Artificial intelligence is set to transform the insurance industry. According to one study, 99% of insurers are either already investing or making plans to invest in Generative AI (GenAI).

They're targeting ways to streamline operations, improve decision-making and enhance customer service. The opportunity is clear: One study found that GenAI could reduce payouts by between 3% and 4%, and drive a 20-30% reduction in loss-adjustment expenses in claims alone.

Yet intent does not guarantee success. Insurers are not investing in AI to secure small gains, but that is often what happens because they start with off-the-shelf AI models.

The limitations of AI models

These services deliver incremental benefits, and there is certainly value in that. What they do not offer is a fundamental change in how firms work and operate.

The reason is that these off-the-shelf models are often positioned as quick, plug-and-play services that need little time to integrate. They might work well on simple, repetitive tasks such as answering internal queries but struggle on more nuanced, complex work such as underwriting, claims processing, and fraud protection.

At the heart of these limitations is data quality. Generic solutions are trained on generic datasets lacking industry- or use case-specific context, and as a result, deliver generic responses. They also risk providing outdated and incorrect information, hurting customer satisfaction and increasing compliance and regulatory issues.

Can models be improved?

There are ways to tackle these challenges. Techniques including retrieval augmented generation (RAG) and fine-tuning do allow insurers to adjust performance:

● RAG involves engineering prompts using specific knowledge to provide context

● Fine-tuning deploys private data to teach models how to respond to prompts

These approaches can enhance accuracy, but they require insurers to have high levels of internal AI literacy. The success of RAG and fine-tuning efforts is closely linked to having subject matter experts who can work on prompts, know the right data to include in training, and review output quality. In other words, it is resource-intensive for relatively little gain; AI remains an assistant to human decision-making.

Ultimately, even off-the-shelf models that have been enhanced through RAG or fine-tuning are only transformative at an application or service level. They can accelerate processes, but what can be achieved is limited by what humans themselves can accomplish.

The real transformation – AI agents

The real opportunity is found in agentic AI, software that performs tasks autonomously with at most limited human intervention. Agents make decisions, plan and learn from new information to complete work defined by clear instructions, delivering focused and precise execution.

They adapt and learn in real time, rather than needing to be retrained regularly. As such, once deployed, they remain aligned with current business needs and industry conditions, ensuring reliability and adaptability.

What sets agents apart from off-the-shelf models is that as well as completing the task, agents can redesign processes to work more effectively.

For instance, a customer has a car accident and needs to make a claim. The traditional process will have been shaped by a human workforce and will likely involve multiple steps, all of which require different information, with no chance of progressing until the previous step has been completed. As a result, a decision can take weeks, if not months, as all parties are consulted.

With an agent-led process, the same information will still be required, but it could all be gathered simultaneously from multiple sources and assessed in real time. It does not matter whether it is incident details, imagery, vehicle value, historic claims data, or the cost of a replacement; the agent can continually gather and review to make a decision in days if not hours.

Complete transformation with full-task automation

This is full-task automation. It allows insurance companies to not just get more work done, with higher levels of accuracy, but to transform how they operate in ways in which they previously never had the time or resources to do. They can:

• Scrutinize every claim like it was the only one they received that day
• Predict risk with deeper levels of accuracy
• Personalize products to a degree that makes the customer feel like they have been created just for them
• Be confident that the potential for fraudulent claims is hugely reduced

AI agents will have a truly transformative effect on insurance, but only if they are deployed at scale. One agent is a pilot; hundreds is a transformation.

To achieve that requires an enterprise-wide AI infrastructure that supports autonomous operations. This needs to ensure that AI agents can integrate with existing systems, operate autonomously, and allow subject matter experts to configure AI agents according to their specific business objectives.

Investing in hundreds of agents may seem like a huge investment, but it is not; right now, agents are good enough to start taking ownership of simple tasks, and they are continually learning. A year from now, they will be 10 times better at 1/100th of the cost. The cost barriers to entry are dropping dramatically.

That also means that competitors will be using them. Can a firm compete against other providers that deploy agents alongside human teams of specialists?

Unlocking operational transformation

AI models, procured off the shelf, will not transform insurance. They will drive incremental benefits, and if deployed at scale those benefits may improve the bottom line by a percent or two. They will not completely overhaul processes that have been designed and built to accommodate the limitations of human workforces.

Only agents can deliver the operational level transformation insurers need, when AI no longer assists teams but takes full responsibility for tasks. By adopting autonomous agents, insurers will be in a stronger position to unlock the opportunities AI offers.

In today's globalized economy, businesses face numerous challenges, including tariffs and fluctuating supply prices. Trade credit insurance is a critical tool for protecting businesses engaged in international trade and mergers and acquisitions (M&A) against these uncertainties. It provides protection against non-payment risks, ensuring financial stability during uncertain times. This article delves into how trade credit insurance mitigates risks from tariffs and fluctuating supply prices and the strategic role it plays in M&A transactions.

Managing Tariffs and Supply Price Fluctuations

Protection Against Non-Payment

When tariffs are imposed, not only does the cost of importing goods rise, but so does the cost of those goods themselves. This makes it significantly harder for buyers to pay the agreed-upon amounts and to meet their payment terms. Many businesses face delayed or non-payment for goods from companies unable to afford the tariffs. Trade credit insurance can cover the seller if the buyer is unable to pay due to these price increases and can help businesses maintain their cash flow and financial stability despite economic challenges.

Assessing Buyer Risk

Trade credit insurers assess buyers' financial stability before transactions occur, helping businesses understand the risk involved. This is particularly important when tariffs raise the cost of doing business with foreign customers or suppliers and in industries with volatile commodity prices. These risk assessments provide peace of mind, allowing businesses to maintain their cash flow even if customers face payment delays or defaults.

Enhancing Financing and Cash Flow

Trade credit insurance helps businesses secure better financing options, as lenders view it as a risk mitigation tool. This can be particularly valuable when dealing with international trade where tariff and supply price volatility can affect cash flow. It ensures that, even if a customer defaults on a payment due to financial strain caused by tariffs or rising supply costs, the business can recover a significant portion of the amount owed.

Role in M&A Transactions

Trade credit insurance can play a vital role in M&A by mitigating risks associated with the buyer's or seller's receivables and improving the financial stability of the parties involved.

Mitigating Credit Risk

In M&A, trade credit insurance protects against the risk of uncollected receivables for both buyers and sellers.

If the company being acquired has significant accounts receivable, trade credit insurance can protect the buyer against the risk that those receivables may not be collected. For example, if the target company has customers who owe large sums, and those customers default or face financial difficulty, the insurer covers the losses. This reduces the buyer's concern about the quality of the receivables when evaluating the value of the target company, giving them confidence that these risks are covered, ensuring smoother financial integration after the acquisition.

Enhancing Valuation and Financing

While insurance can lead to higher valuations, it also eases financing for buyers, as lenders are more willing to provide funds knowing the risk of default is mitigated. This stability is crucial for securing working capital post-transaction.

Having trade credit insurance can increase the valuation of the target company by ensuring that a certain portion of receivables is protected against non-payment. It helps the buyer assess the true value of the target company's financial assets, making the target more attractive by reducing the perceived financial risk.

Ensuring Continuity and Reducing Disputes

Trade credit insurance supports business continuity post-acquisition by ensuring the customer base remains stable and reducing disputes over the quality of receivables.

Trade credit insurance ensures business continuity post-acquisition by stabilizing customer relationships and covering future payment defaults. It also reduces disputes, providing protection if there are concerns about outstanding payments or the quality of the target company's receivables. This facilitates smoother negotiations and integration. It also ensures both parties are covered if there is a default post-acquisition. For example, the buyer may demand that the seller take on the risk of bad debts from the receivables, and trade credit insurance can help facilitate this process.

In addition, trade credit insurance can ensure that the seller's customer base remains stable post-acquisition. If the target company has long-standing relationships with customers, the buyer may be worried that any disruption in those relationships could affect future revenue streams. Trade credit insurance provides the buyer reassurance by covering potential future customer payment defaults.

Reducing Escrow Needs

In M&A deals, escrow accounts are often set up for contingencies like bad debts. Trade credit insurance can reduce the need for escrow by assuring that receivables are covered in case of non-payment, simplifying the transaction process.

Trade credit insurance provides a safety net that mitigates risks from tariffs and volatile supply prices, enhances the value of target companies in M&A transactions, and improves the financial stability of both buyers and sellers. By fostering smoother transactions, reducing post-acquisition risks, and enabling more favorable financing terms, trade credit insurance plays a vital role in ensuring business continuity and growth in an unpredictable economic landscape.

In the complex world of insurance underwriting, mental health presents a growing challenge. The stigma surrounding mental health may or may not be slowly dissipating, but insurers continue to face a critical question: How can they accurately assess mental health risks while providing a positive customer experience?

A study by Reinsurance Group of America (RGA) offers compelling answers, leveraging behavioral science to transform the disclosure process.

The mental health disclosure dilemma

Certain mental health conditions can affect mortality and morbidity outcomes, making accurate disclosures crucial for risk assessment. However, the persistent stigma and confusion surrounding mental health often leads to reluctance in sharing this information. RGA's 2023 Mental Health Survey highlighted this challenge, with 55% of insurers reporting difficulties in underwriting and managing mental health-related claims.

The crux of the problem lies not only in applicants' hesitancy about disclosing but also in the very design of the questions they face. Traditional approaches often fall short, creating ambiguity and emotional discomfort that hinder accurate responses.

Five key behavioral science findings

RGA's behavioral science team conducted an extensive experiment involving 4,049 participants from the U.S. and Australia. The study tested various techniques to improve how customers interpret, process, and ultimately answer mental health questions. The results were striking, offering a new paradigm for insurance underwriting.

The study revealed five key findings:

1. Specificity encourages disclosure

Providing a specific list of mental health conditions increased disclosure rates 17%. For example, instead of asking, "Have you ever been diagnosed with, suffered from, sought medical advice for, or received treatments for any mental health condition? Some examples include anxiety, post-traumatic stress, depression, or schizophrenia," an altered question that asks "Have you ever been diagnosed with, suffered from, sought medical advice for, or received treatments for any of these mental health conditions," followed by a list of 12 possibilities, yielded a more accurate yes/no response.

This approach reduced cognitive load, making it easier for applicants to recall and process information. Surprisingly, this more detailed question only took an average of six seconds longer to complete, with no negative impact on customer experience.

2. Normalizing mental health issues increases openness

Adding a de-stigmatizing statement that normalizes reporting mental health challenges boosted disclosure rates by an additional 10%.

This statement reads, "It is increasingly accepted by people everywhere that recognizing and taking care of our mental health is important. In fact, a recent study showed that the number of people reporting mental health conditions has increased by 20% since 2014, and many adults now take active steps to manage their mental health."

This simple addition increased openness and did not hurt overall user experience.

Better still, pairing this statement with a list of specific conditions noted in the first key finding added only 12 seconds to completion time.

3. Segmenting stigma enhances detailed disclosures

Separating mental health conditions into distinct questions based on associated stigma levels increased the likelihood of participants disclosing multiple conditions. This approach improved disclosure rates for both highly stigmatized conditions, such as schizophrenia, and less stigmatized ones, such as stress and sleep disorders.

4. Balancing detail and emotional comfort helps

While detailed disclosures are valuable for underwriters, they can be emotionally challenging for customers. The study found no difference in disclosure rates between asking about "any conditions" vs. specific conditions. However, participants reported feeling more embarrassed when asked about specific conditions, highlighting the importance of emotional context in question design.

5. AI is an unexpected ally

Intriguingly, half of participants self-reported that an automated chatbot would be their most comfortable channel for disclosing mental health conditions. This preference aligns with psychological distance theory, suggesting that non-human interfaces may provide a more comfortable environment for sharing sensitive information.

Implications for the insurance industry

These findings have far-reaching implications for insurance underwriting and beyond. By incorporating behavioral science principles into question design, insurers can:

• Improve the accuracy of risk assessments
• Enhance the customer experience during the application process
• Potentially increase uptake of insurance products by making the process less daunting

Moreover, these insights can be applied to other sensitive areas of the insurance journey, such as claims forms and personalized claims conversations.

The future: A human-centered approach

As the insurance industry evolves, embracing a human-centered approach to underwriting becomes increasingly crucial. By understanding the cognitive and emotional factors that influence disclosure, insurers can create more effective, empathetic processes that benefit both the company and the customer.

The possibility of future integration of AI and chatbots presents exciting opportunities but would also create challenges requiring more detailed exploration. As these technologies advance toward reality, finding the right balance between psychological comfort and honest disclosure will be key.

By embracing these behavioral science insights, the insurance industry can take a significant step forward in addressing the mental health underwriting challenge.

Read the full report, "Improving Mental Health Disclosure for Insurance Underwriting."

References

• https://www.psychiatry.org/patients-families/stigma-and-discrimination

• https://www.cambridge.org/core/journals/bjpsych-open/article/creating-a-hierarchy-of-mental-healthstigma-testing-the-effect-of-psychiatric-diagnosis-on-stigma/3F7F87E0D4F50412B2A90072CFD8B995

• https://www.rgare.com/knowledge-center/article/how-can-life-insurers-improve-the-dtc-applicationprocess-a-behavioral-science-analysis

• Giesbrecht, G. F., Müller, U., & Miller, M. (2010). Psychological distancing in the development of executive function and emotion regulation. In B. W. Sokol, U. Müller, J. I. M. Carpendale, A. R. Young, & G. Iarocci (Eds), Self and social regulation: Social interaction and the development of social understanding and executive functions (p. 337–357). Oxford University Press.

• https://www.abi.org.uk/products-and-issues/choosing-the-right-insurance/health-insurance/mental-healthstandards/

• https://www.rgare.com/knowledge-center/article/searching-for-simplicity--improving-customercomprehension-in-life-insurance-through-behavioral-science

• https://www.rgare.com/knowledge-center/article/bringing-order-to-complexity-in-claim-forms-an-rgabehavioral-science-study

For going on 30 years now, executives have complained about being "Amazoned": Even though they work for companies that aren't even remote competitors of Amazon, the executives are held to the same high standards for customer service that Amazon has trained people to expect.

In Amazon's annual letter to shareholders this year, CEO Andy Jassy goes on at length about how Amazon keeps raising those standards. Some of the letter is boilerplate that anyone who's cracked open a management book has seen, and some isn't readily applicable to the insurance industry, but I think two of his points could help the insurance industry greatly. They relate to having a "why culture" and figuring out what NOT to do.

Let's have a look.

Jassy's letter includes the obligatory bragging about recent accomplishments and projections of future wins but has a middle section that I found quite smart, on framing decisions. He says Amazon has long divided choices into "two-way and one-way door decisions"—decisions that you can walk back and decisions that you can't. 

He writes:

"But, both of these constructs assume the door is unlocked. A lot of invention is about trying to open doors that have historically seemed bolted shut. And, over the past 30 years, we’ve found one of the most important keys to unlock these doors has been a simple question: 'Why?'

“'Why does this customer experience have to be this way?' 'Why can’t it be better?' 'What are the constraints—why must we accept them?' 'Why can’t we invent around that?' 'Why will it take so long to get to customers?' Why?"

Jassy says Amazon focuses on always improving its "WhyQ."

Insurers can't be as nimble as Amazon. Regulation means there aren't as many "two-way door decisions" in insurance. Even if you can reverse a decision, it may take years to unwind, say, the issuance of policies as part of a move in a new direction.

But there are still an awful lot of things in insurance that are done a certain way because they've always been done that way, even though they're inefficient and customers dislike them. Insurers should be asking: Why do we still use so much paper? Why has our "digital transformation" just moved us to digital versions of those paper forms? Why do claims take so long? Why does underwriting take so long? Do we really have to ask all those questions? Why, why, why?

Insurers have been asking lots of those questions over the past decade and have made progress, but "why?" is a question that has to be asked constantly, and we have a long way to go. 

The second point in Jassy's letter that I think could greatly benefit insures relates to the first. It's about continually asking what you should stop doing because it gets in the way. 

Jassy writes:

"Last fall, I asked teammates across the company to send me bureaucracy examples that they were experiencing. I’ve received almost 1,000 of these emails.... As leaders, we don’t always see the red tape buried deep in our organizations, but we can sure as heck eliminate it when we do. We’ve already made over 375 changes based on this feedback.... We are committed to rooting out bureaucracy that ties up time and dispirits our teammates."

I'd bet a similar exercise would net all kinds of gains at any company of any size in the insurance world. When I started taking week-long bicycle trips and was wondering how much clothing to carry, I read a slogan that made a huge difference: "If you take care of the ounces, the pounds will take care of themselves." The same is true of bureaucracy. There doesn't even have to be a huge aha! moment for the exercise to surface a host of seemingly benign practices that add up to real frustration for both employees and customers. (I wrote at length on this sort of exercise two years ago.)

I'm by no means saying Amazon is a paragon in every way. Many employees have complained about how they're treated. Lots of companies that sell through Amazon have sued it, alleging a variety of unfair business practices. Customers gripe about how the site has been junked up with ads. Amazon and its founder, Jeff Bezos, have drawn flak in some quarters for business decisions that seem to be designed to curry favor with the new Trump administration. 

But Amazon built a massive business through a relentless focus on the customer, and insurers could learn from Jassy's advice.

Cheers,

Paul

P.S. Two other recent articles make useful points on improving the customer experience.

In "Why 54% of Customers Are Disappointed: 5 CX Mistakes Your Business Can't Afford," Bernard Marr says companies these days can capture data on every touchpoint with a customer and tend to use the data well in sales and marketing but often don't have clear insights and a real strategy for using the information to improve the customer experience. Marr also says companies make a mistake by restricting employees' authority. "This is one," he writes, "that most of us have probably experienced – a receptionist who can't offer a room upgrade because they aren't authorized to, or a retail assistant who can’t offer a refund without permission from their manager."

In "Human-Centered, Mission-Driven: What Insurers Can Learn from the Hospitality Field," Ralph Mucerino and John Bruce Tracey write that insurers should adopt a Ritz Carlton sort of mindset toward service. We won't get all the way there. Insurance is too different a business. But we can make a lot of progress.

Beyond the usual mortality risk metrics

Munich Re Life US recently collaborated with predictive health data analytics firm Klarity to analyze novel variables in the expansive UK Biobank dataset that are not typically considered in life insurance underwriting. The team filtered the Biobank data to simulate an “insurable population”. 

Grip strength, sleep duration, and resting heart rate(RHR), in particular, show compelling evidence of being effective in refining and accurately segmenting mortality risk. The study also evaluated the predictive power of muscle strength by exploring the relationship between dominant hand grip strength (GS) and mortality risk.

Overall findings

Sleep duration

Seven hours of sleep per night is associated with the lowest mortality risk, while five or fewer hours sharply increases mortality by 50%, underscoring the critical importance of sufficient sleep.  

Sufficient sleep is widely recognized as vital to physical and mental health and as protecting against cardiovascular disease and diabetes. Biobank individuals self-disclosed their typical hours of sleep, with nearly 90% of the “insurable” population reporting an average of six to eight hours per night and a median duration of seven hours.

Resting heart rate

A resting heart rate of 80-89 bpm has a nearly 50% higher relative mortality risk than a resting heart rate of 60-69 bpm. 

Given its well-established link to cardiovascular health, resting heart rate is a critical overall mortality determinant that is part of the health assessment for Biobank participants. With data available for 92% of the “insurable” pool, the average RHR for study individuals is 30 to 174 bpm, with 81% in the 60 to 100 bpm “normal” range. Our analysis shows that confirms that relative mortality risk increases as resting heart rate increases. According to the American Heart Association, a lower RHR can indicate better heart function and cardiovascular fitness.

Grip strength

Grip strength is an indicator of overall physical strength and health and can effectively segment mortality risk across age and gender. 

Dominant hand grip strength (GS) is a dimension of health not previously considered in life insurance underwriting; however, medical research suggests that it is inversely related to mortality risk in adults. With data available for over 99% of the “insurable” pool, we find that for both men and women, mortality improves as GS increases. While this metric differs significantly between males and females (GS of 42 kg vs. GS of 26 kg.), there‘s a strong relationship between GS and mortality across all age groups for both.

Poor GS is associated with an elevated mortality risk, and we find that the lowest GS category is associated with a mortality risk 1.5-2 times as high as the highest GS category, irrespective of age group and gender.

As an overall strength indicator, GS can be linked to physical function and the ability to perform the activities of daily living, such as bathing and eating, particularly or the oldest age group. For this reason, GS could be a useful addition to the traditional underwriting toolkit for mature ages, where frailty may be a concern.  

Takeaways

Sleep has long been recognized as vital to physical and mental health. Resting heart rate (RHR) is a critical determinant of overall mortality due to its link to cardiovascular health. Dominant hand grip strength (GS) is a measure of muscular strength, which medical research suggests is inversely related to mortality risk in adults. 

It bears repeating that attributes like grip strength require controlled conditions for accurate measurement, making their widespread use in life insurance underwriting challenging. Regardless, these findings highlight the potential for carriers to enhance their underwriting processes by incorporating next-gen attributes, and wearable technology provides an opportunity to access real-time, continuous data that could further enhance their predictive power for mortality risk.

Click here to read the full article, which provides further details on the data set and methodology. This study is part of our series examining the potential of third-party data sources to enhance life insurance underwriting.

Sponsored by ITL Partner: Munich Re

Pity the poor underwriters.

They're having to estimate repair and replacement costs over the next year even though it's hard to know what U.S. tariff policies will be in a week. Any changes could radically alter costs for auto, homeowners and many other lines of insurance. And that's even before you start factoring in potential second- and third-order effects, such as possible sharp drops in employment and plunges in purchases of cars and homes, leading to a global recession. 

I claim no crystal ball — and, like many, mistakenly thought Trump's first-term focus on the stock market would restrain him at least a bit from implementing tariffs that investors detest. But a recent article is helping me see a through line in what look like a scattershot series of actions and rationales. By explaining Trump's motivations, the article sheds some light on how long this tumult will last. 

I'm sorry to say I'm less optimistic than I was before reading it.

It's certainly easy to be confused. Trump and some key advisers say he's just trying to negotiate reductions in trade barriers for U.S. goods — while others insist no negotiation is possible. Some say the goal is to produce a massive revenue stream, while others promote a contradictory goal: reviving American manufacturing. (The more goods are made in the U.S., the less the country imports and the less tariff revenue it receives.) 

One goal — reviving masculinity in the U.S. through the hoped-for profusion of manufacturing jobs — is so far out in the future and depends on so many variables that it strikes me as really just designed to produce chyrons like the one on Fox's "The Five" on Monday. It read, "Trump's Manly Tariffs." 

But a Stanford history professor writes in the New York Times that "there is order amid the chaos, or at least a strategy behind it. Mr. Trump’s tariffs aren’t really about tariffs. They are the opening gambit in a more ambitious plan to smash the world’s economic and geopolitical order and replace it with something intended to better serve American interests."

Jennifer Burns, who has written biographies of Milton Friedman and Ayn Rand, says the Trump plan "seeks to improve the United States’ global trading position by using tariffs and other strong-arm tactics to force the world to take a radical step: weakening the dollar via currency agreements. This devaluation, the theory goes, would make U.S. exports more competitive, put economic pressure on China and increase manufacturing in the United States."

Talk of weakening the dollar is wonky, so Trump, instead, makes impossible promises about eliminating the income tax and paying off the national debt, while his friends on Fox rhapsodize about "Manly Tariffs." 

But the existence of a unifying theory suggests that Trump hasn't been entirely making this up as he goes along, despite his seeming fickleness. The plan suggests three other things, too -— none of them good for those poor underwriters or their employers:

• That the chaos caused by the tariffs could last quite a while. Trump knew the market would crash and went ahead anyway, so he's much less likely than in his first term to change course to appease investors.
• That prices of imported car parts, lumber, etc. will still surge even if tariffs are removed. Depending on how big a devaluation of the dollar Trump forces, the price of imports could be just as high, denominated in dollars, as if he'd slapped a 25% or so tariff on them.
• That the disruption will be profound if Trump actually goes all the way with his plan. Burns quotes one of the two architects of the economic plan as writing that U.S. military protection should be used as a bargaining chip during negotiations on getting countries to increase the value of their currencies. So the international military/security order could be convulsed at the same time that global trade is being turned upside down.      

I'm still not convinced that Trump will be able to see his plan through to the end. Some of Trump's billionaire backers and some corporate leaders have begun to publicly criticize the tariffs. Even Elon Musk went after Peter Navarro, who has been a public face for Trump in arguing for tariffs. Some Republican senators have likewise complained, and Congress has the power to overturn the declaration by Trump of a national emergency that he's using to levy tariffs. 

The incipient "Hands Off" movement says 150 groups organized 1,300 anti-Trump protests over the weekend, and some drew tens of thousands of people. The group has announced what it hopes to be an even bigger set of protests in two weeks. Polls show Americans souring on Trump's economic policies, and the big round of tariffs haven't even kicked in yet. (They apply starting on Wednesday, April 9.)

Given how completely Trump has cowed Republicans in Congress, it's hard to imagine a revolt on tariffs any time soon. But the closer we get to the midterm elections next year, the more likely it is that Republicans fearing an election loss could turn on Trump if his policies aren't improving people's lives. And it's almost impossible for major gains to show up that quickly. Building a factory and staffing it up can take years — it's not like Apple can just pull its business from China next week and start assembling them in the U.S. 

Besides, what CEO is going to start planning to invest hundreds of millions or billions of dollars in U.S. factories while the new rules of the road aren't at all clear — and when a new president in 2029 could just reverse all the executive orders that Trump is using to define those new rules? It's not like CEOs want to build a factory, operate it for a year and then mothball it because it's again cheaper to do the work overseas. 

Revolts tend to happen slowly, then suddenly, so it's not possible to know when so much pressure might build that Trump would wake up one morning, declare victory and move on. But insurers are stuck in a confusing mess in the meantime, and the trouble will linger even longer the more time that passes before some degree of certainty returns — and it seems that Trump is truly committed to tariffs in his second term.

Pity the poor underwriters.

Cheers,

Paul

P.S. Here is an interesting column by Karen Tumulty at the Washington Post on the historical roots of Trump's fascination with tariffs. She traces them to the 1980s and Chrysler CEO Lee Iacocca, who complained bitterly about what he called unfair advantages given to Japanese companies, which he said would let them hollow out the U.S. economy. Instead, the U.S. economy left Japan's in the dust in the 1990s, as its insular economy collapsed — but Trump remembers the complaints, not how history played out. 

P.P.S. In other Elon Musk news, here is an update on his promised introduction of a fleet of robotaxis in June — a plan I've been updating you on and have been highly skeptical about. Some analysts say Tesla has made huge progress toward full autonomy over the past three years, but they also say the company is nowhere close to being ready for full autonomy in two months. Here is a related, 25-minute video from CNBC that shows you what Teslas can and can't do. Again, there's plenty of praise, but the money quote is from Philip Koopman, a professor at Carnegie Mellon University and an authorty on AV safety. He says of Teslas: "Without a human driver, they're completely unsafe."

When the facts change, I change my mind. What do you do? – John Maynard Keynes.

The Baby Boomer generation was the first to experiment with "the Great Risk Transfer" (GRT), a trend in which risks were transferred from institutions to individuals. In employee benefits, it manifested as workplace pensions moving away from defined benefit (DB) plans to defined contribution (DC) plans. Behavioral economics played an important role in this workplace pension risk transfer.

The moment of truth to assess the effects of the GRT and behavioral steering experiments is upon us, and the telltale signs are not too exciting. Baby Boomers are facing a retirement crisis, with fewer savings in their pension pots at the point of retirement or facing the prospect of running out of money while in retirement. Given the checkered results of the GRT, governments are embarking on a new wave of systemic pension reforms. The essence of these reforms is not to reverse the original risk transfer and make institutions fully responsible again but to strike a middle ground where they co-own some of the risk. The reforms also limit some of the autonomy individuals previously had but did not use wisely.

This article discusses risk transfer in workplace pensions, how behavioral economics influenced the course of retirement savings and retirement spending, how the original plan unfolded, and the continuing course-correction efforts.

Baby Boomers and the Retirement Crisis

Baby Boomers, currently in their late 50s to 70s, are either entering retirement or have already moved into a retired life. Representing a sizable portion of the population, they are facing a retirement crisis. Those who are at the doorstep of retirement are realizing that they have not saved enough, whereas those who are already retired are sensing the grim prospect of running out of money. They now have limited choices: postpone retirement, "unretire" and return to work to save for a few more years, or rely more heavily on Social Security for their income. Whatever the choice, they face a significantly reduced quality of life compared with what they once envisioned.

The cause of this predicament can be traced to the GRT, for which this generation became a "poster child" (euphemistically speaking) or a "lab rat" (if using a dysphemism). GRT is a term used by the Institute and Faculty of Actuaries (IFoA), UK, to refer to a trend of transferring risks from institutions—such as employers, the state, and financial service providers—to individuals. Prominent examples of the GRT trend include the steady shift from DB to DC pensions, the move from annuities to drawdown, fewer investment products with guarantees, and insurance products that are increasingly priced based on individual risk profiles rather than group-based pricing (Institute and Faculty of Actuaries, 2021).

A review of the retirement saving journey of employees after the GRT came into effect shows mixed results. The Homo Economicus population, or "economic man," representing a small fraction of the population—who is consistently rational, narrowly self-interested, and pursues subjectively defined ends optimally—might have done well by maximizing benefits from DC plans. In contrast, Homo Communis, or the common man representing the vast majority, who is far from rational, has not fared as well.

It is becoming increasingly evident that many retirees are sliding into a retirement crisis, with median savings significantly lower than what is needed to achieve a reasonable standard of living in retirement (Chater & Loewenstein, 2022). By following the principles of libertarian paternalism, these plans allowed individuals to withdraw or liquidate their pension pots when they needed money, leading to substantial retirement fund leakage. Pension experts now ruefully opine that staying with DB plans might have produced better outcomes, as those funds are professionally managed and can cross-subsidize long-lived pensioners with contributions from those who pass away early. Meanwhile, individual participants in DC plans must worry about outliving their funds.

As the experimental generation steps into retirement, another shortcoming is revealing itself. During the long accumulation phase, DC plans focused too much on the means and methods of accumulation, while the importance of decumulation was largely overlooked. The decumulation landscape, which was previously monopolized by the option to annuitize, gained another choice: withdrawing in lump sums or flexible installments. Annuities lost their appeal, and purchasing one came to be seen as a poor financial decision due to low returns on the money invested. The longevity risk of outliving one's savings faded from the personal finance risk radar.

There is now a late realization that while accumulating a pension pot was once considered the biggest challenge, an even greater challenge lies in efficiently accessing that pot to generate a regular stream of income that protects against inflation and rising expenses in retirement.

The Making of the Retirement Crisis

The workplace pension, also known as an occupational pension or employer-sponsored pension, is one of the great welfare success stories. A workplace pension requires an employer to contribute a sum of money to a pool of funds set aside to pay for a worker's future retirement benefits. The money accumulated during an employee's working years is used to pay a guaranteed, predetermined income for life once the individual reaches a certain age and retires. As labor-intensive, manufacturing-driven economic growth swept across the developed world, providing a workplace pension was seen as the perfect way to retain talent and reward employees.

When the Baby Boomers started working, the workplace pensions that existed were DB plans. Employers or institutions shouldered the entire spectrum of responsibilities, including enrollment, contributions, managing investments, and paying a sum of money on a regular basis sufficient for employees to live on for the rest of their lives. The amount of pension paid was linked to employees' wages or salaries, their length of employment, and other factors. These plans remained the most preferred or even the only workplace pension plan until labor-intensive work flourished.

DB plans started facing extreme financial stress due to the changing nature of work, an increasing old-age dependency ratio, rising life expectancy, declining fertility rates, interest rate volatility, and overall economic fragility. As pensions were guaranteed in DB plans, regulatory frameworks mandated that the funds meet their liabilities even in the worst demographic and economic scenarios. This requirement led employers to invest pension funds in assets considered safe, with minimal interest rate and inflation risks, while also producing regular cash flows to match pension payouts.

The advent of globalization and the shift in drivers of economic growth from manufacturing-based to information- and knowledge-based industries changed the course of industries. The number of new employees in traditional companies declined. The concept of employee persistence also changed, as careers no longer followed a linear pathway but were increasingly disrupted as individuals switched from one job to another. The cascading impact was that the equation between the inflow of money from contributions and the potential outflow as future pension payments began to dangerously signal a possible collapse.

This, in turn, triggered a series of reforms to reduce risk for employers and contain pension expenditures. There was a drive to gradually transfer the financial burden of retirement funding and related risks onto the employees, who were the participants and beneficiaries of the plans. What followed was a well-orchestrated shift from DB to DC plans. Institutions started to mitigate their risk by closing DB plans and sponsoring DC plans to avoid liability shortfalls.

In DC plans, both the employer and employee contribute to the employee's retirement account, and the funds are invested across various financial instruments. The employee controls the amount contributed, chooses where it is invested, and bears the investment risk. The pension benefits the participant receives are based on the amount contributed, as well as factors such as market performance, inflation, fees, and taxes. The transfer of financial risk was candy-wrapped and made to appear lucrative through the application of behavioral economics principles, creating the impression that it provided employees with more freedom, control, choice, higher returns, and personalization.

The Effects of the Great Risk Transfer

The risk transfer conveniently shifted the herculean task of managing long-term retirement finances from qualified professionals and institutions to amateurs and unskilled individuals. Workplace pension plans comprise two phases: accumulation, when money is saved during the working life to create a pension pot or nest egg; and decumulation, when the pension pot is accessed, and the nest egg is hatched to generate a structured payment after retirement. 

In DB plans, both phases remain closely integrated. The plan is tightly controlled and inflexible, giving employees no control or decision-making power. Despite appearing authoritarian or overly paternalistic, these plans were well-designed to account for the limitations of individual participants and were inherently more forgiving in nature. They provided equal financial security to employees across a wide spectrum of financial literacy. An employee participating in a DB plan could be unskilled or inexperienced in managing long-term finances, yet the plan ensured a financial safety net for the rest of their life.

In contrast, DC plans follow the principles of libertarian paternalism—an approach that respects individual freedom while aiming to nudge people's choices in a direction that enhances their well-being. The plan offers employees ownership, greater choice, control, and flexibility. The two phases of the plan are unbundled, and the employee must make decisions during both the accumulation and decumulation phases. Though the plan is more accommodating, it is unforgiving if employees make mistakes. Employees must actively enroll in the plan, determine the appropriate savings amount, efficiently manage their investment portfolio for several years, and then make the right decisions on how to access their retirement funds—whether by converting their pension pot into an annuity for regular income, opting for a lump sum, or choosing flexible withdrawals. If they misstep in any of these activities, the consequence is a punishment in the form of an inevitable retirement crisis faded from the personal finance risk radar.

Finding the Fingerprints of Behavioral Steering 

Given the complexities of risk transfer, the application of behavioral economics played a significant role in shifting risk in workplace pensions. While institutions were primarily motivated by reducing their pension liabilities, there was no equally strong incentive for individuals to willingly assume these risks. Therefore, a compelling motive had to be created to ensure the transfer happened seamlessly. To encourage participation in DC plans, they were marketed through advertisements and supported by regulatory guardrails, giving individuals a sense of security that these plans were highly regulated and safe. The relatively strong economic growth, including multiple bull-market runs in the 1980s and 1990s, acted as a macroeconomic catalyst for the expansion of these plans. This created a "steroid effect," reinforcing the belief that taking on risk and participating in DC plans would lead to a better retirement outcome than remaining in DB plans.

Employers, as sponsors of the plans, partnered with private plan providers to hold company-wide meetings introducing DC plans and encouraging participation. The benefit illustrations projected higher returns from stock market investments compared with the lower but stable returns from traditional pensions. Beyond the potential for higher returns, the plans offered employees a compelling value proposition, including tax-deferred benefits, employer-matching contributions, and plan portability. Packed with hard-to-resist benefits, DC plans quickly became the most common employer-sponsored retirement plan.

Despite the risk transfer, employers had a fiduciary responsibility toward their employees that they had to comply with. They still had the responsibility to provide qualified investment choices, guide employees toward making the right decisions, and ensure that funds were invested appropriately and that the plans were administered judiciously. In the early days, when new-age technologies had not emerged, marketing strategies, advertising, and providing in-person advice through certified advisors played an important role.

After new technologies emerged and do-it-yourself interfaces became the default choice for user engagement, any user interface, from organizational websites to mobile apps, became a digital choice architecture designed to nudge individuals. Choice architecture is the background against which an individual makes decisions and has major consequences for both decisions and outcomes. Changes to the user-interface design elements to guide people's behavior in a digital choice environment were much easier to implement. Consequently, digital behavior-steering initiatives were developed to nudge and steer individuals toward making rational choices. Given the vast time period, spanning several decades, and the comprehensive coverage of subject participants representing different demographic, geographic, psychographic, and behavioral segments, retirement savings became a perfect platform for conducting behavioral-steering experiments.

Choice architecture and nudge prompts were created to improve the participation of individuals and the decisions made by automatically enrolling an employee into a DC plan, creating curated default investment options to simplify the investment process, automatically escalating the contribution made, and, when the participant opted out, re-enrolling or re-escalating the contribution rate. These initiatives showed very promising results and gave confidence that behavioral steering could make up for the deficit in financial knowledge or rational boundedness that individuals are known for.

Behavioral Steering in Workplace Pensions: A Mistake or Misdirection?

Though the GRT was implemented due to economic considerations, the flaw was in the conception and execution in workplace pensions. The advocacy of libertarian paternalism and behavioral steering propositions for retirement did not work as well as initially believed (Chater & Loewenstein, 2022). The approaches were supposedly modeled to be a success based on Homo Economicus. On the contrary, the real world is represented by Homo Communis. These individuals are easily influenced by social and emotional factors and are limited by their own cognitive biases.

In addition, the supposed conclusions on the success of behavioral steering were based on limited-period studies, not longitudinal studies that reflect the multi-decadal time periods that a typical retirement saving accumulation spans across. Only a marginal percentage of the participants in DC plans are financially skilled, whereas the vast majority is unskilled and susceptible to making sub-optimal decisions. Even among those who are supposedly skilled in handling financial decisions, many are only skilled in managing immediate, short-term, or medium-term finances. Retirement saving is altogether a different beast to tame, requiring individuals to have the foresight to plan and pursue the planned path for several decades. This is an extremely rare skill and discipline that even an average Homo Economicus might find lacking.

The very hypothesis that, with the right interventions, individuals—irrespective of their financial knowledge and mental acuity—can make the right decisions to solve the retirement challenge and tackle the risks on their own was flawed and did not yield a binary result. While economic man participating in DC plans benefited immensely by judiciously leveraging deferred taxation, employer match, and better investment options, a rationally bounded Homo Communis choked due to the inability to function under the burden of excessive freedom. The decisions made by them were predominantly driven by immediacy bias, influenced by what they were going through at any given moment.

Research shows that nudges successfully influence initial decisions but lose steam over time. Nudges can make choices more likely but not the behavior that follows them (Polman & Maglio, 2024). Although nudges indeed make people more likely to select the targeted option, they use it less often and for less time compared with people who made the choice without a nudge. The lack of conscious effort might lead people to feel disconnected from their choices, potentially reducing their engagement. It is now acknowledged that the previous line of thinking—that many of society's most pressing problems can be addressed cheaply and effectively at the level of the individual, without modifying the system in which the individual operates—was a mistake (Chater & Loewenstein, 2022). Nudges can contribute substantially to fixing a "broken" policy by helping people make better choices. But behavioral economics, more broadly, should, in the longer term, also help shape the formulation and direction of policy (Loewenstein et al., 2016). Heavy-handed policies that remove individual choice seem to produce superior outcomes compared with nudge approaches that stop short of "forbidding any options" and are "easy and cheap to avoid" (Loewenstein et al., 2016).

It was probably not an optimal thought in the first place to believe that, with the right choice architecture, decisions that span several decades concerning retirement savings could be simplified into a one-time activity requiring a spot decision. To explain this in the context of the Behavior Grid designed by B.J. Fogg, it is like converting the retirement challenge into a dot behavioral change (Fogg, 2007). The grid defines three types of behavioral changes: dot, span, and path, which represent changes made respectively once, for a specific period, or for the long term starting now. The behavioral steering approaches attempted to convert retirement savings into a dot-behavior change, which the participant had to do once and then forget about it. While dot-behavior change might work for something like installing solar panels on a house, it has limited appeal for situations like retirement savings and spending, which require regular follow-through for several decades—essentially a path-behavioral change. According to Fogg, behavior change is a product of three factors: motivation, ability, and prompts (Fogg, 2009). The problem with the behavioral steering approaches experimented with in the DC plans was that they leaned heavily on the strength of triggers or prompts for a successful, permanent behavior change, even when motivation and ability remained unclear or weak. Prompting someone toward saving for retirement is not as simple as buying a consumable.

Lessons Learned: Better Late Than Never

It is clearly emerging that, despite the appeal, the GRT and behavioral steering have not worked well, especially in driving the mass of Homo Communis toward a retirement paradise. It is now accepted that merely setting up defaults cannot guarantee retirement security. Unfortunately, this wisdom is dawning after a colossal failure, with a major percentage of Baby Boomers losing their way in the retirement maze. Governments across several countries have embarked on a new wave of pension reforms. Systemic interventions are being introduced, with some responsibilities shared by the governments or employers, and guidance is being provided. Governments are making changes to the approaches toward both accumulation and decumulation phases. 

Educating people and engaging them to make the right financial decisions is gaining strong momentum. Making qualified and impartial advice available to everyone is another idea that is taking shape. For those who are nearing retirement, assistance and guidance in the form of financial consultations, engagement to plan for retirement, advice to make informed decisions about their pensions, and, in the worst cases where individuals have not saved enough, career consultations to prolong working years are provided.

To protect the interests of those in the early years of the accumulation phase, governments are tweaking workplace pension plans to withdraw some autonomy that the employees previously had but did not use wisely, and to impose some restrictive, hard, paternalistic approaches. Some countries have reformed or are in the process of reforming their pension systems by legalizing mandatory participation. To arrest gross leakages from retirement funds in the form of early withdrawals to meet immediate needs, the retirement fund is split into two parts: flexible and inflexible. Early withdrawals are allowed only from flexible funds, and the inflexible fund is insulated from any intermittent withdrawals. The accumulated portion in the inflexible fund can be used only for the payment of a structured pension at the time of retirement.

Another important product innovation that is gaining wide attention is the creation of collective defined contribution (CDC) plans, also known as target benefit or defined ambition plans. These plans work like a hybrid between DB and DC plans. Instead of contributing to individual accounts, the participants pool their retirement contributions into a single fund. These plans spread the investment risk across all the participants. The contributions are invested to provide members with an income during retirement. The plans provide a target pension, which is based on factors such as salary, length of service, and contribution rate. The income from CDC plans is not guaranteed and may vary based on investment performance and other actuarial factors.

Conclusion

The workplace pension risk transfer effected a paradigm shift in the core philosophy of employee benefits, fiduciary responsibility, and retirement protection. Several countries across the world adopted this model, and now reforms to strike a middle path are gaining momentum. The immediate impact of these reforms will be on the residual late-boomer cohort, which is yet to retire. The lessons learned from the experimental generation will immensely help Generation X, which is currently peaking in the accumulation phase, in seeking appropriate advice and making the right accumulation, decumulation, and retirement decisions.

As for Generation Y and Generation Z, who are respectively in the beginning and growing phases of retirement savings, some new macro challenges are emerging. The nature of work is shifting from regularized employment to freelance or gig work. This poses legal challenges in the definition of employment, the employer-employee relationship, and fiduciary responsibilities. To provide retirement savings to these demographic cohorts, carefully designed regulatory frameworks and new pension plans are needed.

With work itself becoming a matter of personal choice and freedom, behavioral economics and digital choice architecture may become even more important. The personal characteristics and traits of each generation, along with their sensitivity to behavioral steering triggers, differ. Hence, providing everyone with templated advice instead of advice personalized to their situation would not resonate with them. To manage this, it is likely that future digital interfaces will be designed to engage individuals on a personal level and provide continuous behavioral steering prompts to enforce path-behavioral change. While designing any new choice architecture for behavioral steering, it must be borne in mind that it is not the thrill of the journey that matters, but the safety of the travel and reaching the right destination.

References

Chater, N., & Loewenstein, G. (2022). The i-Frame and the S-Frame: How focusing on the Individual-Level solutions has led behavioral public policy astray. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.4046264

Fogg, B. (2009). A Behavior Model for Persuasive Design. https://www.demenzemedicinagenerale.net/images/mens-sana/Captology_Fogg_Behavior_Model.pdf

Hanks, A. S., Just, D. R., Smith, L. E., & Wansink, B. (2012). Healthy convenience: nudging students toward healthier choices in the lunchroom. Journal of Public Health, 34(3), 370–376. https://doi.org/10.1093/pubmed/fds003

Institute and Faculty of Actuaries. (2021). Campaign recommendations. In Institute and Faculty of Actuaries. https://actuaries.org.uk/media/31hbykda/campaign-recommendations-april-2021.pdf

Loewenstein, G., Bryce, C., Hagmann, D., & Rajpal, S. (2014). Warning: You are About to Be Nudged. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.2417383

Loewenstein, G., & Chater, N. (2017). Putting nudges in perspective. Behavioural Public Policy, 1(1), 26–53. https://doi.org/10.1017/bpp.2016.7

Maglio, E. P. and S. J. (2024, May 26). The Problem With Behavioral Nudges. WSJ. https://www.wsj.com/economy/consumers/decision-making-research-behavior-2e5060c1

Mertens, S., Herberz, M., Hahnel, U. J. J., & Brosch, T. (2021). The effectiveness of nudging: A meta-analysis of choice architecture interventions across behavioral domains. Proceedings of the National Academy of Sciences, 119(1). https://doi.org/10.1073/pnas.2107346118

Polman, E., & Maglio, S. J. (2024, April 29). Will Your Nudge Have a Lasting Impact? Harvard Business Review. https://hbr.org/2024/04/will-your-nudge-have-a-lasting-impact

Sunstein, C. R. (2013). Nudges.gov: Behavioral Economics and Regulation. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.2220022

Insurance firms are drowning in data. The average insurer processes over 10 million transactions annually—a figure expected to rise by 29% in the next two years. And the challenge isn't just volume, it's variety, too. That's according to a recent industry report that found that two-thirds of insurance firms handle data from an average of 17 different sources in their premium payment process alone.

Data sources are multiplying, too, and outdated systems are struggling to keep up. This increase in data is driven by the rise of digital transactions, third-party data sources, and regulatory changes. Meanwhile, customer expectations for faster payouts and greater transparency are higher than ever. And new tech-driven players (insurtechs) are setting the standard for faster, more accurate and efficient processes.

New Market Pressures

Legacy systems and manual processes weren't built for this level of complexity, and the consequences are evident. Manual processes are time-consuming and error-prone, leading to delayed settlements and making it difficult to consolidate data across platforms, resulting in data silos and fragmentation. Without centralized, automated data flows, insurance firms lack real-time visibility. However, 90% of firms are considering a new reconciliation solution to address these issues and avoid being left behind during a time of competitive change.

Insurance firms that centralize data management and automate reconciliation will save time, reduce human errors, accelerate reporting, and gain deeper strategic insights—turning data complexity into a competitive advantage.

The Need for Speed

Today's customers expect claims to be settled quickly and with minimal friction. In the U.S. alone, 80% of auto insurance customers are planning to or have already left their current insurer due to the lack of speed and accuracy.

With the insurtech market expected to grow more than 50% from 2024 to 2030 in the U.S., a new benchmark has been set for speed and efficiency to keep customers happy. Digital-first platforms are processing claims in minutes, while traditional insurance firms struggle to keep up with the amount of data being processed.

Insurance firms that automate payment processing and improve back-office efficiency will reduce settlement times, strengthen cash flow, and improve customer trust. So, faster payments don't just enhance customer experience, they create financial stability and free up capital for strategic growth.

Battling Regulatory Pressure

Regulatory standards are becoming more stringent across key markets like the U.S. and the U.K. With varying requirements across states, insurance firms are facing growing pressure to demonstrate accuracy, transparency, and financial control under regulations such as the International Finance Reporting Standard (IFRS 17) and the California Consumer Privacy Act (CCPA). Yet many insurance firms still rely on manual processes for regulatory reporting, raising the risk of inaccuracies, missed deadlines, and penalties.

Implementing automation in regulatory reporting and data reconciliation allows U.S. insurance firms to maintain compliance with greater accuracy and reduce manual efforts. Real-time data validation and automated reporting tools reduce the administrative burden, enabling insurance firms to adapt quickly to changing requirements.

The Power of Centralized Control

To stay competitive, insurance firms need more than small fixes—they need smarter, faster operations. Streamlining processes, improving accuracy, and speeding up service are key to meeting rising customer and market demands. Managing data across multiple platforms and sources is messy and slows operations when time is of the essence. Centralizing data into one system and automating key processes like reconciliations can reduce errors and speed reporting. With cleaner, more connected data, insurance firms can make faster, more informed decisions and respond to market changes with confidence.

Slow payouts frustrate customers and strain cash flow. Automating reconciliations and settlement processes reduces delays, lowers costs, and improves accuracy – boosting customer trust and financial strength. At scale, automating manual processes can lead to an average cost savings of up to 30% within five years for payers.

Regulations are complex and costly to get wrong. Automating reporting and data validation simplifies compliance – reducing risk without adding to operational workloads.

Adapt or Risk Being Outpaced by Competitors

The insurance industry is at a crossroads. Spreadsheets are still integral to financial operations in 90% of organizations. Firms that cling to outdated systems will face higher costs, slower growth, and frustrated customers.

Those who embrace automation and smarter data management will operate more efficiently, improve customer satisfaction, and strengthen their market position.

In today's fast-paced digital world, insurance agencies are constantly looking for ways to improve their customer experience and meet the demand for quick service. Unfortunately, agents are wasting a lot of time on the leg work portion of customer service.

Agents spend 60% of their time each day on tasks related to customer service. That includes finding the right information about their clients' accounts, analyzing that information, communicating it to their clients, and bridging the knowledge gap so the client truly understands what they are receiving. This can equate to around five hours a day! This is a lot of really valuable time that is being lost. Not only does it slow the response time to clients, but it also takes agents away from more valuable interactions with their clients.

Fortunately, technology has played a major role in transforming how accounts are serviced and has evolved to meet customers' ever-changing wants and needs. AI now offers a chance to take this a step further.

There are myriad ways AI can be integrated into marketing and servicing workflows, ultimately freeing agents to focus on what matters most: building relationships and providing personalized advice. As most of us are familiar with AI-augmented chatbots to automate claims management and generative AI to create content, let's take it up a notch and look at how AI can be used in your client communications to ensure consistency and compliance. Using AI to do these things will save your team time while increasing customer satisfaction.

Build Consistency Guardrails

The tone and style of your agency's communications may vary depending on who is sending the emails. Some customer success representatives may take a more formal tone, while others write to their clients like they would their friends. Some may provide many details, while others may believe in keeping email communication brief and to the point. This can be frustrating for your clients. They crave consistency and want to interact with your agency as if the brand is one trusted person.

AI can provide guardrails to your communications, ensuring they remain consistent no matter who sends them. It can detect the tone in an email and suggest changes that will keep it in line with your agency's predetermined brand standards. AI can also help the customer representative properly phrase the information so that it's easier to understand.

Using AI to enforce consistency across client communications will make your customers happy. They will appreciate the steady tone and easily comprehend the information being shared. After all, the information doesn't do your clients any good if they don't understand it! Your team will also be happier because AI will save them time and effort in creating that consistency. Before AI, your team would need to manually review emails for tone and length, but now they can do it with a click of a button. This is a win-win for clients and staff!

Employ Compliance Fact-Checkers

Compliance is crucial for any business dealing with sensitive information. This is especially true for insurance agencies, whose systems house confidential identification and health information. Exposing the information in your agency management system (AMS) would not only upset your clients but could have costly consequences for them. Things like incorrect invoices, stolen bank account information and identity theft will create financial issues and definitely not lead to happy customers. Your agents are human, however, so mistakes can happen. It's easy for numbers to be typed into an email incorrectly or for the wrong customer's information to be included in an email. That's why it's imperative to use technology to safeguard against those mistakes.

Think of AI agents as your compliance fact-checkers. These tools can use the information in your AMS to ensure that you are only sharing the information you want shared. They can even flag information that may belong to another client. Agents no longer need to worry about sharing confidential information with the wrong person. Not only does this save your agency from the consequences of costly E&O issues, but it also saves your team the time of manually double- and triple-checking this information. Most importantly, your clients will be happy knowing their information is safe and secure.

Make AI Your Customer Service Partner

AI offers a wealth of opportunities for insurance agencies looking to elevate their customer service. By strategically implementing AI-augmented tools into the client communication process, agencies can free valuable time for their agents to focus on building stronger customer relationships and providing personalized advice. While it's important to be aware of the potential challenges and frustrations associated with AI, the benefits of increased efficiency, improved customer experiences, and enhanced growth opportunities make AI an invaluable tool for success.

Lately, I've been getting a lot of questions about how to leverage multi-cloud AI capabilities across platforms while minimizing complexity and cost. In my experience working with P&C insurers implementing AI strategies, success depends on strategic clarity rather than the use of the latest technology. The most effective approach balances innovation with pragmatism.

A Business-First Approach

P&C insurers succeed with AI when they start with specific business challenges rather than technology capabilities. Insurers should identify quantifiable business objectives such as claims leakage, underwriting accuracy, customer retention, or operational efficiency. Then link AI initiatives directly to business KPIs aligned on those objectives. This business-first approach will help create a balanced road map of quick wins and strategic capabilities, ensuring investments address actual business needs.

Fit-for-Purpose Strategy

Recently, a client with an M365 E5 subscription was wondering whether implementing Copilot would conflict with their AWS-based analytics platform.

A fit-for-purpose approach would allow insurers to match appropriate AI technologies to specific functions, reducing integration complexity and avoiding the "one-size-fits-all" pitfall that has derailed many AI initiatives.

Here's one way to think about the AI landscape based on the purpose it would serve:

• Core Business Processes: Leverage specialized insurance AI solutions for underwriting, claims processing, and risk assessment through embedded AI in core platforms or third-party integrations. Reserve in-house development for capabilities that create genuine market differentiation or deliver clear ROI.
• Enterprise Productivity: Use the tool aligned with the enterprise productivity suite for everyday knowledge work and collaboration.
• Advanced Analytics: Deploy models via the cloud provider's AI suite for specialized use cases, aligning with the enterprise data management platform and technology stack.

When it comes to leveraging any pre-built AI models and services, including GenAI, insurers should start with defining a framework to leverage them either as-is or fine-tuned, through hosted environments or API integrations, depending on the use case, overall cost and security requirements. The focus should be on speed to value rather than development. Then, organizations should implement a cross-cutting approach that integrates leveraged AI into and enhances solutions across functional areas.

Let's now revisit the question around Copilot and AWS. These platforms serve different purposes in the organization. Microsoft Copilot would integrate with M365 for daily productivity, while AWS would provide the infrastructure for specialized insurance analytics. Microsoft Purview, included with E5, will provide the necessary governance framework to monitor AI usage across the productivity layer. AI infrastructure on AWS will closely align with the organization's broad analytical data architecture.

This approach also helps insurers avoid the common pitfall of implementing GenAI as a technology-first initiative disconnected from real business challenges and outcomes.

Data Architecture

It is critical to prioritize data management, integration, and governance before sophisticated AI implementation. Insurers that approach data as strategic products rather than passive assets gain significant competitive advantages.

I recommend designing a unified data ecosystem connecting structured and unstructured business data into domain-specific data products that mirrors the organization's business architecture. It is also important to implement a data governance framework that ensures consistency, quality, and appropriate controls, and develop robust metadata that gives context and lineage for key data assets. Without this foundation, even the most sophisticated AI strategy will underperform against business expectations, as models will produce unreliable results.

By investing in a strong data architecture first, insurers can establish a reliable foundation for sustainable AI success.

Service Architecture

A robust service architecture enables services to deliver AI, to consume AI, and to be consumed by AI models using standardized protocols. A well-designed architecture helps transform AI from isolated experiments into scalable business capabilities, ensuring investments remain relevant as technologies evolve and new providers emerge.

AI services must be built around core business capabilities rather than technologies, and their effectiveness must be evaluated based on business metrics like loss ratio improvement and adjuster efficiency rather than technical metrics.

Core systems, data infrastructure, and AI capabilities must be connected using standard interfaces, creating an adaptive ecosystem rather than isolated point solutions. This integration serves as the glue between the functional areas mentioned earlier. It's also important to develop test-ready service endpoints and self-service validation interfaces for business users, fostering trust through transparency.

Governance protocols to address data drift, model drift, version control, and compliance-readiness should be baked into this architecture.

Conclusion

Successful P&C insurers understand that the value of AI lies not in specific vendor solutions but in the business capabilities it enables. By prioritizing data architecture and aligning efforts with business outcomes, insurers can navigate the rapidly evolving AI landscape while staying focused on what matters most: reducing complexity, controlling costs, and delivering measurable business impact early and incrementally.