Attend any insurance-focused conference, and you will undoubtedly hear about the high volume of angel and venture capital currently chasing the next unicorn in insurance – the industry equivalent of Facebook or Amazon. In the search for value, investors are first asking questions about how long it will take for innovation to transform the industry’s business model or who will be the major disrupters. Is it the new approach to distribution that will render the current ecosystem extinct? Is it a different approach to underwriting courtesy of new advancements in data and analytics? These are legitimate questions – but what if we are looking in all the wrong places? What if the next Uber of insurance has already arrived? An important component of the value chain in consumer-facing markets is access to the customer. Creating a user-friendly application drives user engagement, and therefore retention. The critical difference in the insurance industry, however, is the lack of engagement. Insurers are challenged with a product that does not lend itself to a frequent interaction with clients. There are typically only two points of contact: sale and claim payment. Though both present opportunities to build customer relationships, this number is orders of magnitude lower than for consumer technology applications. This means insurers need a better solution to work around the value chain to gain control of the customer, one that involves driving down costs. Herein lies an unexpected solution – alternative capital. Alternative capital refers to pools of capital available for the transfer of risk from an insurer to the capital markets, typically in the form of insurance-linked securities (ILS) or special purpose vehicles (SPVs). The impact of alternative (or third party) capital is well understood in the reinsurance world today as global risks have been packaged, turned into portfolios and offered to the very largest providers of capital in the world as an alternative, risk-bearing asset class. The size of the market has grown considerably in recent years, with alternative capacity reaching $86 billion, or 14% of global reinsurance capital, as of the first quarter end of 2017, according to Aon. [caption id="attachment_28547" align="alignnone" width="500"] Alternative capital represents 14% of global reinsurance capital[/caption] The question, though, that investors should be asking is: Who has access to the cheapest capital? The answer to this question, the authors believe, will yield important insights as capital efficiency enables players in this space to get closer to the clients and, more specifically, the source of risk. After providing appropriate scale for disruption in the industry, along with an important framework to understand industry cost of capital, we consider evidence from the banking sector and anecdotal data from current trends in the market to argue that the Uber of insurance is already here. See also: Preparing for Future Disruption…   Disruption at Scale In an era of rampant overcapacity, it is sometimes easy to forget the historical significance of access to risk-based capital. The explosion of new technology applications to challenge the gatekeepers in the industry, funded by a dramatic rise in venture capital for Insurtech startups, has taken a lot of interest among industry participants. Insurance and reinsurance applicants are still, though, fundamentally sending submissions and applications to be accepted as insureds and cedants. There is an offer of risk transfer for premium. Unless platforms that control the customer also start retaining the risk with an equal to or lower cost of capital, the incumbents are still likely to control the system. Until this happens, those who build the best capital platform can offer the best products to attract customers, with the lowest-cost capital translating to lower prices for consumers – a reality supported by the scale of alternative capital disruption. [caption id="attachment_28548" align="alignnone" width="500"] Alternative capital currently on risk in 2016 dwarfs VC funding in Insurtech[/caption] While the scale of insurtech investment itself is enough to make serious people sit up and take notice, it is the speed, velocity, sources and efficiency of capital that will drive the future direction of the industry. To assess the long-term impact to the industry of these dueling solutions to disruption, an understanding of the drivers of industry cost of capital is required. Insurance Company Cost of Capital While ILS as a practice is nearly 20 years old, we are still in the first inning of how this once-niche part of the global reinsurance and risk transfer market will expand its influence. The current problem the industry is facing is that it does not do a good job of differentiating capital sources with specific levels of risk. Investors looking for a 5% or 7% or 11% return all fund risk whether at the 1-in-50 return period or 1-in-250 return period. This capital inefficiency not only lowers margins but also increases the cost of capital and perpetuates a system that hinders new product development at the expense of the end consumer. The use of third party capital gives underwriters the ability to cede remote, capital-intensive risk off their books and onto a lower-cost balance sheet. This matching of different types of risk with different pools of capital produces a leaner, more customer-focused, lower-priced risk transfer market, ultimately benefitting the end consumer from cheaper reinsurance products. The industry also cares about returns. Meeting analyst expectations has become increasingly difficult as pricing and investment yields have declined. Because insurance companies operate in a market that is becoming more commoditized, insurers and reinsurers need to either find another way to increase returns or figure out a way to lower their cost of capital. Alternative capital providers have a fundamental advantage in this respect, as “…pension fund investors are able to accept lower returns for taking Florida hurricane risk than rated reinsurers, for whom the business has a high cost of capital.” Similarly, the opportunity cost of capital for a typical venture capital fund (17% before management fees and carried interest) outstrips the hurdle rate for pension fund investors who turn to catastrophe risk as a diversifying source of return. Similarities to the Banking Sector The importance of cost of capital in determining winners and losers in capital-intensive industries can be clearly seen in the banking sector. When banks accessed lower-cost capital through capital market participants, their balance sheets were essentially disintermediated as they no longer had to finance the bank’s capital charges. By going straight from the issuer to capital and removing the need to finance these expensive capital costs, the aggregate cost of the system (value chain) was reduced. This benefitted consumers and drove retention in the same way we normally associate technology disintermediating distribution in other industries to drive down costs. Just as loan securitization transformed the banking industry, so, too, can risk securitization change the economics and value proposition of insurance for the end consumer. Capital disintermediation offers new operating models that connect the structurers of risk with the pricers of risk, until disrupters prove they can do this better. Because insurance underwriters and mortgage loan originators retain differing levels of risk ceded to the capital markets, gaining access to the most efficient forms of financing is an increasingly important battleground for players in the insurance space. When investors no longer tolerate capital inefficiency and increasing returns proves challenging, value chain disruption is a real threat because the lack of client proximity and customer engagement provides no competitive moat. A comparison between the recent growth in U.S. P&C industry direct written premiums and growth in U.S. P&C industry surpluses shows just why capital efficiency is so important. [caption id="attachment_28549" align="alignnone" width="500"] Renewed focus on capital efficiency as industry surpluses overtake industry premiums[/caption] Industry surpluses have increased at a rate more than double that of industry premiums ($418 billion increase vs. $200 billion increase) since 2002. The surplus growth relative to premiums highlights the steep drop in capital leverage from 1.4 in 2002 to 0.8 in 2016. As insurers sit on larger stockpiles of capital, they are relying less on underwriting leverage to juice up returns to meet their cost of capital. Hence, pricing power and cost controls have taken on increased importance. The Current Fall-out The growth of alternative capital has fostered innovation, both directly and indirectly, among industry participants across the value chain. Under the weight of efficient capital, industry players can take a step beyond leveraging insurtech to create value within the system – they can collapse it. In fact, we are seeing this happen right now. ILS funds such as Nephila are bypassing the entire value chain by targeting primary risk directly. Nephila’s Velocity Risk Underwriters insurance platform enables the ILS manager to source risk directly from the ultimate buyer of insurance. From consumer to agent to the Nephila plumbing system, homeowners’ risk in Florida can be funded directly by a pension fund in another part of the world. See also: Innovation: ‘Where Do We Start?’   Innovation is a healthy response, and this evolution of the reinsurance business model, while producing losers, will produce a leaner, more customer-focused, lower-priced risk transfer market in the end, ultimately benefitting everyone. The opportunity set for alternative capital to benefit the industry is not just limited to retrocession and access to efficient capital. Leveraging third party capital to provide better products and services to clients allows insurance companies to not only expand their current product offerings but also extend the value proposition beyond price alone, with different capital sources offering different structural product designs, terms, durations and levels of collaterization. Consider solar. The insurtech solution to creating a product that will provide protection for this emerging risk might include distribution and analytics. In contrast, partnering with third party capital providers can allow an insurance company to create a product with unique features such as parametric triggers that could solve a problem the traditional reinsurance industry, with its complicated and long casualty forms, has long struggled with. In doing so, third party capital can help grow the market for insurance in a way that insurtech has not yet achieved. Conclusion Two of the most common counterarguments for alternative capital’s Uber-like impact on the industry are that it remains largely untested capital in the face of a significant event and that we’ve hit a ceiling on the limit of third party capital in the traditional reinsurance market. On the first point, as of this writing, it is still too early to assess how existing and new potential third party capital investors will respond to hurricanes Harvey and Irma in coming renewals. However, the combination of the ILS market’s established track record, experience paying claims and fund managers staffed by people who have long experience as reinsurers provides support that investor interest post-event will remain strong. As for the second, in a static world, that claim may be true. But we believe there is plenty of room to grow in this market from both a supply and demand perspective. On the supply side, ILS accounts for only 0.6% of the global alternative investments market, with ample room to increase assets under management, according to the latest survey on alternative assets from Willis Towers Watson. On the demand side, around 70% of global natural catastrophe losses remain uninsured, and these risks are only growing. Insurers will require protection against aggregation and accumulation risk, and increasingly see the natural home for the tail risk in the capital markets and ILS. We have witnessed alternative capital’s ability to attract risk to capital and lower aggregate cost of the entire value chain. Many in the industry are monitoring how insurtech will sustainably attract risk and attract customers over the long term. In the meantime, the companies that have used their time wisely in the soft market by increasing capital efficiency to source, return and attract new forms of capital the quickest will be well-positioned in the space and can happily partner with great distribution partners of their choice. Even so, the players with the lowest cost of capital can accept risk more quickly and easily and can develop products more cheaply and with unique technological features for consumers. Insurtech will still play a pivotal role in shaping the future of the industry. Most of the participants in the insurtech space are in the early stages of capital formation. As investment scales and business models mature, insurtech should help leverage the proliferation of new sources of information and data pools to advance the securitization of different types of risk. Access to enhanced analytics helps make the underwriting and funding of more intangible risks, such as reputation and contingent business interruption, more sustainable, thereby increasing the participation of third party capital in lines of business outside property catastrophe. In the mid-1990s, far less than 1% of global reinsurance capacity was alternative capital; by 2016, its influence had grown to nearly 15%. Here is our question for the next industry conference: What will alternative capital look like in 2030?

Mobile has drastically changed the way we shop, travel, pay our bills and even pay each other. But there’s one area of our lives that it hasn’t changed enough: the way we manage our health. Mobile-focused health represents one of the biggest challenges – and opportunities – facing the healthcare industry. As more consumers connect their homes and lives across devices, particularly their phones, healthcare professionals must harness mobile health technologies and move toward a complete, mobile-optimized user experience. While most insurers already offer mobile apps, they often fail to create an experience that is both functional and intuitive. As our 2016 Digital Healthcare Survey revealed, digital health resources have been embraced by Americans of all ages, especially by younger Americans, with 82% of Gen Y and 67% of Gen X having accessed at least one digital health resource in the past 12 months. Of the digital resources offered by health insurers, mobile apps have the greatest potential to enhance Gen Y and Gen X member understanding and autonomy, but awareness of the apps and their functionality is low. Many Gen Y and Gen X members consider mobile access to their insurance a key resource, but only one-third (32%) are actually aware of whether their insurer even offers a mobile app. This represents a significant missed opportunity, for insurers and consumers alike. See also: A Road Map for Health Insurance   Fortunately for insurers, creating a mobile app doesn’t need to be overly complicated. The fundamental function of a health plan app is to provide members with access to the resources that are applicable to and useful for the mobile experience. However, many apps present far more than this – plan information, including balances, claims data and ID card information as well as coverage and benefits rates for health services, profile and account management options and customer service centers. For most customers, mobile apps don’t need all the resources and attributes of full sites – customers just want a mobile health experience that is intuitive, functional and fits in with their daily routine. So, what functionalities should insurers be looking to include in their latest mobile app versions? Take a page from financial apps, such as PayPal and Venmo, and offer a way for consumers to pay with ease. Incorporating payment features for claims and premiums, as well as push notifications alerting members to coming bills, would likely lead to more timely payments. UnitedHealthcare is one of the few providers that allow members to pay for a claim on its app directly by entering bank account information and then pre-filling most other important information, such as amount and payment recipient. Create visual representations, such as charts, graphs and progress meters, to help consumers better understand aspects of their plans like deductibles and coinsurance. Presenting plan balances and claims data not only improves the aesthetics of a page, but also provides members with a summary of data that may be easier to process. For example, rather than displaying how much of the plan’s deductible and out-of-pocket maximum the member has met, has remaining and has in total within a list format, use an interactive chart or graph to provide expedient summaries of data without sacrificing any detail – a particularly important feature on a mobile app given the limited space. Integrate health data from wearables to mobile apps (and vice versa) to encourage consumers to exercise regularly or eat healthy. Health assessments and connecting fitness apps to track movement are the most commonly rewarded activities, currently recognized by a majority of insurance platforms. Some insurers, such as UnitedHealthcare and Humana, are ahead of the curve, offering separate health and wellness reward program apps that employ push notifications to remind members to keep up with goals, such as “remember to get between seven and eight hours of sleep tonight" and "you have 2,000 more steps until you reach your goal for today.” See also: A Road Map for Health Insurance   While the healthcare industry overall still has a long way to go, digital health companies and startups have leveraged advancements in technology to enhance the mobile health experience for consumers. As functionality continues to improve and usage increases among younger members, the need for effective member support will become critical. Insurers should take note and make mobile health a priority – including functionalities and resources to help members better manage their health. We’ll all be better off as a result.

The commercial property and casualty (P&C) insurance industry has a classification problem. The overlapping use of standard industrial codes (SIC), Insurance Services Office (ISO), National Council on Compensation Insurance (NCCI), North American Industry Classification System (NAICS) and a dozen or so other carrier-imagined coding systems are creating a growing number of problems for insurance companies and agents alike. Ambiguity, misuse, line of business (LOB) specificity, misunderstanding and straight-up miscategorization leads to missed sales opportunities, higher underwriting costs and unexpected exposure to risk down the road. It’s time for the industry to examine risk categorization and take a fresh shot at solving the classification problem. The problem is, many of the fundamental issues stem from limitations of the underlying classification systems themselves. Dude, SIC Standard Industrial Classification (SIC) is like the mainframe computer of classification systems. Wildly out of sync with the modern world, it somehow is still extraordinarily prevalent in the wonderful world of insurance. SIC was first established in the 1930s as a way for government agencies to speak the same language with one another. Obviously, American businesses of the '30s, '40s, '50s and '60s looked very different than the business of today. SIC was developed mostly as a taxonomy for an industrial economy. Attempt to look up the SIC for web developers, for example, and you won’t find it. Many other modern businesses are also missing, because the SIC system itself was retired in 1997. SIC includes broad classifications and generalities, which are largely too unspecific to be of much use in classifying risk. Consider a fairly common code: 5812: Eating Places. Naturally, one would think immediately of “restaurants,” right? Unfortunately, that code also encapsulates industrial feeding, dinner theaters and sports arena concessions. While many insurers would likely write the GL on a restaurant down the street, few would be inclined to touch industrial feeding (whatever that is). What’s good about SIC is that it’s insurance agnostic and has applicability to all lines of business. What’s bad about SIC is that it’s too broad and, perhaps even more of a deal breaker, it stopped being updated two decades ago. See also: Future of Securities Class Actions   The Last DJ There’s a line in a song by Tom Petty & the Heartbreakers that wonders, “the boys upstairs want to see how much you'll pay for what you used to get for free.” This comes to mind anytime anyone advocates for a proprietary classification system. Setting aside the often-high licensing cost, these schemes are also almost always hardwired to the insurance industry, often because they have roots in rating. There are at least two significant consequences to this hardwiring. First, it creates a user experience challenge. This manifests itself to the policyholder, be it on an insurance form or a website. For an industry outsider, categorizing a business using an insurance code is just plain hard. This user experience deficiency also extends to the agent or CSR who must first be trained in the codes themselves before being able to begin to apply them. Second, and of increasing consequence, these values are not found in publicly available data. As insurers and intermediaries continue to integrate with any and all available public data to improve underwriting and reduce sales friction, proprietary, industry-specific classification is always something that will have to be derived after the fact, creating an exposure for error. With the continued use of proprietary classification schemes, industry participants are paying a bundle to speak a language that’s foreign to the rest of the economy. LOB-Specific Systems Workers’ compensation (workers’ comp) is perhaps the best example of a LOB with a classification system developed for pricing. Workers’ comp codes are exceptional for pricing workers’ comp. These codes do the job perfectly, but a coding system designed to attribute premium to payroll classes is not the same thing as a classification system to categorize the aggregate risk associated with the operation of the business. As an example, any insurer that has ever written workers’ comp will know that two of the top “classes” are clerical and outside sales. But what does that business with those exposures actually do? Is a business with clerical exposure an office full of desk workers, or is it a couple of managers at an industrial chemicals factory? The problem with workers’ comp codes, in particular, is that they describe what specific employees at the business do, not what the business itself does in the big picture. Expressing risk appetite is challenging, if not impossible, using payroll codes as a basis, and it’s difficult to convert these codes into eligibility for other LOBs for cross-selling other lines. A higher-level description of what the business actually does is required for portfolio underwriting — payroll codes are not sufficient. The case for NAICS NAICS is provided for free by the U.S. Census Bureau. NAICS classification provides very detailed descriptions of what it is that the business actually does, and, thusly, what exposures come with its operation. NAICS replaced SIC in the late ‘90s. In terms of detail, SIC codes are like on old tube TV, and NAICS is like a 4K flat panel. This high-definition classification has been updated to reflect the service economy and has a much lower level of detail describing exactly what a business does. The eating places conundrum discussed earlier in the context of SIC is a great example. With NAICS, it’s actually possible to code the restaurant down the street differently than a snack vendor at a stadium. As an insurer responsible for underwriting such risks, a high-definition description of the business operations is tremendously useful for pricing and underwriting. Clearly, not all eating places have the same exposures, so NAICS is a fantastic way for commercial insurers to express appetite. (Cafes, yes. Dinner theaters, no.) NAICS is also updated on a five-year rotation, most recently in 2017. Each revision builds on the last, incorporating new and emerging business types. Furthering the case for NAICS, the U.S. government has incorporated it as the go-to system for business classification, meaning that publicly-available data sources can and do provide these codes. See also: Chatbots and the Future of Interaction   A Call for Class Action Adoption of NAICS as as a standard for commercial P&C risk classification benefits the industry in all phases of the policy life-cycle — from reducing friction in how products are distributed; to reducing exposure to mis-categorized risk; and to enabling portfolio underwriting, all while supporting the new businesses being created in the modern economy. It’s long past time we all got on the same page.

We have all become totally wrapped up in the “Innovation Theater.” Some have complained of song and dance routines; others, like Steve Blank, have described their view of “innovation theater” based on the current obsession of setting up outposts, mostly in Silicon Valley. We seem to be layering on more and more activities to grab the attention and spread out our innovation resources. This exploring of different innovating activities includes a growing number of many of our larger companies (perhaps with deeper pockets) trying a variety of creative thinking concepts. At first, they try to imitate startups — going to Silicon Valley, taking organized bus tours, guiding money to accelerators, attending the latest “fashion show”/conference/workshop, hiring a prophet or innovation guru, setting up offices/layouts/hot desks/open formats with creative areas, encouraging the “casual” and changing the environment to how they think innovation will work and thrive. Honestly, the list gets longer by the day. Is this for excitement and buzz, trying to show that large organizations are hip, reacting to market and shareholder pressure? Or is there a real underlying purpose of building innovation capabilities in well-thought-through ways? All this activity shows that innovation should be treated as different I have absolutely no problem at all with any of these activities. They are creating and changing the innovation environment, climate and culture. But to draw out some of the necessary differences that are required to create a clearer innovation environment, let’s start with the need for different attention, different attitudes and different treatment to connect it all. Innovation needs to be treated for what it is: emerging in design, not established in its routine and maintenance of what has been achieved to date. Innovation holds the future, and that is always uncertain, unpredictable and uncomfortable. Yet, is the sum of all these efforts actually moving that innovation needle forward? Are we seeing an accelerating growth from our innovation efforts, especially from large corporations? It is highly questionable when most of our larger organizations are mostly short-term-focused and highly risk-adverse. Innovation simply struggles to sit alongside( or even within) the business when there is this overriding need to be cautious and horizon-one-driven. We often end up with better incremental innovation, whereas the radical or distinctive innovations still seems to remain elusive. Step back before we take on more of the innovation theater Perhaps we need to step back before we keep investing in all this worthwhile “innovation theater” and restate some of the very basic needs on what innovation needs to achieve. There are some basic tenants for innovation.

1. In all our activity and pursuit of innovation, we need to maximize the value. We need to learn how to make better investment decisions. This is not to drown ourselves in even more metrics that are aligned far more to the mature organization. Most of these don’t even work in innovation exploratory work. It is the art of constructing better investment decisions that simply increases the return on innovation investment with better value outcomes.
2. We need to find sustaining ways to balance our investments. All the initiatives that require innovation investment need to be managed effectively. You can’t hide from this fact, otherwise innovation remains ad hoc, often discovered by chance or simply copying others to keep growth moving along. Balanced portfolio’s and healthy pipelines are essential to deliver the best message to the ones supplying the money of “we are focused, back us.”
3. We continue to fall into those easy traps of not aligning innovation to the business strategy. We fail to have a robust understanding of alignment and often get caught up in the excitement of a discovery that is totally disconnected from this alignment but was a wonderful experience to work upon. We do need to align innovation to corporate need.
4. We fail to establish different time horizons of innovation need. Finished innovation outcomes do not neatly fit into the calendar year; they need to be seen and evaluated differently, based on their complexity, newness to the world and value potential. The three frames of thought, referred to as the three-horizon framework, break innovation down into three horizons. Horizon 1: goals or outcomes that contribute to the immediate plan and can be budgeted in good, granular ways with solid detail. Horizon 2: goal objectives that move concepts or ideas forward but have longer-term horizons before they yield a outcome return, where some actual spending gets accounted for in multiple years where you can provide reasonable forecasts or milestones towards validation. Horizon 3: ideas that offer the potential for a new state of innovation that explore many unknowns but are working toward a new future state. This final horizon is where pilot money is allocated and projected out over learning activities and agreed milestones; where understanding and recognizing investigation may involve years of exploration, connecting up multiple dots and growing recognition of different degrees of failure. Planning and acknowledgement of each horizon need must be recognized as distinct, accounted for in their differences and laid out in some form on innovation roadmap to cover all three for balanced progression.

See also: Innovation Maturing Into Major Impacts Speeding up innovation comes in thoughtful ways There is this missing element of knowing the best speed of innovation, and, surprisingly, this can come through managing a robust portfolio management system to speed acceleration and acceptance. If we don't have this clearly defined pathway or roadmap, we can't create the culture of innovation that is required to offer a sustaining route. A route where people, ideas, concepts, processes and tools can be developed, refined and improved to sync on a sustaining basis. If we allow innovation to disconnect and remain fuzzy, we enter the world of guesswork, lowering the identification of leadership buy-in. Getting this pathway established takes lots of hard work, and it needs a broad governance understanding and a combination of piloting and shepherding of all its areas across the parties that have a vested interest. I have continually recommended an overarching integrated innovation framework approach (a work mat methodology) to work through as it clarifies and communicates the innovation intent throughout the company. It is the “combination effect” of strategic alignment, with innovation investment providing thoughtful resource allocation and accessing good information and data that progressively reveals itself as the innovation potential we are in search off. This requires high levels of visibility across all parties or stakeholders to deeply appreciate how innovation is contributing into the total corporate picture. Many often argue that by having a structured innovation process, this portfolio management approach slows down the process. Established and accepted tools inhibit freedom of selection, and established processes often take away essential decisions and flexibility. Perhaps. Regretfully, in large companies you do need to optimize. It's engrained in the way business is done. Operationalizing innovation simply doesn't crush creativity or entrepreneurial spirit; it helps provide checks and balances and assess risk if it is more visible or structured for the multiple voices to relate to. The coordination of this is hard work, often trapped within the stage-gate process, which can, if allowed, be a pity if it becomes too overbearing in its process evaluation needs. I liked the “What’s Next? After Stage-Gate” provided by Dr. Robert Cooper, the creator of the Stage-Gate, that suggests taking a triple-A approach (adaptive, agile and accelerated) toward modernizing the gate-process. The table of the next-generation system — alongside the established, well-known Stage Gate method within this document — offers a view on am ideas-to-launch methodology that reflects more of today’s reality. Look a little deeper into the portfolio management system Firstly, managing the portfolio management system is a skill. It needs to account for this balanced investment in innovation to achieve different horizon objectives of business need. These innovation investments need funding, resourcing and tracking as they progress to move toward those business objectives. It is made up of desired business objectives and the concepts that contribute to this. The portfolio is dynamic and should be evolving and optimizing. It needs to be visible, and it really requires a robust software solution to allow traceability, tracking and monitoring — meaning the portfolio needs to be structured and repeatable. To invest in such a system, you can assess its return through different criteria. You can check the work undertaken if it aligns into the company strategy. Decisions can be tracked, traced and measured for time. You can look at things based on resources to bear and development decisions to be made. You can be alerted to slippage and delve into the reasons why. Additionally, the system can visualize the dynamics within the product portfolio to allow for better decisions and recognition of innovation’s contributions. One critical point of any portfolio management system is setting the right investment criteria It's not simply ideas that need to be within the PPM. Concepts, development initiatives and in-market products need to be included if you want a clear line of sight into your innovation portfolio. There are multiple perspectives to determine their level of fit and importance. For example, you have the expected financial value (revenue, achieved and projected — ROI, NPV, IRR, RONA, etc.) that constantly change as data and information change; they are highly dynamic and valuable if they are constantly updated. Then there’s the need to establish the customer/consumer value, the strategic value, the degree of alignment and the outline of feasibility and the ongoing indication of resource demands. There also needs to be a note made of dangers to be flagged if there are signs of slippage or changes in key criteria that require collective management attention. CISCO has developed a “Value at Stake” that looks at the value often left on the table because of a lack of digital investment or thinking in broader value generating ways. This investment criteria takes on a whole forward-looking assessment when you look through this type of value-by-the-stake concept. Establish a basic but critical set of criteria for a good product portfolio assessment We must establish a consistent set of messages to guide the conversation and assessment for those to recognize the innovation contribution value. So often we fight shy of attempting these, and this is even more often the reason why innovation fails. Of course, finding, validating and explaining these in clear factual ways is hard, because much in innovation is unknown. Avoiding the attempt to quantify simply relegates innovation to a side show. Management is sensible enough, in most cases, to recognize much can be open to interpretation and still needs clarifying. We must accept we don’t live in a perfect world and that innovation is balancing emerging ideas within a set of risk judgements. I would suggest embracing this part of any discussion robustly and taking it as essential — it prepares you for strategic management. Strategic criteria in any assessment You have to show the rate of change, risk and transformation that everything is undergoing — your markets, your competitors, etc., anywhere technology changes. Get a good handle on your competitors' perception, directions of travel and strengths/weaknesses. You need to be ready to highlight competitors' specific initiatives or recent patent activities that show emerging hotspots and might even point to emerging opportunities. Know and be able to clarify your position in the industry, across markets and in customer perceptions. There needs to be a healthy discussion on the appetite of risk, experimentation, learning and impact of technology. See also: Innovation Pivots: 10 Lessons Learned   The specific innovation criteria assessment The specific innovation criteria assessment includes the uniqueness of the idea or concept, probabilities of success from different perspectives of technical and commercial challenge, the current and predicted cost to completion and the timing and any next decision points where investment is required. Then you need to discuss the barriers or potential of ease of copying by others and the forecasted durability of any competitive advantage or new emerging ones. These need evaluating and rating within the portfolio to provide a clear understanding of the value of innovation. Summary If you don’t have a good, robust portfolio management system in place for innovation, you will struggle. No amount of innovation theater will make up for this basic need. Portfolio management is a cost that it requires dedicated resource commitment and investments (in specialized software and dedicated people schooled in project management and strategic and tactical evaluations). A good management portfolio has a central role to play in contributing to your innovation activities. If the innovation theaters generate sound and excitement, how you house and capture the value will determine if the efforts are well-directed, well-centered and fit for purpose. The article was originally published on Hype.

Once an injured worker has low back pain, chances are pretty good he’ll have a recurrence. In fact, new research shows that after an acute episode of LBP, one-third of people will have another episode within a year; and the odds of a recurrence triples after two episodes. Those statistics are somewhat surprising, given that recurring LBP can be avoided. Gustavo C. Machado and his colleagues at the University of Sydney’s School of Public Health tracked more than 1,000 LBP patients in Australia from 2011-12. Their results were published by the American Physical Therapy Association. “We know if you do exercise and receive advice you reduce the risk of having a recurrence,” said Machado, lead author of the study. “Some research shows if you do exercise, you reduce your risk of a back pain episode by 35%; and if you do exercise and get some sort of education or information, that risk is reduced by 45%. So it’s a big reduction in risk. “ With LBP one of the most prevalent causes of workers’ compensation claims, the industry could prevent unnecessary pain among workers and save significant dollars for payers by better educating workers on LBP. LBP An estimated one-third of workers’ compensation claims involve LBP, with direct costs of more than $14 billion annually. Some 1 in 4 workers with LBP remain out of work for up to six months, increasing the chances of permanent disability. Workers of any age can experience LBP, though it typically affects those over the age of 40. The reasons for LBP can vary with age. “With older workers, you see degenerative changes,” said Daniel Sanchez, a physical therapist and VP of Operations for OnSite-Physio, a company that provides PT to injured workers on site. “For younger workers, the LBP tends to be muscular soft tissue in nature. But both degenerative and soft tissue LBP can be recurring.” See also: The State of Workers’ Compensation   Acute episodes of LBP last for no more than six weeks, while chronic LBP continues for at least three months. While injured workers with acute LBP should avoid vigorous activity, that does not mean they should be bedridden. “The first thing initially is to rest, but don’t over rest,” Sanchez explained. “Depending on the severity, maybe a day or two of not overdoing it, but not being laid out in bed all day. Just not doing heavy exercise; no movements that aggravate the pain.” The simple act of reaching down to pick up something can aggravate the pain. For the first couple of days, injured workers should “take it easy, allowing the body to heal,” Sanchez said. Machado concurred. “For those patients, the best thing is to remain active. So do not stay in bed, just keep on the move. Try to keep up with routine activities, such as work,” he said. “It’s also very important to receive advice and education about pain, about recurrences.” The majority of LBP cases resolve within a few weeks. For those that linger, a different regimen is needed. “A quick fix for LBP does not exist,” Machado said. “The latest research shows taking pills doesn’t help much. Exercise and education seem to be the key for treating back pain, reducing back pain and preventing recurrences. The problem is people are not engaged in exercise.” Exercises There are a variety of exercises touted as best for addressing back pain: pilates, yoga and strengthening among others. They are equally effective. “The more research that comes out, the more it’s clear there is no one exercise that is better,” Machado said. “The best is the one you like to do, they one you’re going to engage in, the one you’re going to stick to.” A combination of exercises that improve stability, flexibility and strength will increase mobility and range of motion, thereby reducing pain. When muscles are tight and weak, the joints cannot move properly, Sanchez said.

• Strengthening exercises. The most effective of these, according to Sanchez, are those that target the muscles in the front and back of the spine; the deeper, transverse abdominis that support the spine. Planks are among the most effective and can help the intrinsic, tiny muscles that attach to each vertebrae. These provide postural support which is very important.
• Flexibility exercises. These, especially to the lower extremities, are extremely important. The lumbopelvic hip complex includes muscles that attach from the lower extremities, such as the hip rotators, hamstrings and hip flexors. Exercises that target these areas can relieve tight hips, which helps relieve chronic LBP.
• Posture. Practicing sound posture, good body mechanics and lifting habits are also helpful, Sanchez said. Having a neutral spine is the goal.

Future of LBP The research on LBP should include more robust studies looking at recurrence, Machado said. In the meantime, he and his colleagues are analyzing other aspects of LBP. “The main one we are looking at is over diagnosis and treatment, because a main issue is that people get lots of X-rays and imaging and that’s usually unnecessary,” Machado said. “Lots of people also get opioids. We know that’s a big problem, especially in the U.S. It is not helpful. There are few benefits and really serious risks for side effects.” Machado has a trial study starting soon in Sydney. Along with others, he is also looking into technology; specifically, smart phone apps that claim to help back pain. See also: 25 Axioms Of Medical Care In The Workers Compensation System   “We found over 69 apps to download. They’re making big promises but have not been tested for effectiveness,” he said. “We are planning to do another study in a few months on a specific app that could be promising… This one recommends a 10-week exercise program, mainly strengthening. The problem is there is no research testing this app as to whether it’s effective in reducing pain.” For now, Machado hopes the latest study will provide guidance for providers treating patients with LBP. “We didn’t know how common recurrence was; it shows one-third after recovery will have another episode,” he said. “That’s something a clinician can use to educate a patient when they come, to say ‘look, a third of people have a recurrence, so engage in exercise.’ They can use this to educate them now.” For more information visit https://www.apta.org or http://www.onsite-physio.com.

Major websites all over the world use a system called CAPTCHA to verify that someone is indeed a human and not a bot when entering data or signing into an account. CAPTCHA stands for “Completely Automated Public Turing test to tell Computers and Humans Apart.” The squiggly letters and numbers, often posted against photographs or textured backgrounds, have been a good way to foil hackers. These are annoying but effective. The days of CAPTCHA as a viable line of defense may, however, be numbered. Researchers at Vicarious, a Californian artificial intelligence firm funded by Amazon founder Jeffrey P. Bezos and Facebook’s Mark Zuckerberg, have just published a paper documenting how they were able to defeat CAPTCHA using new artificial-intelligence techniques. Whereas today’s most advanced AI (artificial intelligence) technologies use neural networks that require massive amounts of data to learn from (sometimes millions of examples), the researchers said their system needed just five training steps to crack Google’s reCAPTCHA technology. With this, they achieved a 67% success rate per character — reasonably close to the human accuracy rate of 87%.  In answering PayPal and Yahoo CAPTCHAs, the system achieved an accuracy rate of greater than 50%. See also: The Insurer of the Future – Part 3   The CAPTCHA breakthrough came hard on the heels of another major milestone from Google’s DeepMind team, the people who built the world’s best Go-playing system. DeepMind built a new AI system called AlphaGo Zero that taught itself to play the game at a world-beating level with minimal training data, mainly using trial and error — in a fashion similar to how humans learn. Both playing Go and deciphering CAPTCHAs are still clear examples of what we call narrow AI, which is different than Artificial General Intelligence (AGI) —  the stuff of science fiction. Remember R2-D2 of “Star Wars,” Ava from “Ex Machina” and Samantha from “Her?” They could do many things and learned everything they needed on their own. The narrow AI technologies are systems that can only perform one specific type of task. For example, if you asked AlphaGo Zero to learn to play Monopoly, it could not, even though that is a far less sophisticated game than Go; if you asked the CAPTCHA cracker to learn to understand a spoken phrase, it would not even know where to start. To date, though, even narrow AI has been difficult to build and perfect. To perform very elementary tasks such as determining whether an image is of a cat or a dog, the system requires the development of a model that details exactly what is being analyzed and massive amounts of data with labeled examples of both. The examples are used to train the AI systems, which are modeled on the neural networks in the brain, in which the connections between layers of neurons are adjusted based on what is observed. To put it simply, you tell an AI system exactly what to learn, and the more data you give it, the more accurate it becomes. The methods that Vicarious and Google used were different; they allowed the systems to learn on their own, albeit in a narrow field. By making their own assumptions about what the training model should be and trying different permutations until they got the right results, they were able to teach themselves how to read the letters in a CAPTCHA or to play a game. This blurs the line between narrow AI and AGI and has broader implications — in robotics and in virtually any other field in which machine learning in complex environments may be relevant. See also: Seriously? Artificial Intelligence?   Beyond visual recognition, the Vicarious breakthrough and AlphaGo Zero success are encouraging scientists to think about how AIs can learn to do things from scratch. And this brings us one step closer to coexisting with classes of AIs and robots that can learn to perform new tasks that are slight variants on their previous tasks — and ultimately the AGI of science fiction. So R2-D2 may be here surprisingly sooner than we expected.

This is the fourth in a series. You can find the first three parts here, here and here. The Insurer of the Future’s use of blockchain will depend on whether it is a new entrant or a traditional player. For new entrants, blockchain will be at the core of both their business model and their operating model. The insurer will use blockchain to:

• Underpin a series of smart-contract-enabled parametric insurance products (if event X happens, and "oracle" Y confirms that, then pre-agreed sum of money Z is paid out automatically); and
• Maintain secure policy records significantly more cheaply than its legacy competitors.

If the Insurer of the Future was a traditional player, it’s more likely to be using blockchain as a "bolt on," supporting new products that wouldn’t otherwise be cost-effective. The insurer might, for example, use blockchain ledgers to support micro-insurance policies. An example could be insuring jewelry just for the time its owner plans to wear it this evening. Or providing top-up insurance to participants in the gig economy, lasting just for the length of each gig. See also: Blockchain: Basis for Tomorrow   But whether the Insurer of the Future is a new entrant or an existing insurer, blockchain will be just one of a number of new tools at its disposal. This is one area in which new technology will be incremental to the industry rather than truly disruptive. Unless you think otherwise?

Disruptive Innovation First things first, let me set some context. A question I get asked a lot by insurers and industry folk is, “why should we be interested in what Lemonade are doing?” It’s a great question and exactly what they should be asking. (I also point out that they need to be really interested in what ZhongAn is doing, as well). To massively over-simplify and paraphrase Clayton Christensen, Lemonade has brought simplicity, convenience and affordability to a marketplace where the existing offering is complicated, expensive and inaccessible. This is why the incumbent insurers need to take note when Lemonade pays a claim in three seconds. Otherwise, they could end up like DEC. Once the market leaders in minicomputers, DEC dismissed the rise of PCs, only to watch helplessly as IBM and Apple ate their lunch with personal computers. Or Kodak, the inventor of digital photography. The company was too wedded to an outdated business model that relied on people printing their photos. That was until it was too late, and Kodak went from being the world’s fourth largest brand to bankrupt in less than two decades! Now, it might have taken about 15 years for the demise of Kodak and about 10 for DEC to wake up and smell the coffee. The point being that disruptive innovations don’t take hold overnight; they need time to gain traction and build momentum. But in this digital age, this speed of change is increasing. This is the key characteristic in the World Economic Forum’s definition of the 4^th Industrial Revolution. It took Google just five years to hit a $1 billion in revenues. And Amazon only four! Just think about this for a second. A decade ago, we didn’t have the iPhone, the iPad, Kindle, Uber, AirBnB, Android, Spotify, Instagram, WhatsApp, 4G. Could you imagine life without these now? Could you conceive that insurance is going to change and for the better? You trust me, and I will trust you There is another reason why incumbent insurers should be watching Lemonade very closely. It has addressed the fundamental issue with insurance and customer perception, which is trust, behavior and the conflict of interest. There’s a ton of research and data that shows customers don’t trust insurers. And for good reason. Insurers make the product complicated by using fancy jargon that Joe and Josephine Bloggs can't understand. Insurers get paid up front and then create hurdles and barriers when the customer rightfully asks the insurer to do what they’ve already paid them to do. And worse, the customer has to prove they are not a liar to the insurer’s satisfaction before a penny is paid out. “Insurance fraud has become a self-fulfilling prophecy for incumbent insurers,” Daniel said. “They don’t trust customers to be fair and honest. This drives their behavior toward customers. And guess what, customers respond accordingly. Which justifies the insurer’s behavior in the first place. It’s a vicious circle that neither side can break.” See also: Lemonade’s New Push: Zero Everything   Lemonade’s virtuous circle This conflict of interest doesn’t exist in the Lemonade business model. By operating as a tech platform that is also an insurance carrier, Lemonade has separated cost of operations from the pool of risk capital. It has also raised the bar when it comes to total cost of operations at 20% GWP. Lemonade don’t profit from non-payment of a claim (in the way an incumbent insurer does). The company starts by trusting customers to make honest claims. Which is why Lemonade pays out straight away, with around a third of claim payouts fully automated. No human intervention at all. Lemonade accepts that there are a few bad apples but works on the premise that most of us are fundamentally decent people. It is usually at this point that the diehards and old laggards of the insurance industry start throwing fraud and loss data at me. Citing decades of data that proves Lemonade will eventually crash and burn under the weight of inflated and illegal claims. My response is always the same “hands up everyone who is a bad person.” Of course, no hands go up because the vast majority of us are decent, respectful, honest people. Which is why Lemonade has now had six, yes ,SIX, customers who have handed claims payouts back. Just think about this for a moment. A customer makes a claim (in seconds), gets paid (immediately), finds the situation has changed (later), realizes he got paid too much (oops!), then gives the payment back (you kidding me?). Could the customer’s behavior be directly related to Lemonade’s behavior? Yes, certainly! You only have to look at customer behavior at Grameen Bank in Bangladesh to see that trust can be relied  upon. Here, unsecured personal loans are repaid on time without the need for credit scores and debt collection agencies. You don't have to take my word for it, either. Hot out of the oven is this video of Lemonade customers in New York. So, what’s the story, one year on? Lemonade has been true to its word on the subject of transparency. Throughout the year, the company has published its numbers, warts and all, for everyone to see. Building and maintaining trust is fundamental to Lemonade’s business model, and this starts with being open and honest. Daniel has shared with me the latest numbers, and they are very impressive. I won’t repeat them here, because I know the team will be posting them all shortly in the latest Transparency Chronicles. They’re proud of the numbers, and rightly so. See also: Lemonade: World’s First Live Policy   All I will say is that Daniel and the team have steered a considered and thoughtful course in their first year. They could have chased the numbers, as many first year startups would do, only to regret the quality of business they end up with. But Lemonade's team has stuck to their knitting, have impressive growth numbers, a quality customer base completely aligned to the brand and are now licensed in 18 states (with more to follow). “Our job has only just started,” Daniel said. “Over the next year, we will continue to make insurance easier and better for our customers. One area we’ve started to look at now is the underlying insurance language and the products that form the heart of all insurance.” Are you surprised? You shouldn’t be! Lemonade is a highly professional startup and will no doubt become the definitive case study for exactly how “it” should be done. But has this surprised Daniel? “There are two things that have surprised us this year,” Daniel told me. “First, the extent of the warm reception we’ve received across the industry and from customers. We hoped customers would like us, but we never took for it granted. “After all, you can’t beta test a new insurance company. The MVP (minimally viable product) approach simply doesn’t apply to insurance. It’s regulated and has to be the real deal from the get-go, right first time. So, for us, having customers put their faith in Lemonade from Day One has been very satisfying. “The second is that our faith in humanity and behavioral economics has been affirmed. There will always be people who want to game the system, but on the whole, all our expectations about customer behavior have been exceeded. “Who would have thought we would have six customers who gave their claim payouts back. That is very gratifying and also humbling for us. And gives us encouragement to continue doing what we are doing.” Lemonade is live; insurance will never be the same again! For me, I’m convinced. Historians will look back to Sept. 21, 2016, the day that Lemonade opened for business, as a watershed for the insurance industry. Which means, of course, that the key question now is, who among the incumbent insurers will provide the Kodak moment? The one who simply missed that the world had changed until it was too late.

We know that early intervention is critical to prevent delayed recoveries for injured workers. One of the challenges has been to identify those at higher risk of poor outcomes. Fortunately, we have the tools to determine which patients are more likely to develop chronic pain and languish in a disability mindset. The process is fairly simple and backed by strong, research-based evidence. With increased awareness among payers, providers and other industry stakeholders, we can prevent creeping catastrophic claims, help injured workers regain function quicker and significantly reduce workers’ compensation costs. Reasons for Getting Stuck Psychosocial risk factors used to be little more than a buzz term among workers’ compensation professionals. While those of us who’ve worked extensively with chronic pain patients understood that psychological issues can easily derail a workers’ compensation claim, the research that proves this to be true has become widespread only in recent years. In fact, some of the most recent research says that psychological factors can be more of a predictor of poor outcomes than the underlying medical conditions. We now know for certain that the biomedical model of disease does not hold true for everyone, and the biopsychosocial model of illness must be considered. Where the first is based on the idea that a physical ailment can be cured through medical solutions, the second acknowledges that some people have an underlying psychobiological dysfunction that has clinically significant distress or disability. They are the injured workers who can greatly benefit from early identification and intervention. Inadequate coping skills and a lack of knowledge of what is causing their pain can drive delayed recoveries and overuse of treatments and medications. Chronic pain is the final common pathway of this delayed recovery. See also: A Biopsychosocial Approach to Recovery   Research validated through meta analyses, prospective studies and control group studies shows that injured workers with delayed recoveries typically have:

• Catastrophic thinking
• A history of anxiety or depression
• Anger and perceived injustice about their plight
• An external locus of control
• Minimal resilience

They may also have fear avoidance, meaning they engage in little to no physical activity out of fear they will injure themselves more and experience increased pain. There are myriad reasons why some people have these issues. The cause could be childhood and life experiences, their relationship and interactions with their environments, issues in the workplace or home or other reasons altogether. It’s important that we identify injured workers with these issues as soon as possible after their injuries. Pain Screening Questionnaires One of the most effective ways to pinpoint injured workers with psychological issues is through specially designed, self-administered questionnaires. The one we use to identify patients at risk of developing chronic pain and disability is the Pain Screening Questionnaire (PSQ). The PSQ was developed by a Swedish professor of clinical psychology and is used in many countries. It has been shown through studies to accurately predict time loss, medical spending and function — but not pain. The PSQ takes about five minutes to complete and consists of 21 questions that focus on the injured worker’s:

• Pain attitudes, beliefs and perceptions
• Catastrophizing
• Perception of work
• Mood/affect
• Behavioral response to pain
• Activities of daily living

The injured worker is asked to rate on a scale of 1 to 10 things such as, "How would you rate the pain you have had during the past week?"; "In your view, how large is the risk that your current pain may become permanent?"; and "An increase in pain is an indication that I should stop what I’m doing until the pain decreases." Depending on the score, the injured worker is categorized as low risk, moderate risk, high risk, or very high risk. Those on the lower end of the scale are most appropriately managed through take-home educational materials on chronic pain. Moderate-risk injured workers are good prospects for a self-managed workbook style intervention. High- and very-high-risk injured workers should be referred for additional assessment and an intervention program, such as cognitive behavioral therapy (CBT). See also: Impact on Mental Health in Work Comp   In a program of early identification and intervention, Albertson’s Safeway found 12% of injured workers scored high. Those affected were referred to CBT. After an average of just six CBT sessions, a large percentage of them were able to return to work. Because of the results, primary treating physicians who work with Albertson’s injured workers have been referring them to the program earlier in the claims process. Conclusion It is estimated that 10% of workers’ compensation claims consume at least 80% of medical and indemnity resources. The vast majority of these are injured workers with delayed recoveries due to psychosocial risk factors. With solid science backing up the successful identification and interventions of these employees, we can prevent needless disability and substantially reduce workers’ compensation costs.

Why should you care about machine learning (ML)? ML essentially refers to the phenomenon of computers and other devices that appear to be intelligent because they can learn from patterns in the environment to achieve a specified goal or predict an outcome. These patterns could come from a variety of sources, such as images, voice, free text or even structured data. The machine’s ability to anticipate or interpret the information given can make it seem almost human-like. For example, in the case of workers’ compensation claims, you can apply machine learning to feed the machine claim notes, and it will tell you which job class the claim is for. What makes this possible is ML’s ability to look at combinations of features in data and learn from them to make a variety of associations. Imagine a machine looking at claimant age, diagnosis codes and Part, Nature and Cause (PNC) codes in tandem to instantly determine the likelihood of litigation. Machines now have this ability, thanks to the increasing variety and volume of available data points. They can generate specific algorithms to decode a seemingly infinite number of patterns to make massive amounts of information truly useful. Why Does This Matter to Workers’ Compensation? Teams are constantly trying to get in front of their claims so that they can be proactive rather than reactive. To do that, they need a variety of signals sent to them in real time to figure out the best strategy for a claim. ML helps provide these signals much faster than ever before, and there are four key reasons why. 1. Machine learning can handle many, many combinations in the data … fast. Think of all the data you have to sort through to assess a claim. ML could be applied to increase speed and accuracy as well as to simplify the entire process and unlock invaluable predictive insights. Case in point: To get an accurate prediction of the projected cost of a claim, you might need to juggle as many as 45 different data features, including PNC codes, claimant information and diagnosis information. If you assume 10 values for each feature, which is a conservative estimate (International Classification of Diseases (ICD) codes have as many as 80,000-plus potential values), the number of possible combinations can reach a mind-boggling 10^45. ML algorithms, however, can navigate these combinations and associate patterns with specific data characteristics in minutes. Not only that, they can also determine which set of features contributed most to the outcomes. Talk about the ability to improve claim teams’ efficiency! See also: Machine Learning – Art or Science?   The ability for ML to sift through all these combinations of features and use all their interactions is absolutely game-changing. 2. It can handle “holes” in the data. Aside from quickly sorting through multiple data points and turning them into something meaningful and applicable, ML helps address another common problem: the presence of gaps in some of the fields in the claim or bill data, especially in the early stages of a claim. Techniques like data augmentation, where models are trained on various versions of a claim that expose different levels of gaps, can help ML models tolerate data “holes.” While it’s always important to get the best data one can for a claim, it’s equally important for models to be able to operate with incomplete information. ML makes it possible to move forward despite imperfect or incomplete claims. 3. It can handle changing data. Workers’ compensation claims are also constantly evolving, sometimes dramatically. What started as a neck injury can evolve into a spinal injury. In an ideal world, a claims team would be notified as soon as a major shift happened in one of their claims instead of waiting until the traditional 30-day, 60-day and 90-day check-in points. ML helps here, as well. Because ML models can handle gaps in the data, they can also instantly navigate the changing nature of claims, alerting claims teams immediately of changes. Not only can ML tackle changes in claim data itself, it can also handle changes in the overall operations of a claims team. A robust retraining schedule helps the ML models stay current so that they are perpetually unearthing new patterns to better assist when claims operations or macro-effects, such as new regulations, occur. 4. It can work with more natural forms of data, such as free text, voice and images. While this is all excellent news for claims teams interested in harnessing the power of ML to drive better care, I’ve saved perhaps the best benefit for last. One of the biggest advantages of ML is its ability to handle not only structured datasets but unstructured as well. What does this mean? ML can detect new patterns out of things like natural text and images. This has never been possible before. A simple application of this is in the use of ICD codes. Claims that have a combination of ICD-9 and ICD-10, for example, can be really clunky to deal with because the mapping between them is a little complicated. But with ML and natural language processing (NLP) techniques, we can use the ICD descriptions instead of the pure codes to unearth relevant themes and topics. Those become features for the models, and no more mapping is required. This is an area within ML that is evolving rapidly. We can see it all around us with devices like Amazon’s Echo and Apple’s Siri. In all likelihood, ML will generate a hotbed of activity in workers’ compensation predictive analytics, as well. Imagine a scenario where a claims examiner could get suggestions for doctors to recommend to the injured worker, all while they are typing in their claim notes. See also: Machine Learning: a New Force  Despite All Its Power, ML Is Just a Tool No matter how awesome and transformative ML seems, it’s not a magic bullet. At its core, it’s a useful new set of tools meant to empower claims examiners to do what they do best — be a coach whose main goal is getting the injured workers through their recovery. ML can redefine the role of the examiner as a coach and a problem-solver while removing large parts of their “rote” work. This enables examiners to spend more time talking to injured workers. Tack on the fact that ML can improve the efficiency of the entire claims process — from the ability to assess and process more claims faster and more accurately to getting employees the right care throughout their journey so that they can return to work and resume their lives — and ML has the capacity to provide data-driven insights that will help employees feel better faster, all while reducing costs overall. Deploying a combination of tools like ML with empathetic examiners who have strong problem-solving skills will elevate the entire workers’ compensation experience — for claims teams, employees and the companies they represent. It has the capacity to ultimately drive better care. While ML can’t do it all, it can modernize and fundamentally transform workers’ compensation. This article was first published in Claims Journal.