Tag Archives: natural language processing

Best AI Tech for P&C Personal Lines

Artificial intelligence technologies are everywhere. The great leap forward in AI over the past decade has come along with an explosion of new tech companies, AI deployment across almost every industry sector and AI capabilities behind the scenes in billions of intelligent devices around the world. What does all of this mean for the personal lines insurance sector? SMA answers this question in a new research report, “AI in P&C Personal Lines: Insurer Progress, Plans, and Predictions.”

The first step toward answering this question is to understand that AI is a family of related technologies, each with its own potential uses and insurance implications. The key technologies relevant for P&C insurance are machine learning, computer vision, robotic process automation, user interaction technologies, natural language processing and voice technologies. It’s a challenge to sort through all these technologies, the insurtech and incumbent providers that offer AI-based solutions and where each insurer will benefit most from applying AI.

The overall value rankings indicate that user interaction technologies fueled by AI are at the top of the list for personal lines insurers. Every insurer has activity underway, mostly by leveraging chatbots for interactions with policyholders and agents or using machine learning for guided data collection during the application process. Insurers see high potential for transformation in policy servicing, billing and claims – areas where routine interactions can be automated.

Robotic process automation is in broad use across personal lines, although the RPA technology is viewed by many as more tactical. There is high value related to streamlining operations and reducing costs, but most wouldn’t put it in the innovative category.

Machine learning and computer vision have great potential for personal lines in both underwriting and claims. The combination of computer vision and ML technologies applied to aerial imagery is already becoming a common way to provide property characteristics and risk scores for underwriting. Likewise, images from satellites, fixed-wing aircraft and drones are frequently used for NATCAT situations. And AI technologies will be increasingly applied to these images for response planning.

There are many other examples. But for the purposes of this blog, the main question – which technologies are most valuable – has been answered. AI-based user interface (UI) technologies, machine learning (ML) and computer vision demonstrate the best combination of high value today and transformation potential for the long term.

But perhaps the more important question is not which technologies are valuable, but rather where AI technologies are most valuable in the enterprise. The short answer is that there are so many potential value levers and so many unique aspects to different business areas and lines of business that it is difficult to select just a couple of high-value areas. That said, it is relatively apparent that underwriting and claims both present major opportunities, and activities are already underway there. There are great possibilities for AI in inspections, property underwriting, triage, fraud, CAT management, automated damage assessment, predictive reserving and other specific areas.

See also: Stop Being Scared of Artificial Intelligence

There is no shortage of opportunities for AI in personal lines. Fortunately, there are increasing numbers of tech solutions in the market and growing expertise in the industry involving AI technologies and how to apply them. Ultimately, we expect to see a pervasive use of AI technologies throughout the insurance industry. Some will become table stakes. Others will define the winners in the new era of insurance.  

Sentiment Analytics Can Drive Growth

Insurers have a near-constant stream of unstructured data at their disposal that can be used to drive growth by improving policyholder retention and identifying cross-sell and upsell opportunities. One of the challenges for insurers is sorting through this mountain of unstructured data quickly to gain an accurate understanding of the sentiment of their customers in real time.

The last few years have seen sentiment analytics become a critical component of customer feedback strategies for companies of all sizes. Sentiment analytics uses a combination of natural language processing (NLP), machine learning (ML) and deep text analytics to bring out the nuances hidden in the text.

Sentiment is easier to translate and analyze than it is to express. Sentiment analytics, also referred to as opinion mining, is a technique to abstract the underlying sentiment from textual data. Usually, this customer feedback is unstructured data flowing in from multiple channels, such as:

  • Voice messages
  • Claims
  • Adjuster notes
  • Medical records
  • DMV reports
  • Surveys
  • Underwriter notes
  • Email
  • Call center logs
  • Social media posts

The idea is to understand not only the nature of the feedback but also to derive context out of it. However, sentiments are complicated. Analyzing sentiments, even more so.

Domain-Specific Sentiment Analytics

The complexity of spoken language makes it difficult if not impossible to derive sentiment accurately every time. Teaching a machine to understand such things as tonality, cultural lingo and slang, or the ability to discount grammatical errors, and comprehend rhetoric such as irony or sarcasm are all difficult at present. Existing sentiment analytics tools are not equipped to identify the true sentiment of these types of dialogues.

Although sarcasm is a problematic form of language to detect, there are other complex statements that machines are learning to comprehend. Consider the following statement: “Rocketz Insurance Company has always offered me great pricing, but at times I have not been happy with their response time for questions about my policy.” This sentence has a negative as well as a positive connotation, and sentiment analytics come into play. The first part of the sentence can be identified as a positive feeling, and the other half is identified as negative.

See also: 3-Step Approach to Big Data Analytics  

Sentiment analytics specific to industry lines play a key role. The accuracy of identifying the sentiment of data can be increased by training the system (machine learning) on a specific domain, such as the insurance industry. For example, the terms “garaging” and “towing” have a greater meaning in the insurance industry as opposed to, for example, manufacturing. Therefore, if a client makes a comment about either, it would have more meaning for insurance than other industries.

Driving Growth

In the insurance industry, sentiment analytics can be used in a multitude of ways that directly affect business, such as:

  1. Improving retention rates
  2. Identifying cross-sell/upsell opportunities
  3. Identifying trends

Improving Retention Rates

Having the ability to quickly and easily identify the sentiment of policyholders whose auto policy will renew in 60 days or less is a good example. Let’s say we have 1,000 auto policies that are up for renewal by the end of 2019, and the priority for the renewal team is to contact policyholders who are “detractors.” The sentiment of their conversations and interactions, regardless of the channel, has a low score. The challenge for the renewal team is: Who do they contact first? Are all “detractors” equal in their dissatisfaction with their auto policy? And what may be the root cause or causes of their dissatisfaction?

The renewal team can target these policyholders with a strategy to retain these policyholders with insight on why the policyholder may be unhappy before contacting them. Maybe it was a bad claim experience, or they are unhappy with their premium and started shopping elsewhere. It’s not enough to simply understand who a “detractor” is; you need to understand why.

Identifying Cross-sell/Upsell Opportunities

Using the example of our 1,000 auto policies that are up for renewal, what about the policyholders who are “promoters” and happy with their auto coverage? This is an ideal time for the renewal team to contact these policyholders and thank them for their business as a minimum. But is there a cross-sell or upsell opportunity here? For the renewal team focusing on this segment of policyholders, it would be helpful to have some idea why their sentiment is high before contacting them with a potential offer.

Identifying Trends

In addition to identifying immediate opportunities that can be acted on, sentiment analytics can help insurers understand trends such as:

  1. The sudden demand for a product type
  2. The like or dislike of a specific customer experience

Many homeowners are considering cyber insurance to protect themselves from identity theft and invasion of privacy. Sentiment analytics can provide insurers who currently do not offer cyber insurance a heads-up that maybe they should consider offering cyber insurance as a cross-sell/upsell opportunity.

See also: Predictive Analytics: Now You See It….  

Everyone wants to identify and correct bad customer experiences, especially a bad claims experience. What about good customer experiences? Going back to our auto policy example, something as simple as having the ability to easily obtain an auto insurance card online, or easily reach a customer service representative. can be a positive customer experience that sales and marketing may want to promote.

Conclusion

Sentiment analytics can help insurers sort through a continuous stream of unstructured data to identify opportunities for increasing revenue and identifying trends. Currently, sentiment analytics is not perfect, but focusing on a specific domain such as the insurance industry will increase accuracy.

Sentiment analytics can be a powerful tool if leveraged starting at the earliest touch point, even if it begins with a small set of customers.

Insurtech: Revolution, Evolution or Hype?

Artificial Intelligence (AI), machine learning, the Internet of Things (IoT), blockchain, robotics, quantum computing — the terminology of technology is staggering enough, let alone understanding what it is and how to use it. While some of the advances in technology are more noise than anything useful, many of these developments can be quite valuable for our industry — especially in claims management and risk control.

Fortunately, there are experts with a good understanding of insurtech and how we can make it meaningful for our companies and our injured workers. Three of them helped us break down the latest technological developments and provided insights into how they can benefit the workers’ compensation system during our most recent Out Front Ideas webinar:

  • Guy Fraker, chief innovation officer for Insurance Thought Leadership
  • Jason Landrum, global chief information officer for Sedgwick Claims Management Services
  • Peter Miller, CPCU, president and CEO of The Institutes/Risk and Insurance Knowledge Group.

What Is Insurtech?

Simply put, insurtech is using technology to improve efficiencies and provide a better customer experience in the insurance industry. Many startup companies have entered this space in the last few years, although the majority are not as helpful as they may first appear. Companies come out with apps or, as our speakers said, shiny new toys that drive user experience and seem really cool but do not add value. One speaker said many companies offer solutions to problems that do not exist.

The more mature, robust companies — those with an actual product that covers the life cycle of the value chain or a significant gap in it — are in the minority but have the most potential to make a difference. They offer strategic and comprehensive solutions. There are not too many organizations with this capability, so there are tremendous opportunities.

The right technology, when properly deployed, has the capability of making a significant impact. Two of the most meaningful advancements for our industry are machine learning and AI. So what are these terms and how do they differ?

Machine learning is actually an application of AI. Where AI is basically computer-based logic, machine learning uses statistical techniques to allow computer systems to learn from data without being explicitly programmed. From the data, it defines a formula to predict an outcome, thereby making it meaningful.

See also: Insurtech Ecosystem: Who Will Eat Whom?  

Some TPAs and carriers are using this technology to quickly flag claims that could be in danger of adverse development. The computer takes a claim, runs it against data on other claims and can determine if it is likely to become severe. As soon as the machine learning model detects something different about a claim — something a human would not be able to identify as a potentially huge loss — it alerts the claims manager to intervene and manage it more carefully. The effect is to drive the outcomes of claims in a more positive way.

For injured workers, technology is being leveraged to provide apps that provide easy access to claim information. Injured workers can find out where they are in each step of the process without having to call the adjuster.

In the consumer market, AI and machine learning are being used to apply natural language processing and determine what the person is actually saying or asking a computer. Think Alexa or Google Home. Our speakers predict it will soon become commonplace for humans to interact easily with machines.

Implementing this technology may seem overwhelming to organizations, especially if they try to adopt it on a large-scale basis. Instead, companies should have a narrowly defined plan and seek real solutions.

Industry Initiative and Blockchain

One of the exciting potential uses of newer technology in our industry is something called the RiskBlock Alliance. This not-for-profit industry consortium is meant to provide a framework that the industry owns: a standardized way of looking at data. It is based on three technologies:

  1. The Internet of Things
  2. Blockchain
  3. Data analytics

The confluence of these technologies is profound. In a nutshell, the IoT is the network of electronic devices that can digitally capture and exchange data. Blockchain enables the storage of this data along with rule sets. It can execute automated instructions based on the data and the rules applied to it. It also allows for data sharing among organizations in a secure way. A couple of examples demonstrate the significant savings and benefits to the industry:

  1. Proof of coverage. For example, a short-haul trucking company must provide proof of insurance for every load each driver takes; approximately eight hauls per day, per driver. It adds up to about 200,000 times each day that proof of coverage must be executed. There are different insurers involved with each load. It takes the company about 30 minutes to get a proof of insurance for each load. However, using the sharable platform of blockchain means the proof of insurance per load can be available in a matter of milliseconds.
  2. Sharing policy information when subrogation comes into the equation. Say there is an auto accident between two vehicles, each with a different insurance carrier. Initially, both insurers start paying the insureds while they sort through the details to see who is at fault. Once fault is established, payment between the two carriers must be settled. Right now, this is done manually at an estimated annual cost of $300 million to the industry. Using blockchain, the policy information could be housed in a secure environment and the settlement done instantly. Putting policy information in an automated process on an aggregate basis could save tremendous amounts of money and time.

Challenges and Opportunities

While there are some challenges in implementing new technologies, there are also many opportunities. Many of the more rote tasks of handling claims can be done faster by technology, freeing claims managers to provide the human touch that is so necessary in so many workers’ compensation claims. Spending more time with injured workers, showing them concern and empathy, results in better outcomes for them and lower costs for payers.

One challenge is the need for data standardization, something RiskBlock is targeting. This could level the playing field and provide opportunities for smaller insurers to grow more quickly.
Incorporating aspects of insurtech into the daily workflow can be challenging, especially because there are so many innovations and ideas at play. It is important to try to harness that enthusiasm and apply it to a framework that captures the best ideas and develops them into solutions.

Another potential challenge is that our industry is so heavily regulated, and regulated differently in each jurisdiction. That means that some insurtech solutions may work in one area but not another. Caution is required before jumping on something that may not be workable.

One challenge that can be easily overcome is changing the mindset that implementing new technology requires many people. It does not. Moving into the insurtech space is best done in a constrained way, with just two or three people involved. As one speaker said, “It’s not about thinking outside the box. It’s about building the box.” Every game-changing organization like Microsoft and Facebook started with a team of just three or four people.

See also: Key Challenges on AI, Machine Learning  

From a healthcare standpoint, one of the best opportunities from insurtech is the ability to get in front of pain, which can also be referred to as pre-pain or pre-hab. As healthcare technology advances, we will be able to help workers and their families understand what to expect in terms of pain before they undergo surgery, for example. We can help them be better prepared, facilitating better and shorter recoveries.

The Future

With the maturation of insurtech companies, our experts expect the number of startups will slow in the next couple of years. Instead, existing companies will return with innovations.

The tremendous amount of data available in the future will help level the playing field between larger and smaller carriers. This is because the smaller carriers will be able to participate in data sharing initiatives to have access to analytics way beyond what their own data could provide. Data aggregation insurtech companies are going directly to the source for data, such as partnering with auto manufacturers to access data from their onboard computer systems.

Insurtech will also allow pharmacists to match DNA to prescriptions to determine if they are feasible. Also, robotics can be used to handle riskier or repetitive tasks. Rather than replacing workers, the technology allows them to engage in more meaningful responsibilities. Using AI to process routine, medical-only claims may even result in eliminating some steps. We may find straight-through processing can be done quickly and efficiently.

One of the most exciting uses of new technology is to eliminate losses by removing risks. Insurtech can be used to detect when and how certain actions will likely lead to injuries, allowing humans to set up systems to prevent those conditions. The ability to avoid losses would truly transform our industry.

How to Use AI in Customer Service

How to manage the increase in incoming unstructured information is a key challenge in the insurance industry—we explore how Accenture’s Machine Learning Text Analyzer can achieve this using historical data.

How do you approach customer service and policy administration within your organization? In this blog post, I’ll demonstrate how artificial intelligence (AI) and a raised AIQ can help you get the most out of your data. (For the other articles in this series, click here.) To do this, I’ll discuss how insurers can use machine learning to analyze texts.

How can insurers use AI in customer service and policy administration?

The customer service and policy administration workforce can make their lives easier by using AI to:

  • Understand and act on external emails and requests.
  • Automate call center and webchat services—helping companies get on with more intricate work.
  • Enable self-service queries on policy issuance, endorsements, cancellations and renewals—using virtual assistants, for example.
  • Process unstructured data, which means fewer mistakes and better customer service.

How does AI improve customer services and policy administration?

AI enables more efficient administration processes. Insurance executives plan to invest in seven AI-related technologies in the next three years. They are: 

  • Machine learning; 
  • Deep learning; 
  • Natural language processing; 
  • Video analytics; 
  • Embedded AI solutions; 
  • Robotic process automation; 
  • Computer vision. 

See also: Policy Administration: Ripe for Modernizing  

In addition to increasing the efficiency of administration processes and enhancing analytical insights, AI technologies also benefit customer services through:

As I will show in the use case below, the customer service and policy administration workforce can use machine learning to process information faster and with greater accuracy.

Use case: Machine Learning Text Analyzer (MALTA)

Insurers today must figure out how to manage the exponential increase in incoming unstructured data. Eighty percent of data generated is unstructured, and the volume continues to grow exponentially. Forty percent of business executives complain that they have too much unstructured text data and don’t know how to interpret it.

Insurers face three main challenges:

1. Too much unstructured information

  • A large amount of information comes in through a variety of channels;
  • Incoming data is structured as well as unstructured;
  • Much of the workforce is occupied with processing unstructured information;
  • A large amount of unstructured information exists within the organization.

2. Too many communication channels

Customers use a large variety of channels to communicate with their insurance company, such as e-mail, contact forms, the service desk (e.g. ticketing), letters and applications.

3. The information is not linked to business processes

  • Workers lose a lot of time when they have to identify received information and allocate requests to the right channels;
  • They also lose time owing to inefficient processes caused by breaks in the system;
  • This prolongs the response time to clients;
  • Humans are prone to errors, which creep in at all points.

Solution: Machine Learning Text Analyzer (MALTA)

Now, insurers can automate the analysis and classification of incoming text by applying machine learning and using historical data.

How does MALTA work in customer service and policy administration?

MALTA can analyze any incoming documents, for example when customers send their policy documents via email.

These documents can be analyzed and classified using natural language processing methods and machine learning algorithms. MALTA is also trained with historical data, which enables it to classify, understand and extract information.

In the next step, MALTA links your customer’s policy document to business processes, prompting different functions to take action. Depending on the business and architecture set-up, MALTA or the output of the API triggers a process chain, a robot or an agent so that the necessary processing steps can be executed.

See also: In Age of Disruption, What Is Insurance?  

Benefits of MALTA

MALTA is flexible, customizable, independent, multilingual, state-of-the-art and end-to-end; using Accenture’s machine learning text analyzer, insurers can:

  • Increase classification accuracy and efficiency, and reduce errors.
  • Create individual learning models based on training data.
  • Deploy the solution on-premise, not only in the cloud.
  • Automate repetitive tasks, allowing employees to focus on more complex work.
  • Categorize new requests immediately and send them to the relevant departments.
  • Use state-of-the-art models and tools.
  • Work on a platform-independent web service.
  • Carry out classification outside regular business hours.
  • Cleanse data and extract and evaluate features.
  • Link robotics and process automation tools to classification.
  • Set up and train employees with minimal effort.

In addition to customer services and policy administration, insurers can use MALTA across other parts of the enterprise, for example:

Are you ready to power up your business with AI? Download the report on How to boost your AIQ for more insight.

Emerging Technology in Personal Lines

Personal lines insurers are investigating emerging technologies and developing strategies and plans related to individual new technologies. Technology is advancing so rapidly that it is even difficult to define what should be considered an emerging technology. For the past several years, SMA has been tracking 13 technologies that many consider to be emerging. These include technologies such as autonomous vehicles, AI, wearables and the Internet of Things. In our recent research, five of these technologies have emerged as “power players” for personal lines insurers, based on the level of insurer activity and the potential for transformation. The specific plans by insurers for these and other technologies are detailed in the SMA report, Emerging Tech in Personal Lines: Broad Implications, Significant Activity.

See also: 2018’s Top Projects in Personal Lines  

Some big themes for emerging tech in personal lines stand out:

  • Artificial Intelligence dominates. AI is often a misunderstood and misused term. However, when specific technologies that are part of the AI family are evaluated, much activity is underway – by insurers, insurtech startups and mature tech vendors. Chatbots, robotic process automation (RPA), machine learning, natural language processing (NLP) and others are the subjects of many strategies, pilots and implementations.
  • The Autonomous Vehicle frenzy is cooling.There is still an acute awareness of the potential of autonomous vehicles to dramatically alter the private passenger auto insurance market. But there is also the realization that, despite the hype, the transition is likely to be a long one, and the big implications for insurers are probably 10 or more years out.
  • The IoT is going mainstream. Discussions continue about the transformational potential of the IoT for all lines of business. But rather than just talking about the possibilities, there is now a great deal of partnering, piloting and live implementation underway. We are still in the early stages of incorporating the IoT into strategies and insurance products and services, but their use is becoming more widespread every day.
  • UI Options are dramatically expanding. The many new ways to interact with prospects, policyholders, agents, claimants and others should now be considered in omni-channel plans. Messaging platforms, voice, chatbots and more are becoming preferred ways to communicate for certain customer segments.

See also: Insurtech and Personal Lines  

Certainly, other trends and much emerging tech activity are happening outside these main themes. Wearables, new payment technologies, drones, blockchain and other technologies are being incorporated into strategies, pilots and investment plans. The next few years promise to be quite exciting as advancing technologies spark more innovation in the industry.