The Evolving Landscape of Homeowners Insurance

The homeowners insurance market is under intense pressure due to a combination of factors:

• Population Shifts and Real Estate Development: Growing populations and new developments in high-risk regions exacerbate underwriting difficulties.
• Unpredictable Weather: Increased frequency and severity of weather-related events make risk prediction more complex.
• Market Disruptions: Factors like social inflation, macroeconomic pressures, rapid innovation demands, and heightened competition are intensifying the strain on insurers. Litigation management costs surged 19% from 2018 to 2023 for the combined P&C sector, reflecting an approximate $24 billion loss adjustment expense (LAE).

These issues are compounded by a surge in consumer insurance shopping, a rise in higher-risk policies, and an increase in long-time policyholders switching carriers. Insurers are forced to make tough choices, such as raising premiums, exiting markets, or discontinuing certain coverage lines. These pressures are preventing the industry from achieving necessary positive outcomes critical for future profitability.

The Burden on Underwriters

Amid the ongoing market uncertainty, underwriters particularly are facing significant challenges:

• Overwhelming Submission Volume: An influx of submissions strains underwriting resources.
• Inaccurate Risk Prediction: Traditional methods based on demographic data and zip codes are proving to be inadequate, necessitating more precise risk assessment tools.
• Data Quality Issues: Poor-quality or unstructured data and manual processes further complicate risk assessment.

Underwriters are questioning why the process is so burdensome and how it can be alleviated. The crux of the issue is underwriting profitability, crucial for maintaining healthy bottom-line results. Fortunately, there are modern solutions to address these goals.

Modernize Risk Assessments with Behavioral Predictions

Traditional underwriting relies heavily on demographic data and location. Enhanced risk assessment goes beyond traditional methods such as ZIP Codes and credit scores by incorporating individual behavior and decision-making patterns.

Underwriters today can access a wealth of data outside traditional risk variables to deepen the understanding of their policyholder’s risk profile. By utilizing AI-powered behavioral predictions, which incorporate information about consumer activity, interests, buying choices, etc., underwriters can more accurately predict outcomes relevant to insurance underwriting performance.

To remain competitive, insurers must invest in this digital transformation and reduce reliance on complex, manual processes. In fact, a McKinsey analysis found that the most successful carriers are those leveraging the latest technologies to optimize underwriting capabilities. Enhanced approaches for success involve adding:

• Comprehensive Risk Profiles: A detailed view of prospective and existing policyholders to start transforming underwriting workflow.
• AI and Person-Level Insights: Using AI to access powerful person-level insights about customers and their individual risk propensities, which directly impact underwriting profitability.

Leveraging Behavioral Predictions Across the Insurance Value Chain

Using behavioral intelligence represents a transformative shift in insurance underwriting. By incorporating policyholder behaviors into risk evaluation, insurers gain a deeper understanding of risk profiles and individual customer nuances, leading to a more customer-centric approach to coverage. This intelligence offers specific predictions that enhance risk assessment:

• Identifying policyholders with a high propensity to seek an attorney at first notice of loss (FNOL) or a likelihood to litigate.

• Predicting claims frequency and severity for current policyholders and prospective customers.

• Assessing the likelihood of non-payment or early cancellation.
• Determining which prospects are most likely to convert to new customers and predicting their lifetime value.

When insurers identify policyholders with a higher propensity for risk, they can proactively manage these policies more precisely using unique identifiers. Meanwhile, they can handle other policies based on different qualifiers that indicate varying levels of risk. This becomes a game-changer for insurers’ ability to predict and review for underwriting.

Empowering Underwriters

Underwriters are tasked with building profitable books with targeted risk profiles. To be successful, they need to make quick, effective, and accurate assessments of the profitability of each policyholder. However, they are often working with limited or convoluted information and are under considerable time constraints.

Utilizing person-level intelligence in underwriting allows underwriters to focus their craft and expertise on the most complex risks. By integrating person-level intelligence, insurers can:

• Identify High-Risk Insureds Early: Early identification of high-risk individuals allows underwriters to focus on the most complex cases, improving resource allocation.
• Improve Efficiency: With simply using a name and address, significant risk insights can be obtained in seconds.
• Enhance Risk Handling: Focus on complex cases while automating the handling of low-risk applications, improving overall workflow efficiency.
• Provide Personalized Service: Offer coverage tailored to individual risk profiles, moving beyond traditional factors like location and credit scores.

Future-Proofing Insurance Underwriting

For P&C insurers, integrating person-level intelligence into underwriting processes offers a more precise and complete view of the policyholder risk profile. This approach helps insurers better prepare for uncertainty, respond to market volatility, avoid adverse selection, and achieve profitable, sustainable growth. These AI-powered behavioral predictions empower insurers to:

• Predict and Manage Risks More Accurately: Identify high-risk individuals earlier and adjust policies accordingly.
• Enhance Customer Understanding: Gain insights into customer behavior, such as propensity to litigate or likelihood of early cancellation, improving risk management strategies.
• Improve Underwriting Profitability: Achieve more accurate risk assessment and better manage underwriting resources, contributing to healthier bottom-line profitability and sustainable growth.

Using Advanced AI Risk Assessment with Pinpoint Predictive

A more comprehensive and inclusive risk assessment requires a deep understanding of the individual behind the policy, as well as incorporating insights into an insurer’s decision-making processes as part of risk analysis.

Pinpoint Predictive empowers underwriters by enabling them to make smarter, more equitable assessments of risk, accurately identifying high-risk and low-risk individuals. This enhanced accuracy in underwriting workflows helps insurers better serve their customers by identifying and quantifying individual risk earlier and more accurately.

By bridging the gap between the most powerful behavioral predictions made by the world’s leading tech companies and the specialized requirements of the insurance industry, Pinpoint is delivering unmatched risk-selection capabilities at various points along the insurance value chain.

Conclusion

The homeowners insurance sector faces unprecedented challenges. Behavioral intelligence, centered on individuals, represents the next generation of technology, transforming policyholder risk assessments and offering insights into future customer risks. As the industry adapts to these new technologies, the focus will shift toward more informed, efficient, and customer-centric underwriting practices, paving the way for a more resilient and profitable insurance market. Insurers that integrate these advanced risk assessment tools and insights will ultimately be the most successful in enhancing their underwriting processes, addressing the evolving risks associated with new developments and weather events, and ultimately improving financial outcomes.

Improve your Underwriting Outcomes with Pinpoint

With predictions available earlier and more accurately than any other risk solution on the market, Pinpoint is transforming the P&C insurance industry and helping underwriters drive better outcomes with an AI-powered, real-time solution for precise risk selection. For more information about how Pinpoint can help you, visit www.pinpoint.ai or contact info@pinpoint.ai.

The post Addressing Homeowners Underwriting with Behavioral Risk Predictions first appeared on AAIS.

Challenges Facing the Construction Industry and the Impact on Insurance

The construction industry relies heavily on the insurance sector to safeguard its key assets, projects, equipment, and personnel. Rajaraman noted that currently, the industry is seeing robust demand. “This is largely fueled by fiscal policies enacted during the pandemic era and initiatives like the CHIPS Act, which have boosted both residential and commercial infrastructure projects,” he said. “Consequently, construction costs have soared steeply.”

On top of this, the increase in natural calamities has impacted both the frequency and magnitude of insurance claims. New risks are adding to these challenges. “While the industry traditionally faced losses primarily from events like floods, windstorms, and theft, it now confronts novel exposures, such as those related to renewable energy, hailstorms, and other catastrophic events,” Rajaraman stated. “These factors have created a challenging market for construction project insurance.” While many of these challenges are beyond the control of insurance companies, he believes staying cooperative demands they innovate their processes by leveraging technology and reducing costs.

How Insurers Can Optimize Costs Associated with Construction Insurance

There are specific strategies Rajaraman recommends for reducing overall costs related to managing insurance products and programs. “We are seeing an increase in the prominence of MGAs in the builder’s risk space, where they can use their market expertise and favorable cost structures for better risk assessment and pricing,” he explained. “A U.S.-based mutual carrier approached us with a request to set up an AAIS-based builder’s risk program for three distinct MGAs, each targeting different customer segments. We recommended a streamlined approach where programs, such as the AAIS builder’s risk standard, serve as the foundation without necessitating individual implementations for each MGA.” Instead, each MGA’s view of the AAIS builder’s risk program was tailored to align with their program-specific requirements. According to Rajaraman, this model significantly reduced the costs associated with establishing the programs and all three programs were successfully launched within a span of two weeks. Additionally, by circumventing the traditional underwriting process, the carrier also gained visibility into strong and weaker classes of risks, providing valuable insights into opportunities for improvement.

Ensuring Data Security and Isolation for Multiple MGAs or Carriers Sharing the Same Instance

Rajaraman has found that MGAs and carriers want to ensure there is no compromise on their data; they need to keep their data secure and isolated from other insurers on the cloud. “We prioritize the safety and security of each insurer’s data by implementing separate databases per insurer,” he shared. “This means that each insurer’s information is stored in their own isolated repository, completely segregated from other insurance companies’ data. There is also no mixing or overlap of data between insurers, thus maintaining confidentiality and integrity of each insurer’s sensitive information.” This helps insurers enhance their data security and stay compliant with regulatory requirements and industry standards. Rajaraman recommends this approach for carriers and MGAs looking to benefit from the scale of cloud speed and security as well.

To view the full interview with Anand Rajaraman, click on the video above.

The post Methods for MGAs to Cut Construction Insurance Costs first appeared on AAIS.

Why is GenAI Different?

In our opinion, beyond everything else, the following three characteristics of GenAI make it very different from previous generations of AI/ML:

• Generally Pre-Trained: GenAI tools come out of the box with the impressive ability to understand the English language and reduce the overhead of training task-specific language models.
• Ability to Analyze/Synthesize Complex Language: GenAI can effectively summarize text and create new content with little training.
• Large Context Abilities: GenAI tools can work on large volumes of text to draw inferences.

Insurance Complexities and Limitations of Traditional AI

Insurance is saddled with enormous manual processes that make it error-prone, expensive, and slow. A lot of insurance processes still rely on exchanging documents up and down the value chain. Brokers need to fill out complex forms, review multiple quotes and policies to effectively inform their insured clients about their choices, and then check policies thoroughly for errors. Issuing certificates, managing mid-term alterations, and renewals are all still largely done manually. Insurance companies are also still analyzing forms sent by brokers, reading claims reports and matching them against policy coverage, issuing reinsurance bordereau, and performing many other similar tasks manually.

Traditional ML tools cannot process and analyze insurance language or extract insights from these complex documents. Standard workflow and robotic processing automation (RPA) technology do not support true AI-guided workflow or AI-guided human review and decision-making. This is where GenAI can make a huge difference.

The BluePond.AI Approach to Making GenAI Work in Insurance

While GenAI tools natively bring a high degree of English language understanding and analysis capability, they do not understand insurance concepts. To make effective software for automating human tasks, machines need to understand the insurance terminology, the context of a process or purpose of a document, the references between various documents (e.g., the policy associated with a claim assessment report), different coverage types, and more.

Therefore, to solve complex insurance process automation and document-understanding tasks, BluePond.AI has built our proprietary “P&C Domain Reference Library” to serve as a repository for insurance language, entities, values, fields, coverage, and document types. This reference library is the master orchestrator of the repository consisting of traditional and GenAI tools that, when used in symphony, can execute tasks like a trained insurance professional, thus creating a unique P&C insurance “CoPilot.”

Whether you’re an insurer, reinsurer, third-party administrator, or broker, our GenAI-based “CoPilot” approach can help resolve persistent challenges within the insurance industry.

Insurers & MGAs

• Automatic Audit of Delegated Underwriting: GenAI can assist with the automatic monitoring of policies written under delegated underwriting authority by program managers, MGAs, or cover holders, to ensure issued policies comply with underwriting guidelines at the time of underwriting.
• Claims Document Monitoring: GenAI tools can automatically identify incoming documents, extract data enabling the population of the claims system, and flag any major risk or severity factors.
• Ingestion of Loss Assessor Notes: GenAI can identify, match, and extract event details and loss cause/estimate from assessor notes. Plus, it can generate claim trends, seasonality, and complexity insights for underwriting and product development.
• Matching Claims to Policy Coverage: GenAI can offer automatic matching of claims against policy coverage, including looking through language that changes coverage under certain circumstances.
• Policy Checking: GenAI tools can help with checking issued policies against submission data, underwriting instructions, and prior terms to ensure there are no errors.

Brokers and Agents

• Policy Checking: Despite Errors and Omissions (E&O) lawsuits against brokers being on the rise, many brokers and agents struggle to make time for tedious policy checking. With GenAI, policies can be checked thoroughly against expiring quotes and detailed submission schedules when and where required.
• Quote Comparison: GenAI can create easy-to-understand summaries of complex quotes from different carriers/MGAs even if they widely differ in format and inclusions.
• Certificate of Insurance (CoI) Issuance: CoI processing is largely a manual process, with various types of non-standard requests making it a complex, time-consuming, and expensive operation. With GenAI tools, we can automate the understanding of CoI requests, extract the right information from policies, and digitize request processing – taking a big operational load off agents’ shoulders.
• Submission Verification & Inspection Matching: GenAI can intelligently automate submission data verification across various documents and cross-check against third-party data sources.
• Placement Recommendations: GenAI tools build recommendations on the right coverage for the insured and best-suited markets to place the risk based on the extraction of data from historical quotes and policies of similar insureds.

Reinsurers

• Bordereaux Ingestion: Data entry from numerous long bordereaux remains an operational challenge for reinsurers, which also creates information asymmetry and can lead to leakage of premiums and overpayment of claims. With GenAI-based tools, reinsurers can extract granular data from premiums and claims bordereaux, saving time and cost of data entry while also ensuring treaty conditions are being complied with at a granular level.
• Claims–Premium Matching: GenAI can automatically match claims information in relation to the assumed (or ceded) premiums and direct premiums through various contract documents to help improve inaccuracies and manual errors.
• Run-Off Book Monitoring: GenAI tools allow reinsurers to set their own parameters to monitor run-off claims development and flag any adverse or unexpected developments.
• Loss Portfolio Transfer (LPT) Assessments: LPTs carry billions of dollars of investments and need an accurate analysis of the underlying portfolio to avoid the risk of underperformance. With GenAI tools, the bottom-up assessment of an LPT book becomes feasible, with the ability to scan for exact exposures carried in the policies, assessment of reserve adequacy, adverse claims development indicators, and more.
• Audit of Cedent Underwriting: GenAI can perform data-driven audits of cedent underwriting based on ingestion of details from bordereaux and cedent underwritten policies.

BluePond.AI: Taking the Friction out of Deployment

The potential of GenAI in the insurance industry is vast. We foresee there being a significant ability to automate complex human processes, leverage AI to guide human review and decision-making, and extract not just data, but also insurance insights from the language in complex documents. That’s why we created the P&C CoPilot Platform. The platform can analyze documents, extract values and fields, and derive insurance insights from complex language (e.g., coverage comparison from two different types of policy documents). However, one might wonder about the difficulty of executing a GenAI strategy in-house. Even if you have a team with deep experience in software engineering or traditional AI, the fundamental construct of GenAI systems is different so it requires a different approach to execution.

For that reason, we have configured our P&C CoPilot Platform as a ready-to-deploy Platform as a Service (PaaS) for any P&C industry participant. With this approach, you will not waste time procuring, testing, configuring, and integrating the fundamental building blocks that are required for a typical GenAI application. The platform is fully configured and ready for quick deployment either as a fully managed PaaS in your environment or as a multi-tenant Software as a Service (SaaS) in ours. It is designed to connect to various data and document sources, an API gateway to connect to internal and external services, and enable seamless enterprise integration.

BluePond.AI manages the ever-evolving GenAI landscape and connects to various best-of-breed services behind the scenes so that you can focus on the insurance use case. We offer complete execution support on data science and software engineering, and post-production human-in-the-loop processing support if required. Even if you have the right GenAI expertise and infrastructure, our platform can still save up to 12 months of setup time and expenses, while providing use case acceleration with our P&C reference library.

To learn more about BluePond.AI and how you can successfully leverage GenAI, visit www.bluepond.ai.

The post The Art of the Possible with GenAI in Insurance first appeared on AAIS.

Issues Insurance Carriers are Facing

While Kadous believes the biggest issue insurance carriers are facing is related to reinsurance, there are a lot of other challenges affecting their businesses. “There’s a lot of challenges with keeping up with the latest technology,” he said. However, Kadous doesn’t view advancements in insuretech necessarily as a problem, but more of an opportunity for the insurance industry as a whole.

Opportunities in the MGA Segment

Kadous has noticed that a lot of new startups are rolling out as MGAs, which he finds interesting. “What we see there is an opportunity for front-end carriers in particular,” he shared. “But in addition to that, there is an opportunity for AAIS to engage directly with those MGAs, learn more about what they’re doing, and pull them into some of the discussions on our councils.” Kadous further explained how AAIS is creating a new membership class that will allow MGAs to come on board as AAIS non-voting Members, but still be able to utilize its programs directly without needing to pull their front-end carrier in as an AAIS Member. “We think it’s a change for the better,” he emphasized. “This membership option is going to really open up the MGA space and enable us to invite them into our community.”

AAIS: Supporting its Members in a Tough Market

One way AAIS is currently supporting its Members is through its Product Deployment Platform (PDP). “When companies go to adopt a program or update a program, there are always expenses involved from the company perspective,” Kadous explained. “We’re working on our [PDP] and building out some really terrific assets that are going to help in terms of getting a product up and running or updated and delivering things such as data tables digitally.” In addition, AAIS also has Mind Maps, which essentially explains how to program a form’s logic and how to program the entire manual. “Mind Maps tries to simplify things as much as possible to make either a transition or an update much less painful,” Kadous assured.

To view the full interview with John Kadous, please click the video above.

The post AAIS Vice President of Products Reveals Organizational Opportunities in the MGA Segment & Member Support Initiatives first appeared on AAIS.

Trends in Commercial Property

Dunn started off the discussion by requesting overviews of what the panelist are currently seeing in the market. Coan explained that while inflation isn’t a new trend, COVID did have a larger effect than normal. One point that he likes to make clear, is that inflation is not one in the same. It is different regionally and by state. “Florida has a 7.14 average inflation rate across the state, but Fort Myers and Miami are both over nine,” Coan reported. “Fort Myers is over nine because of [Hurricane] Ian, and Miami is over nine because inflation is driven by the people who live [there].” When developing costs, Coan insisted that it is important to keep in mind regional costs all the way down to zip code and census block level. “When you’re underwriting, you need to understand that different regions are experiencing inflation and cost factors differently.”

Inflation is affecting every American and when it comes to the confluence of factors in the renewal books, Heine has noticed there tends to be a mismatch of pricing terms and conditions. “We’re urging our clients to take a good healthy look at the renewal book to make sure that they’re managing that risk,” he said. “When you’re adjusting values, you have to look at everything.”

Current Challenges and Actions Carriers Can Take to Address Them

Coan reported that more than 50% of homes are short to value. “These [catastrophic] losses all over the country have shown that no one has a book that’s a true value,” he emphasized. “Everyone thinks they have a book that’s insured to value, but you don’t. When you run books of business, you’re going to see the valuations of those books are more than 50% undervalued.” Coan admitted that this is a common problem, even more so on the commercial side. “They get looked at less often,” he stated, “and there’s less information that the carrier is getting from the broker or the agent to write those policies.”

Heine believes carriers are under great pressure due to the tightening capacity. “There are a lot more questions at the portfolio and individual property levels when it comes to the middle market and large commercial,” he said. “At that level of discipline, folks are able to show that they actually have a command over their book and a process to help manage that. Part of what we’re seeing is a gap in terms of what these folks are insuring for and what the reinsurers are looking at.”

In terms of action to help address these problems, Coan believes it starts with data and technology. “You need to have third-party data built in because you can’t have all the information on this risk,” he advised. “It has been proven that vendors like Betterview and e2Value can help adjust the data, dissect that data, and provide customers with really good information on these risks.” Coen stressed that it is important to take advantage of these advances in the marketplace. “If you’re not in a more digital world doing things like incorporating an API and pulling in real-time information so you can give a real-time solution, then you’re going to fall behind,” he claimed. “Everyone understands new technology, so the excuse that customers aren’t going to want to do that is not true. You need to bring in the right data to process your claims, your underwriting, and your submission policies.”

The Way Forward: Progress of Insurers & Technology/Data Working Together

The integration of new technology such as API has been helpful for companies like Betterview when it comes to strategy. “Resources are at a premium wherever you go and or customers were absorbing that,” Heine explained. “Now, it’s like someone flipped that discussion and we’ve decided to absorb that time, effort, and cost. Overall, technology has advanced the actual value of what folks can get over a period of time.”

While there may be some hesitancy to get involved with data and technology providers, especially from legacy carriers, Coan believes it will pay off in the long run. “It actually makes that traditional underwriting and claims process more robust because you’re able to filter out those risks that are functioning and priced correctly and focus more time on where your risk is really valuable,” he assured. “So, you’re not losing any of the traditional core values your company may have. You’re actually enhancing them.”

If you would like to view the presentation again in its entirety, please click the video above.

Questions? Please don’t hesitate to reach out to any of the featured speakers through the contact information below.

Stephanie Vasey

Product Manager of Commercial Lines and Inland Marine (AAIS)

stephaniev@aaisonline.com

John Dunn

Vice President (OneShield)

jdunn@oneshield.com

Jeff Heine

Chief Revenue Officer (Betterview)

jheine@betterview.com

Skip Coan

Vice President (e2Value, Inc.)

scoan@e2value.com

The post Webinar: Property Preparedness & Bracing for the Future of Commercial Insurance first appeared on AAIS.