Recently, I was invited to present on the impact of big data and advanced analytics on the insurance industry at the NCSL Legislative Summit. This talk couldn’t have been timelier, as the insurance sector now tops the list of most disrupted sectors. Some of the culprits and causes for this top spot are related to the speed of technological change, changing customer behavior, increased investments in the insurtech sector and new market entrants, such as homeowners and renters insurance startup Lemonade. A significant driver of this disruption is technological change – especially in big data and advanced analytics.
Here are 10 key trends that are affecting big data and advanced analytics – most of which have a hand in disrupting the insurance industry:
Size and scope – Big data is getting bigger and faster. With connected cars, homes and buildings, and machines, the amount of data is increasing exponentially. Investments in IoT and Industrial IoT, 5G and other related areas will only increase the speed and amount of data. With this increased volume and velocity, we will not be able to generate meaningful insights from all of this data without advanced analytics and artificial intelligence.
Big data technology – Big data technology is moving from Hadoop to streaming architectures to hybrid “translytical” databases. While concepts like “data lakes” and NoSQL databases mature, new technologies like Apache Spark, Tez, Storm, BigTop and REEF, among others, are creating a constant flow of new tools, which adds to a sense of “big data in flux.”
Democratization – The democratization of data, business intelligence and data science is accelerating. Essentially, this means that anybody in a given organization with the right permissions can use any dataset, slice and dice the data, run analysis and create reports with very little help from IT or data scientists. This creates expectations for timely delivery, and business analysts can no longer hide behind IT timelines and potential delays.
Open source movement – The open source revolution in data, code and citizen data scientist is accelerating access to data and generation of insights. Open source tools are maturing and finding their way into commercial vendor solutions, and the pace of open source tool creation is continuing unabated; the Apache Software Foundation lists more than 350 current open source initiatives. This steady stream requires data engineers and data scientists to constantly evaluate tools and discover new ways of data engineering and data science.
Ubiquitous intelligence – Advanced analytics – especially various types of artificial intelligence areas (reference to my AI report post) – is evolving and becoming ubiquitous intelligence. AI can now interact with us through natural language, speak to us, hear us, see the world and even feel objects. As a result, it will start seamlessly weaving itself into many of our day-to-day activities, such as using a search engine or sorting our email, recommending things to buy based on our preferences and needs, seeing the world and guiding us through our interaction with other people and things without our even being aware of its doing so. This will further heighten our sense of disruption and constant change.
Deep learning – Deep learning, a subset of the machine learning family (which itself is just one area of AI), has been improving in speed, scale, accuracy, sophistication and the scope of problems it addresses. Unlike previous techniques, which were specific to the different type of data (e.g., text, audio, image), deep learning techniques have been applied across all different types of data. This has contributed to reduced development time and greater sharing and broadened the scope of innovation and disruption.
MLaaS – Machine learning, cloud computing and open source movement are converging to create Machine Learning as a Service (MLaaS). This not only decreases the overall variable costs of using AI but also provides large volumes of data that the machine learning systems can further exploit to improve their accuracy, resulting in a virtuous cycle.
Funding – Big data funding peaked in 2015. However, funding for artificial intelligence, especially machine learning and deep learning, has continued to attract increasingly significant investments. In the first half of this year, more than $3.6 billion has been invested in AI and machine learning. This increased funding has attracted great talent to explore difficult areas of AI that will be disruptors of the future economy.
Center of Excellence: As organizations continue to obtain good ROI from their initial pilots and proof-of-concepts in analytics, automation and AI efforts, they are increasingly looking toward setting up centers of excellence where they can train, nurture and grow the talent. The exact role of the center changes based on the overall organizational culture and how the rest of their business operates – centralized, federated or decentralized.
Competitive landscape – The big data landscape continues to grow, and the AI landscape is expanding rapidly. Deep learning companies are growing the fastest across multiple sectors. Competition among startups – as well as incumbents that want to stay ahead of potential disruption – is creating a vibrant ecosystem of partnerships and mergers and acquisitions that further the disruptive cycle.
Jeff Heepke knows where to plant corn on his 4,500-acre farm in Illinois because of artificial intelligence (AI). He uses a smartphone app called Climate Basic, which divides Heepke’s farmland (and, in fact, the entire continental U.S.) into plots that are 10 meters square. The app draws on local temperature and erosion records, expected precipitation, soil quality and other agricultural data to determine how to maximize yields for each plot. If a rainy cold front is expected to pass by, Heepke knows which areas to avoid watering or irrigating that afternoon. As the U.S. Department of Agriculture noted, this use of artificial intelligence across the industry has produced the largest crops in the country’s history.
Climate Corp., the Silicon Valley–based developer of Climate Basic, also offers a more advanced AI app that operates autonomously. If a storm hits a region, or a drought occurs, it lowers local yield numbers. Farmers who have bought insurance to supplement their government coverage get a check; no questions asked, no paper filing necessary. The insurance companies and farmers both benefit from having a much less labor-intensive, more streamlined and less expensive automated claims process.
Monsanto paid nearly $1 billion to buy Climate Corp. in 2013, giving the company’s models added legitimacy. Since then, Monsanto has continued to upgrade the AI models, integrating data from farm equipment and sensors planted in the fields so that they improve their accuracy and insight as more data is fed into them. One result is a better understanding of climate change and its effects — for example, the northward migration of arable land for corn, or the increasing frequency of severe storms.
Applications like this are typical of the new wave of artificial intelligence in business. AI is generating new approaches to business models, operations and the deployment of people that are likely to fundamentally change the way business operates. And if it can transform an earthbound industry like agriculture, how long will it be before your company is affected?
An Unavoidable Opportunity
Many business leaders are keenly aware of the potential value of artificial intelligence but are not yet poised to take advantage of it. In PwC’s 2017 Digital IQ survey of senior executives worldwide, 54% of the respondents said they were making substantial investments in AI today. But only 20% said their organizations had the skills necessary to succeed with this technology (see “Winning with Digital Confidence,” by Chris Curran and Tom Puthiyamadam).
Reports on artificial intelligence tend to portray it as either a servant, making all technology more responsive, or an overlord, eliminating jobs and destroying privacy. But for business decision makers, AI is primarily an enabler of productivity. It will eliminate jobs, to be sure, but it will also fundamentally change work processes and might create jobs in the long run. The nature of decision making, collaboration, creative art and scientific research will all be affected; so will enterprise structures. Technological systems, including potentially your products and services, as well as your office and factory equipment, will respond to people (and one another) in ways that feel as if they are coming to life.
In their book Artificial Intelligence: A Modern Approach (Pearson, 1995), Stuart Russell and Peter Norvig define AI as “the designing and building of intelligent agents that receive percepts from the environment and take actions that affect that environment.” The most critical difference between AI and general-purpose software is in the phrase “take actions.” AI enables machines to respond on their own to signals from the world at large, signals that programmers do not directly control and therefore can’t anticipate.
The fastest-growing category of AI is machine learning, or the ability of software to improve its own activity by analyzing interactions with the world at large (see “The Road to Deep Learning,” below). This technology, which has been a continual force in the history of computing since the 1940s, has grown dramatically in sophistication during the last few years.
This may be the first moment in AI’s history when a majority of experts agree the technology has practical value. From its conceptual beginnings in the 1950s, led by legendary computer scientists such as Marvin Minsky and John McCarthy, its future viability has been the subject of fierce debate. As recently as 2000, the most proficient AI system was roughly comparable, in complexity, to the brain of a worm. Then, as high-bandwidth networking, cloud computing, and high-powered graphics-enabled microprocessors emerged, researchers began building multilayered neural networks — still extremely slow and limited in comparison with natural brains, but useful in practical ways.
The best-known AI triumphs — in which software systems beat expert human players in Jeopardy, chess, Go, poker and soccer — differ from most day-to-day business applications. These games have prescribed rules and well-defined outcomes; every game ends in a win, loss or tie. The games are also closed-loop systems: They affect only the players, not outsiders. The software can be trained through multiple failures with no serious risks. You can’t say the same of an autonomous vehicle crash, a factory failure or a mistranslation.
There are currently two main schools of thought on how to develop the inference capabilities necessary for AI programs to navigate through the complexities of everyday life. In both, programs learn from experience — that is, the responses and reactions they get influence the way the programs act thereafter. The first approach uses conditional instructions (also known as heuristics) to accomplish this. For instance, an AI bot would interpret the emotions in a conversation by following a program that instructed it to start by checking for emotions that were evident in the recent past.
The second approach is known as machine learning. The machine is taught, using specific examples, to make inferences about the world around it. It then builds its understanding through this inference-making ability, without following specific instructions to do so. The Google search engine’s “next-word completion” feature is a good example of machine learning. Type in the word artificial, and several suggestions for the next word will appear, perhaps intelligence, selection and insemination. No one has programmed the search engine to seek those complements. Google chose the strategy of looking for the three words most frequently typed after artificial. With huge amounts of data available, machine learning can provide uncanny accuracy about patterns of behavior.
The type of machine learning called deep learning has become increasingly important. A deep learning system is a multilayered neural network that learns representations of the world and stores them as a nested hierarchy of concepts many layers deep. For example, when processing thousands of images, it recognizes objects based on a hierarchy of simpler building blocks: straight lines and curved lines at the basic level, then eyes, mouths and noses, and then faces, and then specific facial features. Besides image recognition, deep learning appears to be a promising way to approach complex challenges such as speech comprehension, human-machine conversation, language translation and vehicle navigation (see Exhibit A).
Though it is the closest machine to a human brain, a deep learning neural network is not suitable for all problems. It requires multiple processors with enormous computing power, far beyond conventional IT architecture; it will learn only by processing enormous amounts of data; and its decision processes are not transparent.
News aggregation software, for example, had long relied on rudimentary AI to curate articles based on people’s requests. Then it evolved to analyze behavior, tracking the way people clicked on articles and the time they spent reading, and adjusting the selections accordingly. Next it aggregated individual users’ behavior with the larger population, particularly those who had similar media habits. Now it is incorporating broader data about the way readers’ interests change over time, to anticipate what people are likely to want to see next, even if they have never clicked on that topic before. Tomorrow’s AI aggregators will be able to detect and counter “fake news” by scanning for inconsistencies and routing people to alternative perspectives.
AI applications in daily use include all smartphone digital assistants, email programs that sort entries by importance, voice recognition systems, image recognition apps such as Facebook Picture Search, digital assistants such as Amazon Echo and Google Home and much of the emerging Industrial Internet. Some AI apps are targeted at minor frustrations — DoNotPay, an online legal bot, has reversed thousands of parking tickets — and others, such as connected car and language translation technologies, represent fundamental shifts in the way people live. A growing number are aimed at improving human behavior; for instance, GM’s 2016 Chevrolet Malibu feeds data from sensors into a backseat driver–like guidance system for teenagers at the wheel.
Despite all this activity, the market for AI is still small. Market research firm Tractica estimated 2016 revenues at just $644 million. But it expects hockey stick-style growth, reaching $15 billion by 2022 and accelerating thereafter. In late 2016, there were about 1,500 AI-related startups in the U.S. alone, and total funding in 2016 reached a record $5 billion. Google, Facebook, Microsoft, Salesforce.com and other tech companies are snapping up AI software companies, and large, established companies are recruiting deep learning talent and, like Monsanto, buying AI companies specializing in their markets. To make the most of this technology in your enterprise, consider the three main ways that businesses can or will use AI:
Assisted intelligence, now widely available, improves what people and organizations are already doing.
Augmented intelligence, emerging today, enables organizations and people to do things they couldn’t otherwise do.
Autonomous intelligence, being developed for the future, creates and deploys machines that act on their own.
Many companies will make investments in all three during the next few years, drawing from a wide variety of applications (see Exhibit 1). They complement one another but require different types of investment, different staffing considerations and different business models.
Assisted intelligence amplifies the value of existing activity. For example, Google’s Gmail sorts incoming email into “Primary,” “Social” and “Promotion” default tabs. The algorithm, trained with data from millions of other users’ emails, makes people more efficient without changing the way they use email or altering the value it provides.
Assisted intelligence tends to involve clearly defined, rules-based, repeatable tasks. These include automated assembly lines and other uses of physical robots; robotic process automation, in which software-based agents simulate the online activities of a human being; and back-office functions such as billing, finance and regulatory compliance. This form of AI can be used to verify and cross-check data — for example, when paper checks are read and verified by a bank’s ATM. Assisted intelligence has already become common in some enterprise software processes. In “opportunity to order” (basic sales) and “order to cash” (receiving and processing customer orders), the software offers guidance and direction that was formerly available only from people.
The Oscar W. Larson Co. used assisted intelligence to improve its field service operations. This is a 70-plus-year-old family-owned general contractor, which, among other services to the oil and gas industry, provides maintenance and repair for point-of-sales systems and fuel dispensers at gas stations. One costly and irritating problem is “truck rerolls”: service calls that have to be rescheduled because the technician lacks the tools, parts or expertise for a particular issue. After analyzing data on service calls, the AI software showed how to reduce truck rerolls by 20%, a rate that should continue to improve as the software learns to recognize more patterns.
Assisted intelligence apps often involve computer models of complex realities that allow businesses to test decisions with less risk. For example, one auto manufacturer has developed a simulation of consumer behavior, incorporating data about the types of trips people make, the ways those affect supply and demand for motor vehicles and the variations in those patterns for different city topologies, marketing approaches and vehicle price ranges. The model spells out more than 200,000 variations for the automaker to consider and simulates the potential success of any tested variation, thus assisting in the design of car launches. As the automaker introduces cars and the simulator incorporates the data on outcomes from each launch, the model’s predictions will become ever more accurate.
AI-based packages of this sort are available on more and more enterprise software platforms. Success with assisted intelligence should lead to improvements in conventional business metrics such as labor productivity, revenues or margins per employee and average time to completion for processes. Much of the cost involved is in the staff you hire, who must be skilled at marshaling and interpreting data. To evaluate where to deploy assisted intelligence, consider two questions: What products or services could you easily make more marketable if they were more automatically responsive to your customers? Which of your current processes and practices, including your decision-making practices, would be more powerful with more intelligence?
Augmented intelligence software lends new capability to human activity, permitting enterprises to do things they couldn’t do before. Unlike assisted intelligence, it fundamentally alters the nature of the task, and business models change accordingly.
For example, Netflix uses machine learning algorithms to do something media has never done before: suggest choices customers would probably not have found themselves, based not just on the customer’s patterns of behavior but on those of the audience at large. A Netflix user, unlike a cable TV pay-per-view customer, can easily switch from one premium video to another without penalty, after just a few minutes. This gives consumers more control over their time. They use it to choose videos more tailored to the way they feel at any given moment. Every time that happens, the system records that observation and adjusts its recommendation list — and it enables Netflix to tailor its next round of videos to user preferences more accurately. This leads to reduced costs and higher profits per movie, and a more enthusiastic audience, which then enables more investments in personalization (and AI). Left outside this virtuous circle are conventional advertising and television networks. No wonder other video channels, such as HBO and Amazon, as well as recorded music channels such as Spotify, have moved to similar models.
Over time, as algorithms grow more sophisticated, the symbiotic relationship between human and AI will further change entertainment industry practices. The unit of viewing decision will probably become the scene, not the story; algorithms will link scenes to audience emotions. A consumer might ask to see only scenes where a Meryl Streep character is falling in love, or to trace a particular type of swordplay from one action movie to another. Data accumulating from these choices will further refine the ability of the entertainment industry to spark people’s emotions, satisfy their curiosity and gain their loyalty.
Another current use of augmented intelligence is in legal research. Though most cases are searchable online, finding relevant precedents still requires many hours of sifting through past opinions. Luminance, a startup specializing in legal research, can run through thousands of cases in a very short time, providing inferences about their relevance to a current proceeding. Systems like these don’t yet replace human legal research. But they dramatically reduce the rote work conducted by associate attorneys, a job rated as the least satisfying in the U.S. Similar applications are emerging for other types of data sifting, including financial audits, interpreting regulations, finding patterns in epidemiological data and (as noted above) farming.
To develop applications like these, you’ll need to marshal your own imagination to look for products, services or processes that would not be possible at all without AI. For example, an AI system can track a wide number of product features, warranty costs, repeat purchase rates and more general purchasing metrics, bringing only unusual or noteworthy correlations to your attention. Are a high number of repairs associated with a particular region, material or line of products? Could you use this information to redesign your products, avoid recalls or spark innovation in some way?
The success of an augmented intelligence effort depends on whether it has enabled your company to do new things. To assess this capability, track your margins, innovation cycles, customer experience and revenue growth as potential proxies. Also watch your impact on disruption: Are your new innovations doing to some part of the business ecosystem what, say, ride-hailing services are doing to conventional taxi companies?
You won’t find many off-the-shelf applications for augmented intelligence. They involve advanced forms of machine learning and natural language processing, plus specialized interfaces tailored to your company and industry. However, you can build bespoke augmented intelligence applications on cloud-based enterprise platforms, most of which allow modifications in open source code. Given the unstructured nature of your most critical decision processes, an augmented intelligence application would require voluminous historical data from your own company, along with data from the rest of your industry and related fields (such as demographics). This will help the system distinguish external factors, such as competition and economic conditions, from the impact of your own decisions.
The greatest change from augmented intelligence may be felt by senior decision makers, as the new models often give them new alternatives to consider that don’t match their past experience or gut feelings. They should be open to those alternatives, but also skeptical. AI systems are not infallible; just like any human guide, they must show consistency, explain their decisions and counter biases, or they will lose their value.
Very few autonomous intelligence systems — systems that make decisions without direct human involvement or oversight — are in widespread use today. Early examples include automated trading in the stock market (about 75% of Nasdaq trading is conducted autonomously) and facial recognition. In some circumstances, algorithms are better than people at identifying other people. Other early examples include robots that dispose of bombs, gather deep-sea data, maintain space stations and perform other tasks inherently unsafe for people.
The most eagerly anticipated forms of autonomous intelligence — self-driving cars and full-fledged language translation programs — are not yet ready for general use. The closest autonomous service so far is Tencent’s messaging and social media platform WeChat, which has close to 800 million daily active users, most of them in China. The program, which was designed primarily for use on smartphones, offers relatively sophisticated voice recognition, Chinese-to-English language translation, facial recognition (including suggestions of celebrities who look like the person holding the phone) and virtual bot friends that can play guessing games. Notwithstanding their cleverness and their pioneering use of natural language processing, these are still niche applications, and still very limited by technology. Some of the most popular AI apps, for example, are small, menu- and rule-driven programs, which conduct fairly rudimentary conversations around a limited group of options.
Despite the lead time required to bring the technology further along, any business prepared to base a strategy on advanced digital technology should be thinking seriously about autonomous intelligence now. The Internet of Things will generate vast amounts of information, more than humans can reasonably interpret. In commercial aircraft, for example, so much flight data is gathered that engineers can’t process it all; thus, Boeing has announced a $7.5 million partnership with Carnegie Mellon University, along with other efforts to develop AI systems that can, for example, predict when airplanes will need maintenance. Autonomous intelligence’s greatest challenge may not be technological at all — it may be companies’ ability to build in enough transparency for people to trust these systems to act in their best interest.
As you contemplate the introduction of artificial intelligence, articulate what mix of the three approaches works best for you.
Are you primarily interested in upgrading your existing processes, reducing costs and improving productivity? If so, then start with assisted intelligence, probably with a small group of services from a cloud-based provider.
Do you seek to build your business around something new — responsive and self-driven products, or services and experiences that incorporate AI? Then pursue an augmented intelligence approach, probably with more complex AI applications resident on the cloud.
Are you developing a genuinely new technology? Most companies will be better off primarily using someone else’s AI platforms, but, if you can justify building your own, you may become one of the leaders in your market.
The transition among these forms of AI is not clean-cut; they sit on a continuum. In developing their own AI strategy, many companies begin somewhere between assisted and augmented, while expecting to move toward autonomous eventually (see Exhibit 2).
Though investments in AI may seem expensive now, the costs will decline over the next 10 years as the software becomes more commoditized. “As this technology continues to mature,” writes Daniel Eckert, a managing director in emerging technology services for PwC US, “we should see the price adhere toward a utility model and flatten out. We expect a tiered pricing model to be introduced: a free (or freemium model) for simple activities, and a premium model for discrete, business-differentiating services.”
AI is often sold on the premise that it will replace human labor at lower cost — and the effect on employment could be devastating, though no one knows for sure. Carl Benedikt Frey and Michael Osborne of Oxford University’s engineering school have calculated that AI will put 47% of the jobs in the U.S. at risk; a 2016 Forrester research report estimated it at 6%, at least by 2025. On the other hand, Baidu Research head (and deep learning pioneer) Andrew Ng recently said, “AI is the new electricity,” meaning that it will be found everywhere and create jobs that weren’t imaginable before its appearance.
At the same time that AI threatens the loss of an almost unimaginable number of jobs, it is also a hungry, unsatisfied employer. The lack of capable talent — people skilled in deep learning technology and analytics — may well turn out to be the biggest obstacle for large companies. The greatest opportunities may thus be for independent businesspeople, including farmers like Jeff Heepke, who no longer need scale to compete with large companies, because AI has leveled the playing field.
It is still too early to say which types of companies will be the most successful in this area — and we don’t yet have an AI model to predict it for us. In the end, we cannot even say for sure that the companies that enter the field first will be the most successful. The dominant players will be those that, like Climate Corp., Oscar W. Larson, Netflix and many other companies large and small, have taken AI to heart as a way to become far more capable, in a far more relevant way, than they otherwise would ever be.
Softening prices, little or no organic growth and increased competition have characterized most of the commercial insurance environment in recent years. These factors and a relatively benign cat environment continue to attract new types of capital providers (e.g., hedge funds, pension funds, foreign investors, capital markets) looking to diversify their investment portfolios with uncorrelated insurance assets.
Limited organic growth opportunities also have led to a broad consolidation of distributors, with an increasingly large number of private equity-backed brokers looking for short-term gains and opportunities to reduce systemic inefficiency. In turn, this has led to significant carrier investments in automation to facilitate effective and efficient straight-through processing (STP).
More specific responses to market conditions from commercial insurance constituents include:
Distributor response – Distributors are increasingly looking for ways to (1) negotiate more aggressively on individual transactions (e.g., appetite exceptions, non-standard terms and conditions, pricing), (2) operate more efficiently (e.g., customized processes, only partial completion of applications) and (3) exert their bargaining power to gain higher commissions and other sources of revenue (e.g., access to market intelligence).
In addition, brokers are becoming increasingly organized. They are looking to 1) reduce the number of carriers with whom they place business in favor of ones that have a broad underwriting appetite and are easy to do business with and 2) exit the service arena, especially on small commercial accounts where margins are already extremely thin.
Carrier response – Carriers are intensifying their efforts to compete for a “top three” position with distributors by attempting to (1) be easier to do business with (both in terms of technology and personal relationships), (2) increase product specialization and related underwriting expertise, (3) increase their appetite for more hazardous risk and 4) (as a less favored option) lower rates and pricing.
Although more and more carriers have invested in automated underwriting and pricing, broker/agent expectations are only increasing. They not only want to clearly understand a carrier’s underwriting appetite, they also want to get near-real-time quotes on the majority of standard risks without extensive manual data entry on their side.
For now, carriers have avoided being “spread-sheeted” by using proprietary agent portals to increase ease of business interactions, rather than directly integrating with agency management systems and comparative raters. Distributors have not yet increased their demands for the latter two, recognizing that they could lead to a commission squeeze or even losing their appointment if the portability of their book declines with a given carrier.
Customer response – Last but not least, customers’ behaviors and expectations are changing, too. They are becoming more comfortable researching business insurance online, and expect their shopping experience to reflect what they see in personal insurance. However, they are still turning to an agent (whether digitally or in person) to confirm their purchase decision and complete the deal. This is especially the case when businesses mature and risk management becomes more critical to their success.
As all this has been happening, artificial intelligence (AI) has matured significantly, demonstrating that it can markedly improve existing STP. We describe below the AI technologies – including robotic process automation, natural language processing and machine learning – that can increase commercial insurance’s efficiency and effectiveness and thereby benefit investors, distributors and carriers themselves.
Availability and access to large volumes of data, increasing processing power, cloud computing, open-source software and advances in algorithms have fueled the rise of AI from academic curiosity to commercial viability.
The next generation of straight-through processing
Although many carriers are already heavily automated, their initial focus has largely been on automated underwriting and pricing. This has left considerable manual intervention in the issuance process, post-bind audits and other downstream transactions. All of these can be streamlined to further drive down costs. Once carriers move to truly mechanized underwriting, the next step will be to embed third-party data feeds and advanced analytics to drive straight-through processing (STP) of risks.
For example, imagine a small business owner being able to enter just four pieces of information (e.g., business name, business address and owner’s name and DOB) on a policy application and receiving a real-time business insurance quote with the option to immediately purchase and electronically receive policy documents. Furthermore, imagine this approach having no impact on underwriting quality or manual back-end processing requirements for the carrier. Integrating AI techniques and additional internal and external data sources into small business processing have the potential to make this a reality.
A combination of leveraging internal data from prior quotes and policies, integrating external structured data feeds and mining a business’s website and social media presence could provide carriers with enough information to determine a business’s operations, applicable class codes, property details, employment and payroll and other key risk characteristics to underwrite and price low-complexity risks. In cases where more information is needed, dynamic question sets with user-friendly inputs could streamline the application process without sacrificing underwriting quality.
How AI can improve straight-through processing
In addition to immediate cost improvements, commercial carriers
that leverage internal and external data resources and apply AI to commercial processing can benefit from reduced turn-around time, better and more consistent decision-making and improved agent/customer satisfaction.
The carriers that are the first to adopt the latest in AI-enabled straight-through processing will be preferred by their existing agencies, as well as be able to pursue alternative distribution channels that feature a more streamlined, user-friendly acquisition process that accommodates less sophisticated users.
Some of the most promising AI techniques that can help insurers improve STP include:
Robotic process automation (RPA) is an area of AI that could increase STP efficiency and bring down costs at acceptable level of increased risk. RPA automates data entry, third-party data integration, form filling and data validation. More advanced process-mining techniques use machine learning to infer business processes from transaction logs, web and call center logs, email, and workflow logs. They profile the time it takes for different steps of the quote-to-issue process to be fulfilled and, to streamline the process, plot a distribution that enables the identification of outliers. They also track exceptions, and the reasons for them, thereby enabling greater efficiency. RPA is also tracking conformance and compliance with established standards, thereby leading to more consistent and compliant service delivery.
Machine learning is building routing logic and underwriting-related models. For example, a detailed analysis of a commercial book of business over time can identify the need for no- touch, medium-touch or high-touch interaction models. This categorization enables better routing across multi-segment (i.e., small commercial, middle market and large commercial) insurers. In addition, machine learning can inform a wide variety of predictive models.
Using open source technology, PwC has built natural language processing engines that continuously evaluate a large number of news and social media sources and report on key concepts.
Commercial insurers and brokers can use this ontology of “key concepts” to traverse the output, identify drivers of specific risks and refer to articles related to these risks. By indicating the relevance of articles (e.g., via a thumbs up or thumbs down) insurers can “train” the natural language engine to look for specific sources and type of articles. As the system learns over time, it can graph trending topics, the sectors and companies associated with certain risks and the underlying impacts if the risks develop adversely. We also have built a question-answer engine that allows risk experts to make natural language inquiries and retrieve relevant reports and documents to conduct further analysis. With natural language generation, the engine also can create risk profiles for senior management’s consumption.
By coupling deep learning systems with natural language processing, PwC has been able to create powerful risk analysis enablers that enhance and speed up emerging risk analyses. When analyzing text from news sources or social media sources, the system needs to understand the context under which certain words are used. For example, a common word like “run” has more than 645 meanings according to the Oxford English Dictionary. “Deep Learning” or neural network-based machine learning systems can actually capture the context of words within sentences, sentences within documents and documents within a collection of documents.
In closing, even with their increased focus on ease of doing business, there is still much room for carriers to improve. There currently is a clear opportunity for prescient and active carriers to separate themselves from the pack, but doing so will require a competitive mindset that has not traditionally defined the industry. Small and medium commercial carriers must find ways to improve their cost structures to compete profitably in the long term. AI-enabled solutions offer some of the most promising ways to do this.
New investors in the commercial insurance market are increasingly looking for short-term gains and greater efficiencies from the industry.
Moreover, distributors are looking for greater ease of doing business with commercial carriers and have demonstrated a willingness to favor the ones that can meet their expectations.
Commercial carriers have automated quoting in an attempt to facilitate effective straight-through processing. This has increased efficiencies, which has benefited investors and helped improve the distributor experience.
However, many manual processes and inefficiencies still remain. Once carriers move to truly mechanized underwriting, the next step will be to embed third-party data feeds and advanced analytics to drive straight through processing of risks. Recent developments in artificial intelligence (AI) can help carriers do this.
Insurance is the industry most affected by disruptive change, according to the percentage of CEOs who are extremely concerned about the threats to their growth prospects from the speed of technological change, changing customer behavior and competition from new market entrants.
Insurers know they need to innovate to remain competitive. In fact, 67% of insurance respondents to PwC’s 2017 CEO Survey see creativity and innovation as very important to their organizations, ahead of other financial services sectors and the CEO Survey population as a whole. And, insurance CEOs noted that the area they would most like to strengthen to capitalize on growth opportunities is digital and technological capabilities, followed by customer experience (reflecting the connections between the two).
However, the industry’s traditional conservatism and the dizzying pace of technological change has made it difficult to change. As a result, most insurers are looking outside the industry – typically in the insurtech space (e.g., drones, sensors, internet of things (IoT)) – for the best ways to improve their systems, processes and products. And there is no doubt that industry stakeholders think insurtech has real promise: Annual investment in insurtech startups has increased fivefold over the past three years, with cumulative funding reaching $3.4 billion since 2010, based on the companies that PwC’s DeNovo platform follows.
To facilitate a diverse approach to identifying opportunities and potential partners from different industries and specialty areas, an enterprise innovation model (EIM) is table stakes. An EIM facilitates:
New product and service development: Being active in insurtech can help insurers discover emerging coverage needs and risks that require new insurance products and services. As a result, they can improve their product portfolio strategy and design of new risk models.
Market exploration and discovery: Prescient insurers actively monitor new trends and innovations, and some have even established a presence in innovation hotspots (e.g., Silicon Valley) where they can directly learn about the latest developments in real-time and initiate innovation programs.
Partnerships that drive new solutions: Exploration typically leads to the development of potential use cases that address specific business challenges. Insurers can partner with startups to build pilots to test and deploy in the market.
Contributions to insurtech’s growth and development: As we describe below, venture capital and incubator programs can play an important role in key innovation efforts. Established insurers that clearly identify areas of need and opportunity can work with startups to develop appropriate solutions.
Most insurers are looking outside the industry for the best ways to improve their systems, processes and products.
Maintaining awareness, influencing the market and identifying the right partners
To ensure an organization’s innovation efforts are in sync with – or even driving – the latest developments in the market, an EIM needs a formalized yet agile process for identifying and incorporating best practices.
Dedicated assessment of insurtech advancements can allow insurers to identify and promote best practices and key technologies. Moreover, maintaining a close connection with the insurtech market can help a company develop its external knowledge and relationships with innovators. Through this process, insurers can identify potential partners that can help them understand evolving technologies and their applications, and even contribute to developing the capabilities they desire.
With a deeper understanding of the market, capabilities and key players, insurers can be better positioned to facilitate innovation, ideation and design. While some fintech companies already have compelling insurance applications, insurers have a great opportunity to identify and design new potential use cases.
Fast prototyping is key to quickly creating minimally viable products (MVP) and bringing ideas to life. Early-stage startups develop and deploy full-functioning prototypes in near real time and go to market with solutions that evolve with market feedback. The development cycle is shortened, which allows startups to quickly deliver solutions and tailor future releases based on usage trends and feedback and to accommodate more diverse needs. Established insurers can follow the same approach or can partner with existing startups that have a MVP to help them to move to the next stage, scaling.
The ways to accomplish all of this vary based on how the organization plans to source new opportunities and ideas, how it plans on executing innovation and how it plans to deploy new products and services. The following graphic provides examples of EIMs by primary function.
The innovation center
The innovation center (also named “lab” or “hub”) is a structure at a corporate level that bridges external innovation with business unit needs and innovation opportunities. It relies on internal subject matter experts and innovation champions to ignite and drive innovation initiatives at a business unit level. With this model, innovative new products and services go to market under the company’s brand.
The innovation hub provides an outside-in view while promoting innovation internally. With this model, the company dedicates a team to constantly monitor trends and market activity, build and maintain relationships with key insurtech players, identify potential future scenarios and determine new partnership opportunities.
The hub should be managed through business units to effectively innovate (i.e., building prototypes and scaling models). Execution is a key success factor, and we recommend insurers consider complementary innovation models to help promote positive outcomes.
Regardless of the model they use, we recommend that insurers of all sizes consider developing an innovation center and create an external connection based on potential future scenarios.
An incubator can drive innovation from idea to end product by identifying new opportunities and developing related solutions. Although it does require a significant investment of both money and resources, it has proven especially effective in addressing complex problems and devising new approaches to them.
Although the incubator can be internal, external structures typically create unique development environments and attract necessary talent. Via an external approach, ideas come mostly from outside the company and a panel of internal or external innovation specialists provide high-level guidance and approval for the innovation the company is seeking through the incubator.
Although the incubator initially drives innovation, business units typically become involved during the development process. They have an important role, especially when planning to deploy new solutions within the organization. The incubator can wind up as a start-up that can go to the market under its own name.
One of the main strengths of the incubator model is that it facilitates execution. It holds an idea until a prototype is developed and a minimally viable product is available. The gradual involvement of business units during the process enables the model to adequately scale. Upon adoption by its future owner, the incubator and business units can address any related challenges related to operating capacity, cyber risk, regulation and other issues.
Strategic venture capital (SVC)
The SVC model offers the opportunity to participate via stake or acquisition in relevant insurtech-related players. This is a way to influence and shape the development of specific startups (e.g. pushing them to solve specific problems) and acquire key capabilities and talent, and as a way to derive value from strategic investments.
In the SVC model, the company establishes a new ventures division, which sources ideas from the outside. The company provides funding and support for equity, while a SVC from this new structure explores, identities and evaluates solutions and markets new ventures under its own brand. The funds thatSVC invests in a startup help new players augment their capabilities and scale their business model. This could lead to potential market joint ventures, acquisitions or other deals to monetize the initial investment.
Established insurers with SVC arms are usually leaders in specific market segments and therefore leverage their experience and knowledge to select key ventures. To become more active with insurtech, these structures can be linked to innovation centers, thereby allowing companies to connect ventures with business units.
Instead of choosing one model over the other, we propose one that combines key elements from each. Companies can select elements based on their need for external innovation, the availability of talent, their ability to execute and the amount of investment the organization is willing to commit.
EIM operating options
EIM operating characteristics
Bridging the cultural divide
Complicating the need to innovate is the fact that an insurer’s culture often influences an external company’s decision about partnering with it. In fact, according to our 2016 Global
FinTech Survey, more than half of fintechs see differences in management and culture as a key challenge in working with insurers. Insurers also realize this, and 45% of insurance companies agree that this is a major challenge.
Accordingly, insurers will need to assess the availability and compatibility of existing resources and determine how and where they can find what may not currently be available. By clearly articulating the organization’s needs, defining explicit roles and establishing a model for enterprise innovation, an insurer can address any underlying concerns it may have about partnerships.
While insurers can create internal structures to support innovation, most of them will have to enlist external resources in one way or another. In fact, we expect many talented professionals without insurance-specific skills will be the ones who wind up driving innovation.
Attracting and developing innovators
Insurers can create internal structures to support innovation, but – as EIMs stipulate – success ultimately depends on having the right talent. And, most insurers will have to enlist external resources – ones who have an entrepreneurial mindset and who are well-connected to insurtech – in one way or another.
How does a company attract and retain this kind of talent? There are four primary ways:
Acquire the new talent from start-ups. This works well if the acquired company keeps running its business under its own start-up rules, away from the acquirer’s bureaucracy. Otherwise, if there is too much acquirer interference, then retention will be a challenge in a market that covets innovators.
Attract the talent directly from the market. This option typically requires a new mindset from the hiring company in terms of business role, work environment and even location. Establishing a presence in relevant innovation hotspots will help make an offer more attractive, facilitate external connections and demonstrate the insurer’s commitment to letting innovators be free to innovate.
Partner with startups, technology vendors, universities, researchers and other proven innovators. This option represents a major opportunity because it enables the insurer to create the connections to and formal partnerships with new talent. However, while identifying desired capabilities is relatively easy, there will need to be strong alignment of purpose between the organization and the new partners for the relationship to work. In this case, the Innovation Hub should be the most helpful model.
Grow the talent. This option is probably the least disruptive because it doesn’t require external changes. Large organizations have the opportunity to discover talent within their structures. But, the organization will have to ascertain and leverage the mentality and professional background of employees in many different ways. Gamification, internal collaboration groups and other resources can help in the search for potential in-house innovators, but most companies will need a more sophisticated staffing model to develop this talent (e.g., having specific development plans and offering “external” experiences in projects and with partners).
Complementing these options is the insurance industry leadership’s advocacy of new methods to foster change in employee skill sets. According to insurance respondents to PwC’s 2017 CEO Survey,
61% are exploring the benefits of humans and machines working together (considerably higher than any other FS sector), and
49% are considering the impact of artificial intelligence on future skills needs (also considerably higher than any other FS sector).
In response to this rapidly changing environment, incumbent insurers are approaching insurtech in various ways, prominently through joint partnerships or startup programs. But whatever strategy an organization pursues, insurtech’s main impact will be new business models that create challenges for market players and other industry stakeholders (e.g., regulators). In this environment, insurers will need to move away from trying to control all parts of their value chain and customer experience through traditional business models, and instead move toward leveraging their trusted relationships with customers and their extensive access to client data.
For many traditional insurers, this approach will require a fundamental shift in identity and purpose. The new norm will involve turning away from a linear product push approach, to a customer-centric model in which insurers are facilitators of a service that enables clients to acquire advice and interact with all relevant actors through multiple channels. By focusing on incorporating new technologies into their own architecture, traditional insurers can prepare themselves to play a central role in the new world in which they will operate at the center of customer activity and maintain strong positions even as innovations alter the marketplace.
To effectively develop these new business models and capabilities and establish mutually beneficial insurtech relationships, established insurers will need to start with a well-thought-out innovation strategy that incorporates the following:
An effective enterprise innovation model (EIM) will take into account the different ways to meet an organization’s various needs and help it make innovative breakthroughs. The model or combination of models that is most suitable for an organization will depend on its innovation appetite, the type of partnerships it desires and the capabilities it needs. EIMs feature three primary approaches to support corporate strategy, partnering via innovation centers (or hubs), building capabilities via incubators and buying capabilities via a strategic ventures division. Companies can select elements from each of these models based on their need for external innovation, the availability of talent, their ability to execute and the amount of investment the organization is willing to commit.
Even though insurers can create the internal structures that support innovation, most of them will have to enlist external resources in one way or another. Accordingly, they will need to assess the availability and compatibility of existing talent and determine how and where they can find what may not currently be available. Much like with enterprise innovation models, there are certain ways (often in combination) that insurers can locate and obtain the resources they need, including acquiring it, trying to attract it, partnering and growing it internally.
This is Part 3 of a 3-part series. Part 1 can be found here; Part 2, here.
In our first blog post on artificial intelligence (AI), we outlined the challenges of defining AI, and in our second blog post, we described how ubiquitous AI is becoming, defining it as “ubiquitous intelligence.” In this post, we define the continuum of AI as “AAAI”: assisted, augmented and autonomous intelligence.
AI as Assisted Intelligence
Over the past couple of decades, AI has replaced many of the repetitive and standardized tasks done by humans. For example, industrial robots are tackling many manufacturing tasks. Similarly, many administrative tasks such as taking meeting minutes, answering phones and searching for information are all done by some form of an automated system. We call this type of automation — where the AI is assisting humans to do the same tasks faster or better — assisted intelligence. The humans are still making some of the key decisions, but the AI is executing the tasks on their behalf. The decision rights are solely with the humans.
AI as Augmented Intelligence
We are just now moving to the next stage of augmented intelligence, where humans and machines learn from each other and redefine the breadth and depth of what they do together. For example, in a recent client engagement, we carried out 200,000 go-to-market scenarios generated by an AI system for a service introduction. This provided the human decision-makers with a high degree of granularity and specificity regarding the assumptions, future projections and impact of the new service.
While the system learned a lot and modeled the ecosystem, the humans saw the sensitivities and feedback involved in market adoption. Under these circumstances, the human and the machine share the decision rights. In addition, unlike assisted intelligence, in augmented intelligence, the nature of the task fundamentally changes. On a spectrum ranging from no automation to total autonomous operation, each sector, company and individual will set the appropriate level of machine augmentation. Over time, the dial might move more toward totally autonomous, or it might stay somewhere in between.
AI as Autonomous Intelligence
Lastly, we see autonomous intelligence, in some cases, where adaptive/continuous systems take over. They will do so only after the human decision-maker starts trusting the machine (e.g., fully autonomous self-driving cars), or when the cycle time of decision making is so fast that having the human in the loop is a liability (e.g., automated trading). In autonomous intelligence, the decision rights are with the machine and are fundamentally different from assisted intelligence.
The decision to move from augmented intelligence to autonomous intelligence will largely be in our hands and will be made based on a number of different factors — including the speed of human decision making, the technical feasibility of making autonomous decisions, the cost of building solutions and the trust we place in these solutions.
As enterprises contemplate the introduction of AI across their functional areas, it helps to clearly articulate which stage of AI they are aiming for. Are they merely automating repetitive tasks and providing assisted intelligence? Are they fundamentally changing the nature of work by having humans and machines collaborate with each other to make decisions with augmented intelligence? Or are they delegating all decision making with autonomous intelligence?