term. Uber, which has been known as a cowboy, including in driverless cars, has been scrambling since a car in autonomous mode struck and killed a pedestrian in Tempe, AZ, on March 18. It's not clear to me that there's a fundamental flaw in Uber's technology, but the company seems to be moving too fast. It will have to scale back testing and increase security (there was just one safety driver, rather than the usual two, in the car that struck the pedestrian). Uber will also have to pound away at testing a lot more, both on simulators and in the real world, before putting any significant numbers of cars on the road.

Tesla may need to reboot even more. It claims it can get to full autonomy without using lidar, which the rest of the industry is treating as essential. The rotating lidar bulbs on the top of driverless cars surely don't fit the sleek design esthetic at Tesla—maybe lidar becomes acceptable once it can be delivered via just a couple of chips built into bumpers. But solid-state lidar is at least a few years away from being possible in production models, and I don't believe that Tesla can get close to full autonomy in the meantime. Tesla also needs to back off its belief that drivers can easily toggle between autonomous mode and full engagement with driving. Others aren't counting on drivers, who seem to need 10 seconds or more to reengage, and I think Tesla must change its thinking.

Companies taking a more deliberate approach will have to contend with the negative publicity from the recent Uber and Tesla fatalities. But I see no signs that Google's Waymo or GM's Cruise are backing off their plans to launch ride-sharing AV services this year and next in limited, well-mapped environments.

Vivek Wadhwa published an article with us that argues for a moratorium of at least two to three years on any testing around pedestrians. He owns a Tesla and is comfortable using it in Autopilot mode on highways but says local streets are too complicated for now.

Our own Guy Fraker conducted a thought-provoking webinar on the myths and realities of autonomous vehicles that you can listen to here. He provides great perspective both on how far we've come and on how far we still have to go.

Have a great week. (And drive safely.)

Paul Carroll
Editor-in-Chief