The U.S. property and casualty (P&C) insurance industry wrote $975.8 billion in net premiums in 2025, up 5% year over year, and posted a net combined ratio of 92.9, its lowest in more than a decade. The net underwriting gain roughly tripled to about $63 billion. These property and casualty insurance industry statistics, drawn from AM Best, NAIC, Verisk and APCIA, and Triple-I and Milliman data, show an industry that consolidated its recovery in 2025, helped by the first Atlantic hurricane season in 10 years with no U.S. landfall. 

Below are the numbers that matter for 2026, organized by market size, profitability, catastrophe exposure, line of business, and the largest carriers.

Property and Casualty Insurance Industry Statistics: Key Numbers (2026)

Based on full-year 2025 results and first-quarter 2026 data:

1. Net premiums written (2025, U.S.): $975.8 billion, up 5% (AM Best)
2. Direct premiums written by line (2025): $1.11 trillion, up 5% (AM Best)
3. Net combined ratio (2025): 92.9, improved from 96.6 in 2024 (Verisk and APCIA; AM Best)
4. Net underwriting gain (2025): about $63 billion, up from $23 billion in 2024 (Verisk and APCIA) 
5. Net income after taxes (2025): $148 billion, down from $169 billion in 2024 (Verisk and APCIA)
6. Catastrophe losses (2025): 7.6 points on the combined ratio, down from 8.8 points in 2024 (AM Best) 
7. Largest carrier: State Farm, $113.8 billion net premiums written in 2025, up 5.6% (AM Best)  
8. Top 10 carriers: 51.4% of total P&C market share, on NAIC 2024 direct premiums written 
9. Policyholders’ surplus (year-end 2025): $1.2 trillion, up from $1.1 trillion a year earlier (Verisk and APCIA)

10. First-quarter 2026 combined ratio: 92.0, a 7 point improvement year over year (AM Best)

A combined ratio below 100 means an insurer paid out less in claims and expenses than it collected in premiums. At 92.9, the 2025 result clears that mark by more than 7 points. 

How Big Is the P&C Insurance Market?

The U.S. P&C market is measured most reliably by premiums, and net premiums written reached $975.8 billion in 2025, up 5%, according to AM Best. On a direct basis, AM Best puts 2025 direct premiums written by line at $1.11 trillion, also up 5%. 

Globally, estimates vary by research firm because of differing definitions and currency assumptions. Swiss Re Institute puts the global P&C market at about $2.4 trillion, having doubled over the past 20 years, and forecasts global non-life real premium growth at a cyclical low of 0.6% in 2026 before a modest recovery to about 1% in 2027. Commercial research houses publish figures several times larger on broader definitions. For anything you need to defend, use Swiss Re, NAIC or AM Best.

Two forces drove premium growth through 2024 and into 2025: multi-year rate increases catching up to inflation-driven loss costs, and rising insured values in homes, vehicles, and commercial property. Both are now fading. Growth slowed to 5% in 2025 from 8.7% in 2024, and to 2.9% in the first quarter of 2026. 

P&C Industry Profitability: Combined Ratio and Net Income

2025 was the industry’s best underwriting year in over a decade. The net combined ratio of 92.9 was a 3.7 point improvement on 96.6 in 2024, and the largest underwriting profit in ten years. 

Profitability improved across most of the market:

1. Net underwriting gain roughly tripled, to about $63 billion from $23 billion in 2024
2. Personal lines underwriting profit almost quadrupled to $45.7 billion, and commercial lines profit more than doubled to $19.2 billion  
3. Policyholders’ surplus rose 11.4% to $1.19 trillion, a $122.9 billion increase 
4. Net income after taxes eased to $148 billion from $169 billion, because 2024 carried larger one-off investment gains 

Personal lines drove the turnaround. Personal auto improved again in 2025, to a net combined ratio of 91.8, 3.5 points better than 2024. Homeowners posted 88.1, its strongest underwriting performance in more than ten years, even after the January 2025 Los Angeles wildfires. Commercial lines stayed profitable, and the gap with personal lines narrowed further. 

The takeaway for 2026: the profit is real, but it was weather-assisted. Verisk and APCIA attribute the 2025 result more to unusually low catastrophe losses than to any structural shift in risk. Rate increases have now turned into rate decreases, and Triple-I and Milliman forecast a 3.7% contraction in underlying P&C growth in the first half of 2026. 

Catastrophe Losses and Their Impact

Catastrophe losses remain the single biggest swing factor in P&C results. In 2025 they cost the industry 7.6 points of combined ratio, down from 8.8 points in 2024. That 1.2 point swing is most of the year’s improvement, which is why the profit is better read as a quiet year than as a new baseline. 

The January 2025 Los Angeles wildfires drove $33.3 billion of catastrophe losses in the first quarter of 2025 alone, 14.5 points of combined ratio. Then the Atlantic hurricane season produced no U.S. landfall, the first time in 10 years. That combination, a severe first quarter followed by a quiet second half, is what produced the year’s result. First-quarter 2026 catastrophe losses fell to $10.0 billion, or 4.2 points.

Homeowners is the clearest illustration. Its 2025 net combined ratio came in at 88.1, the best in more than ten years, and yet the line absorbed the largest single wildfire loss in U.S. history in the same year. Catastrophe volatility, not day-to-day claims, is what moves property margins, in both directions.

P&C Insurance by Line of Business

Personal auto is the largest single line, and its scale shapes the whole industry’s numbers.

Private passenger auto liability alone accounts for 18.9% of all P&C direct premiums, and adding auto physical damage takes auto to roughly 35% of the market, so shifts in auto pricing move the whole industry’s combined ratio. Progressive overtook State Farm as the largest private passenger auto liability writer in 2025. Workers’ compensation remains the standout profit line, while general liability and commercial auto are the lines insurers are watching most closely through 2026. 

The Largest P&C Insurance Companies in the U.S.

State Farm is still the largest P&C insurer. It wrote $113.8 billion in net premiums in 2025, up 5.6%, and cut its net combined ratio to 98.2 from 105. The top 10 groups together controlled 51.4% of the market on NAIC 2024 direct premiums written, the most recent complete NAIC market share data.

Market concentration is rising. Progressive grew net premiums 11.7% in 2025, the fastest of the 25 largest writers, and overtook State Farm as the largest private passenger auto liability writer with a 19.9% share. Farmers grew 10.4% while posting an 82.5 net combined ratio, the best of the top 25. Liberty Mutual went the other way, shrinking premiums 3.3% while improving its combined ratio 10 points to 90.2. That spread puts steady pressure on regional and mid-sized insurers to compete on price, service, and technology. 

2026 Outlook and Trends

The profitable stretch has continued into 2026, and growth has all but stopped.In the first quarter, U.S. P&C insurers posted a $16.3 billion net underwriting gain against a $1.0 billion loss a year earlier, and the combined ratio improved 7 points to 92.0. Net income more than doubled to $41.8 billion, and after-tax return on surplus reached 3.3% for the quarter. Net premiums written grew just 2.9%, to $250.9 billion.

Full-year 2025 delivered on that forecast: the lowest net combined ratio in over a decade, helped by a hurricane season that spared the U.S. and homeowners results that held up even after the Los Angeles fires. For 2026, Triple-I and Milliman forecast a 3.7% contraction in underlying P&C growth in the first half, against 1.6% growth in 2025, with recovery expected in 2027 and 2028. General liability and commercial auto remain the only major lines forecast to stay above 100.
Three trends to watch in 2026:

1. Rate decreases. Commercial premiums fell 1.2% in the first quarter of 2026, ending a 33-quarter run of increases, then fell 2.0% in the second. Commercial property dropped 6.3%, its steepest fall since 2010. Margins compress from here if losses climb. 
2. Catastrophe volatility. Wildfire, convective storm, and hurricane exposure keep property results unpredictable regardless of the underlying trend. 2025 shows the upside of a quiet year; the first quarter of that year shows the downside.
3. AI and core-system modernization. With prices falling and growth contracting, expense ratio is the lever insurers still control. That is pushing investment into agentic AI inside underwriting and claims, and into the data platforms that let insurers reprice against current loss costs rather than last year’s. 

Common Mistakes When Reading P&C Statistics

Most P&C stat pages report one number without the context that makes it usable. A few pitfalls to avoid,including the one that caught out most 2026 stat pages: underwriting profit rose in 2025 while net income fell, so quoting either one alone gets the year wrong.

1. Confusing direct and net premiums. Direct premiums written ( $1.1 trillion in 2025) count business before reinsurance; net premiums written ($975.8 billion) count it after. Both come from AM Best. They are not interchangeable, and the gap between them is not a rounding difference. 
2. Reading combined ratio in isolation. A 92.9 combined ratio is strong, but 2025 carried only a 7.6 point catastrophe load, its lightest in years. Verisk and APCIA are explicit that the result owes more to unusually low catastrophe losses than to any structural change in risk. 
3. Mixing market-size forecasts with actuals. Global “$X trillion by 2035” projections come from commercial research firms with differing methods, and they can run to twice the figure Swiss Re Institute publishes for the same market. Use them for scale, not precision , and never divide a U.S. actual by a global forecast.
4. Blending personal and commercial trends. Personal auto and homeowners moved in opposite directions from general liability and commercial auto in 2025. Industry averages hide this.
5. Treating one year as a trend. The 2024 rebound followed two loss years, and 2025 improved on it only with the weather’s help. Profitability here is cyclical and rate-dependent.

FAQ

What is the size of the U.S. P&C insurance industry?

The U.S. P&C industry wrote $975.8 billion in net premiums in 2025, up 5% on 2024, according to AM Best. On a direct basis, 2025 direct premiums written by line reached $1.11 trillion, also up 5%.

What was the P&C industry combined ratio in 2024?

The net combined ratio was 92.9 in 2025, a 3.7 point improvement on 96.6 in 2024 and the industry’s best underwriting result in more than a decade. A ratio below 100 signals an underwriting profit.

Who is the largest property and casualty insurer in the U.S.?

State Farm is the largest P&C insurer, with $113.8 billion in net premiums written in 2025, up 5.6%. Progressive is second, and overtook State Farm during 2025 as the largest writer of private passenger auto liability, the biggest single P&C line.

How much did catastrophes cost P&C insurers?

Catastrophe losses cost U.S. P&C insurers 7.6 points of combined ratio in 2025, down from 8.8 points in 2024. The January 2025 Los Angeles wildfires alone drove $33.3 billion of losses in the first quarter, and the Atlantic hurricane season then produced no U.S. landfall for the first time in 10 years. First-quarter 2026 catastrophe losses fell to $10.0 billion.

What is the biggest line of P&C insurance?

Private passenger auto liability is the largest single line, at $211 billion in 2025 direct premiums, 18.9% of the P&C total and up 5.1%. Add auto physical damage at $177.4 billion and auto accounts for roughly 35% of all P&C premiums. Homeowners multi-peril is second, at $188.9 billion and 17%.

Is the P&C insurance industry profitable in 2026?

Yes. 2025 closed with a 92.9 combined ratio and about $63 billion of underwriting gain, the industry’s best result in more than a decade, and the first quarter of 2026 improved again to 92.0 with $41.8 billion of net income. Premium growth has slowed to 2.9%, though, and general liability and commercial auto remain unprofitable.

How concentrated is the P&C market?

The top 10 insurers controlled 51.4% of total P&C market share on NAIC 2024 direct premiums written, the most recent complete NAIC data. Concentration is still rising: Progressive grew net premiums 11.7% in 2025 and Farmers 10.4%, both well ahead of the 5% industry rate.

The post Property and Casualty Insurance Industry Statistics (2026) appeared first on Sapiens.

Reinsurance in life insurance is the practice of a life insurer transferring, or ceding, part of the risk on its policies to another company, the reinsurer, in exchange for a share of the premium. It’s insurance for insurers. Life insurers use it to limit exposure on large or unusual risks, stabilize claims, free up capital, and back long-term guarantees on products that can last 50 years. In 2026, reinsurance in life insurance spans everything from classic mortality-risk sharing to the fast-growing transfer of entire annuity blocks offshore.

This guide explains why insurers use it, what risks get reinsured, the types and structures involved, how it works mechanically, the offshore boom reshaping the market, and the regulation now tightening around it.

What Is Reinsurance in Life Insurance?

Reinsurance is a contract in which one insurer (the ceding company) transfers part of its risk to another (the reinsurer, or assuming company). The ceding company pays a reinsurance premium; in return, the reinsurer takes on an agreed share of the claims.

The key point for policyholders: the relationship doesn’t change. The original insurer still owns the policy, collects the premium, and pays the claim. Reinsurance operates behind the scenes between the two insurers, so the policyholder has no contractual relationship with the reinsurer. 

Why Life Insurers Use Reinsurance

Almost every life insurer buys reinsurance. According to the American Council of Life Insurers’ 2025 Life Insurers Fact Book, 88% of life insurers with life premiums ceded at least part of those premiums in 2024, and 54% of insurers doing annuity business ceded annuity considerations, excluding deposit-type funds. The reasons are consistent:

1. Exceed the retention limit. Every insurer sets a retention limit: the largest risk it will keep on one life. Reinsurance lets it write policies above that limit by ceding the excess.
2. Stabilize results. Ceding smooths claim volatility and protects against single events causing multiple deaths.
3. Free up capital. Reinsurance can reduce required reserves, releasing capital for growth or new products.
4. Access expertise. Reinsurers bring underwriting, pricing, and mortality data that help insurers price difficult risks and enter new markets.
5. Manage catastrophe and pandemic exposure. Reinsurance provides protection against spikes in claims from unexpected events.

What Risks Get Reinsured

Life reinsurance covers more than death benefits. The risks a life insurer can transfer include:

1. Mortality risk. The risk that insured lives die sooner than expected. The classic life reinsurance.
2. Longevity risk. The opposite risk, that annuitants live longer than expected. Central to annuity and pension reinsurance.
3. Morbidity risk. The risk of illness or disability, relevant to health and disability riders.
4. Lapse and surrender risk. The risk that policyholder behavior differs from pricing assumptions.
5. Expense and investment risk. Cost overruns and, in asset-intensive deals, the performance of the assets backing the liabilities.

Types of Life Reinsurance

Reinsurance is classified two ways: by how the treaty is placed, and by how risk is shared.

How Life Reinsurance Is Structured: YRT, Coinsurance, and Modco

Life and pension business uses specific structures that most general explainers skip. The difference between them is what gets transferred: just the risk, or the risk plus the reserves and assets.

How It Works: Ceding, Retention, and Premiums

The mechanics center on the retention limit and the net amount at risk. An insurer decides how much of each policy to keep and cedes the rest. For a $10 million policy with a $2 million retention, $8 million of risk is ceded to one or more reinsurers.

The ceding company pays a reinsurance premium, typically calculated as a percentage of the risk ceded, a percentage of the premium charged, or a rate that varies by the insured’s risk classification. Tracking these cessions, premiums, and recoveries across thousands of policies and multiple treaties is a significant administrative task, and one that life carriers typically run inside systems tied closely to their actuarial and general ledger platforms. 

The Rise of Asset-Intensive and Offshore Reinsurance

The biggest shift in life reinsurance is not about mortality. It’s about assets. In asset-intensive reinsurance, insurers transfer entire blocks of annuity and long-duration liabilities, along with the assets backing them, to reinsurers that are often affiliated with asset managers. Demand from an aging population for guaranteed annuities has fueled the trend.

The scale is large and increasingly offshore:

1. Fitch Ratings puts U.S. reserves ceded at roughly $2.4 trillion in 2024, up from $2 trillion in 2023. . More than $1.1 trillion of that now sits offshore, with Bermuda accounting for 84% of it. 
2. Bermuda’s long-term reinsurance sector held about $1.52 trillion in assets as of September 2025, according to the Bermuda Monetary Authority. BILTIR, the sector’s industry association, reports that more than 80% of its ceded business originates in the U.S. 
3. Cayman is a fast-growing challenger. Reinsurance assets there reached roughly $101 billion at year-end 2025, up from $23 billion in 2020, according to CIMA statistics cited by Cayman Finance. 
4. Many offshore reinsurers are owned by or affiliated with alternative asset managers. Fitch estimated that around 40% of ceded reserves at year-end 2024 involved an entity with an alternative investment manager relationship. 

The benefits are capital efficiency and investment scale. The concerns, raised by the NAIC and the IMF, center on private-credit exposure, affiliated asset managers, valuation of private assets, and potential contagion if these interconnected structures come under stress.

Regulation and Reserve Credit

Reinsurance only helps a cedant’s balance sheet if the insurer can take reserve credit for it, and regulators control that through credit-for-reinsurance rules, which generally require the reinsurer to be accredited or to post collateral. As offshore activity has grown, oversight has tightened.

The most significant recent change is NAIC Actuarial Guideline 55 (AG 55), approved in August 2025. For the first time, it formally extends asset adequacy testing to reinsured blocks: U.S. cedants must demonstrate that liabilities transferred offshore remain fully backed by assets under moderately adverse conditions, and disclose the analysis in year-end statutory filings, starting with the 31 December 2025 annual statement. The NAIC has named life reinsurer investment practices a strategic priority for 2026.

Offshore frameworks are not unregulated. Bermuda applies risk-based capital calibrated to a 1-in-200-year event, holds EU Solvency II equivalence, and has reciprocal status with U.S. regulators, though scrutiny of asset strategies continues on both sides.

The Life Reinsurance Marketplace

Two groups dominate. Traditional life reinsurers focus on mortality and morbidity risk and include RGA, Munich Re, Swiss Re, SCOR, Hannover Re, and Pacific Life Re. A newer group of asset-intensive reinsurers affiliated with alternative asset managers focuses on annuity and long-duration blocks, frequently through Bermuda, Cayman, or Barbados vehicles.

The line between the two is blurring as traditional reinsurers build asset-intensive capabilities and asset managers expand into life risk. Sidecar structures show how capital is entering the market. One example is the Bermuda-based Chariot Reinsurance, co-sponsored by MetLife and General Atlantic, which each hold about 15% of the equity. It completed its first transaction on 1 July 2025, reinsuring around $10 billion of MetLife liabilities.

What Reinsurance Means for Policyholders

Reinsurance is largely invisible to policyholders, and that’s by design. The original insurer remains responsible for the policy and pays the claim; the reinsurer settles with the insurer, not the customer. Because life insurance is heavily regulated at the state level, reserve requirements and asset adequacy testing protect your benefits regardless of how the risk is reinsured. 

Sapiens and Reinsurance

Reinsurance administration is its own discipline, and the system behind it depends on the line of business. Life reinsurance is about managing long-term assets and liabilities, so the platforms that support it sit close to actuarial and general ledger systems. Property and casualty reinsurance is about managing unpredictable claims, so those platforms sit close to risk assessment, claims, and statutory reporting. Sapiens builds reinsurance software for insurers managing their own reinsurance programs, automating core operational processes, handling complex contracts, and reporting from a single data repository. That capability sits on the property and casualty and specialty side of our portfolio. Where Sapiens works with life, pensions, and annuities insurers is core policy administration and the operations around it.

Reinsurance keeps growing in scale and in regulatory complexity and Sapiens has over 40 years of experience navigating the industry. See how we approach reinsurance.

FAQ

What is reinsurance in life insurance?

It is a contract in which a life insurer (the ceding company) transfers part of its risk to another insurer (the reinsurer) in exchange for a share of the premium. It’s essentially insurance for insurers, used to manage risk and capital.

Why do life insurers use reinsurance?

To write policies above their retention limit, stabilize claim volatility, free up capital by reducing reserves, access reinsurer underwriting and pricing expertise, and protect against catastrophe and pandemic spikes. Most life insurers cede at least some premium.

What is the difference between treaty and facultative reinsurance?

Treaty reinsurance automatically covers a whole class of policies under a standing agreement. Facultative reinsurance assesses each policy individually, and is used for large, unusual, or high-risk cases outside the treaty.

What are YRT, coinsurance, and modified coinsurance?

YRT (yearly renewable term) cedes only mortality risk on the net amount at risk. Coinsurance cedes a proportional share of premiums, reserves, and risk, with the reinsurer holding the reserves. Modified coinsurance (modco) is the same but the cedant keeps the assets and reserves.

What is asset-intensive reinsurance?

It is the transfer of entire blocks of annuity or long-duration liabilities, along with the assets backing them, often to reinsurers affiliated with asset managers. It has driven large flows of U.S. life reserves offshore, especially to Bermuda.

Does reinsurance affect my life insurance policy?

No. The original insurer still owns your policy, collects your premium, and pays your claim. Reinsurance operates between the two insurers behind the scenes, with no direct relationship to the policyholder.

How is life reinsurance regulated?

Through credit-for-reinsurance rules that require reinsurers to be accredited or post collateral, and, increasingly, asset adequacy testing. NAIC Actuarial Guideline 55, approved in August 2025, requires U.S. cedants to show that offshore-reinsured liabilities remain backed by assets under moderately adverse conditions.

The post Reinsurance in Life Insurance: How It Works in 2026 appeared first on Sapiens.

Every regulated business runs on decisions. Who qualifies for the loan. How the claim is adjudicated. What the policy costs. These decisions are the business: its strategy, its risk appetite, and its regulatory posture, expressed as logic that runs thousands of times a day.

That logic is one of the most valuable assets a company has. And most companies don’t own it.

If you run a regulated business, you might reasonably disagree. You already handle this, you’d say. It’s configured in your loan origination system, your policy administration platform, the business applications that run your operation. The rules are in there. The decisions get made.

They do. But configuration is not the same as owning your logic, and the difference is where the value leaks away.

Those systems weren’t built to specialize in decisioning. They offer configurability, enough flexibility to get close to what you intend. And “close” is the problem. Your business writes a policy. Someone translates that policy into whatever the system can express. What runs in production is not your intent; it is the nearest approximation the tool allowed. Between what you meant and what actually happens, there is a gap. Most organizations have simply learned to live in that gap.

A platform built for decisioning closes it. Because it specializes in decisioning and carries a real modeling methodology, business intent can be captured with full fidelity: what you decide is what runs. And something more happens along the way. The rigor the methodology demands forces the policy to be explicit. It surfaces the ambiguities and edge cases the original policy quietly left unresolved, so you end up with a sharper, more considered policy than you started with.

The tool doesn’t just record your thinking faithfully. It improves it.

That fidelity compounds. When what runs is genuinely what you intended, you have real control. Real control is what makes you agile. You can change what you mean and see it in production quickly, rather than re-approximating it through configuration every time the market moves.

So why do so many settle for the facsimile? After many years in the field, I keep arriving at the same answer: the barrier was never really the technology, it was the specialization. Closing the gap between intent and production takes a rigorous, disciplined way of modeling decisions, and that expertise has always been scarce. Scarce expertise is a risk, so most organizations make do with what their core systems can configure.

The barrier was never really the technology, it was the specialization.

That is what is now changing. AI, and agentic AI in particular, is remarkably good at exactly this work: reading policy and regulation, translating it into structured decision models, and pushing the business to sharpen its thinking as it goes. The methodology that used to live with a handful of experts is moving into the tools. Full-fidelity modeling, the thing that was out of reach for all but the best-resourced teams, is becoming accessible to far more of them.

And that is only the input side.

The same intelligence that captures your intent on the way in can analyze your outcomes on the way out. Every decision produces a result. AI is just as good at studying those results, spotting where outcomes drift from what you wanted, and feeding that insight back to improve the policy and the models, so the next decisions are better. Intent flows in, outcomes flow out, and the outcomes teach the intent. A closed loop. We call it The Decision Loop.

This isn’t purely aspirational. The observability exists today: we use analytics to watch outcomes, set alerts against thresholds, and test model versions against each other to see which performs. What comes next is the intelligent automation of the feedback itself, where the analysis and the improvement close into a continuous cycle with greater insight and less manual effort.

That is the real destination, and it reframes what “owning your business logic” even means. Ownership was never about capturing your decisions once and considering the job done. It means turning your logic into a living asset: one that runs with fidelity, improves with every outcome, and adapts as fast as your business does. In time, it becomes something more — a partner that thinks alongside you, surfacing insights and testing scenarios so you make sharper decisions about where to go next.

Better decisions, compounding into advantage. That’s the opportunity in front of you.

The post You don’t own your business logic. You configured a copy of it. appeared first on Sapiens.

At the root of the problem is static operating models built around fixed rules, fixed workflows, and periodic (rather than continuous) review cycles. When operating models are static, AI doesn’t fix the weaknesses in the system, it amplifies them.  

Flawed logic runs faster and touches more policies before anyone notices the problem. Inconsistent decisioning multiplies, producing thousands of small discrepancies that erode fairness and consistency across a book of business. And compliance risk doesn’t shrink when a model can’t explain its own reasoning; it grows, because regulators, auditors, and customers cannot accept “the algorithm decided” as an answer when models hurt your business and damage your brand.  

When the Operating Model Is Dynamic, the Story Changes 

The good news is that the same AI capabilities that expose weaknesses in a static environment become genuine accelerators inside a dynamic operating model built for continuous adjustment. Pricing adjusts in near-real time as exposure trends shift, rather than lagging months behind the risk it’s meant to reflect. McKinsey estimates that automation and AI can reduce claims handling costs by 25–30% and compress processing times by 40–50%, figures that translate directly into faster payouts, lower loss-adjustment expense, and a genuinely better experience for policyholders when they need their insurer most. 

Underwriting undergoes a similar transformation. In a dynamic model, underwriting becomes explainable, auditable, and continuously improved through feedback loops that connect decisions back to actual claims outcomes. Instead of a static rulebook applied uniformly for years, the model learns, adjusts, and documents its own reasoning as new data arrives. That’s the difference between an underwriting engine that merely processes applications and one that actively gets smarter about risk over time. 

For claims, the process is adaptive rather than fixed: triage rules recalibrate as loss patterns shift, fraud detection sharpens with every new case rather than waiting for the next model refresh. As with underwriting, decisions are logged and explainable at each step, so a disputed payout or denial can be traced back to the reasoning behind it.  

Scaling, Not Adoption, Is the Differentiator 

A recent forecast Forrester report makes the stakes explicit: AI and automation improve expense ratios at the top 50 insurers by two percentage points, but only for carriers that have scaled beyond pilots into full production. Running a successful pilot in one line of business or one regional office no longer counts as transformation. The insurers capturing real value are the ones that have moved AI out of the innovation lab and into the operational core, across underwriting, claims, pricing, and service simultaneously. 

That distinction matters because adoption is no longer the differentiator, it’s scaling. Nearly every carrier has piloted some form of AI by now. Far fewer have built the dynamic operating model required to disseminate that capability across the enterprise consistently, with the governance, data infrastructure, and change management to support it. Scaling is harder than piloting, and it’s exactly where most transformation efforts stall. 

Helping Policyholders at Their Most Vulnerable  

Internal performance does not stay internal. Whatever gap exists between an insurer’s operating model and the speed of change around it eventually shows up on the customer side of the ledger. When operating models fail to keep pace with volatility, policyholders feel the consequences directly: slower claims resolution when they can least afford the delay, pricing that reflects operational drag rather than true exposure, products that no longer fit evolving risks, and service experiences that erode trust, as well as drive lapse and opt-out.  

For customers, these operational inefficiencies take on a particular weight. They surface as limited access to modern, flexible products, slow onboarding for new policies, and delayed payouts when a claim is filed. Because Life claims often carry significant emotional weight, arriving during bereavement or financial hardship, inconsistent or impersonal service does more damage than it would in a lower-stakes transaction.  

These individual failures don’t stay isolated. They compound at the industry level into the growing divide between total economic losses and the losses that are insured. Analysts now put this global protection gap at more than $900 billion, a figure that represents both a commercial shortfall and the failure of the industry’s core social purpose: to be there, reliably, when people need it the most. 

Closing the Gap Starts with the Foundation 

The way forward is clear. AI doesn’t create operational maturity, it reveals and reinforces whatever maturity already exists. Insurers that treat AI as a bolt-on to a static operating model will find their weaknesses amplified at machine speed. Insurers that invest first in a dynamic, continuously adjusting foundation will find that the same technology becomes a genuine multiplier, for their business and for the policyholders who depend on them. 

For most insurers, the question has already moved past whether to adopt AI. Most have. What’s still unresolved is whether the operating model underneath that AI can keep pace with how fast risk, regulation, and customer expectations are moving. That’s an infrastructure decision as much as a technology one, and it determines whether every future AI investment compounds returns or compounds risk. 

This is the third post in our series on operating model readiness, and the pattern holds across every carrier we work with: the ones that diagnose their foundation honestly, before scaling AI further, are the ones turning automation into measurable outperformance. The ones that skip that step tend to find out, at scale and at speed, that a faster wrong answer is still the wrong answer. 

Want to find out more about what you can do? Check out our whitepaper, “The Performance Gap,” for additional insights.   

The post Does Your AI Accelerate Solutions or Amplify the Status Quo? appeared first on Sapiens.

Climate catastrophes and geopolitical tension are reshaping pricing and loss severity in the property and casualty (P&C) sector. Annual insured losses above $100 billion are now routine. At the same time, claims inflation is outpacing earned premium growth, putting combined ratios under steady pressure. The squeeze is structural, and it’s tightening to the point that even strong rate adequacy can’t keep up when costs rise faster than premiums can be repriced.

Among life insurers, there are also challenges, along with clear bright spots. Stronger investment margins have improved performance, but they require a reset. Long‑duration liability management and portfolio positioning must adjust to a steeper yield curve, while renewed demand for savings and retirement products is lifting volumes. To absorb that growth, operating models need governance that can grow with the business, solid Asset-Liability Management (ALM), and tighter capital planning to capture the expected uplift through 2027.

Supervisory bodies and rating agencies can no longer rely solely on financials. They evaluate operational resilience and technology readiness as core signals of preparedness. What to do? How to change your organization to catch up to this new reality?

System Check: Static Foundation vs. Dynamic Operating Model

If you’re an insurance leader working through this situation, the message is clear: what worked in past cycles won’t carry you forward. Your operating model needs to change. You must learn from data and adjust across pricing, underwriting, product design, and claims, without loosening control.

Most insurers built their operating models for a different era, at a time when the priority was to record and process, not to adapt. Such static models prize stability over flexibility, fragment workflows across systems, and lean on manual controls. Under stress, they force a bad trade: speed for control, flexibility for lower cost. Worst of all, the customer feels it first. Everything slows to a crawl while the business chases compliance.

That cost is real. According to Carrier Management, legacy systems lose millions annually in hidden operational costs. They’re inefficient to run, heavy on your team’s time, and blind to what’s happening across the business in real time. Every day spent maintaining them is a day not spent growing.

A dynamic operating model improves how the business runs. Moving from project-led change to continuous capability means running on SaaS, scaling on demand, and trusting decisions your team can explain. Controls live inside the workflow; they aren’t bolted on after the fact. Data lineage is clear by default. Products iterate faster, without giving up governance.

The Payoff: Sharper Pricing and Faster Claims

For P&C insurers, change shows up first in underwriting and claims. Real-time exposure data lets your pricing team reprice risk as conditions shift, instead of waiting for the next renewal cycle to catch up to loss trends already priced into the market. Claims workflows route routine cases through automation, freeing your adjusters to focus on complex, catastrophe-driven volume spikes, cutting cycle time exactly when speed matters most.

For Life insurers, the payoff is growth without strain. As savings and retirement demand climbs, a dynamic model lets you launch new products, form distribution partnerships, and adjust ALM without a corresponding surge in headcount or manual reconciliation. Governance also stays intact because controls are embedded across the entire solution.

The shift isn’t just technical. It’s a change in how leadership thinks about resilience, not as a fixed state to defend, but as a capability to keep building. Insurers that treat their operating model as a living system, one that absorbs volatility rather than merely surviving it, will set the pace. Those still running on static foundations will spend 2026 and beyond reacting to a market that no longer waits.

Want to find out more about what you can do? Check out our whitepaper, “The Performance Gap,” for additional insights.

The post Beyond Static Operating Models that Keep Insurers Stuck appeared first on Sapiens.

McKinsey’s analysis found that 60% of an insurer’s financial performance is driven not by lines of business but by operational execution and that holds across both hard and soft cycles. The implication for senior insurance leadership is significant: the lines you write matter less than your ability to execute the pricing strategy designed for them. 

That execution rarely fails in a dramatic moment. It typically erodes through three specific decisions that almost every insurer makes, without recognising them as decisions. 

Decision 1: Tolerating the lag between rate change and live portfolio 

A pricing committee decision is not a rate change until it’s consistently reflected across every channel, system, delegated authority and quote engine. In fragmented environments, that gap is often measured in weeks or even months and varies by channel. 

The cost is not theoretical. A two-week delay in implementing a 3% rate increase means three weeks of underpriced business bound before the change takes effect. In subscription and specialty markets, the lag becomes more pronounced because price changes must flow consistently across lead and follow underwriting teams, channels, and delegated arrangements. The London Market is a particularly acute example as pricing intent passes through multiple underwriting entities before reaching the portfolio, and any one of them can dilute it. 

The board-level question is straightforward: how long does it typically take for pricing decisions to be fully implemented across every channel? If the answer is “weeks” or “it depends,” you have a measurable execution gap that no additional pricing sophistication will close. 

Decision 2: Treating underwriting exceptions as individual events 

Exceptions are normal in commercial and specialty business. An underwriter pricing a risk below standard rates because of relationship value, capacity needs, or genuine risk insight is fine. It’s how the business operates. 

The problem is that exceptions are rarely visible in aggregate until they have already shifted the portfolio. In a fragmented operating model, understanding how exceptions accumulate typically requires retrospective reporting usually quarterly or monthly, and almost never in real time. By the time the pattern becomes visible, the drift has already occurred making corrections slower, more difficult, and more disruptive. 

For CIOs and CTOs, this is fundamentally a connectivity question. You already have the data. The missing piece is connecting the systems where exceptions are recorded to the dashboards where portfolio-level risk appetite is tracked so you  see the pattern before it becomes a problem. 

Decision 3: Deploying AI on top of fragmented operations 

AI is now central to strategies for pricing, underwriting and claims. The investment is significant, but outcomes vary greatly, and the difference rarely comes down to the model itself. 

AI is an amplifier accelerating whatever operating model it is embedded in. In an organisation with connected systems, shared data foundations and strong governance, AI closes the feedback loop between pricing intent and portfolio reality faster than manual processes ever could. In fragmented organisations, however, AI amplifies fragmentation. Models trained on incomplete data reach flawed conclusions faster. Decisions made without the visibility of connected systems spread across the portfolio more rapidly. Governance gaps widen as automated decision-making begins to outpace the organisation’s ability to keep up. 

Decisions around AI investment and technology architecture are ultimately the same decision. Before approving any AI deployment in pricing or underwriting, the most important question is not “which model?” but “what will this model be connected to?” 

What “good” looks like 

The insurers maintaining stable, predictable performance through volatile conditions are not necessarily the ones with the most sophisticated pricing models or the largest data science teams. They’re the ones who have closed the gap between pricing strategy and portfolio reality at the operating-model level. 

In practical terms, that means: 

• Rate changes implemented across every channel in days instead of months with actual measurements available on demand 

• Real-time visibility into how much of the live portfolio is being priced within agreed risk appetite tolerance  

• AI-driven pricing decisions operating within governed parameters that can be demonstrated to the board at any point 

These are not aspirational capabilities. In McKinsey’s analysis, the top-quartile commercial P&C performers maintain loss ratios six percentage points below their peers, driven primarily by superior underwriting discipline and execution. On a €1 billion premium portfolio, a six-point loss ratio improvement is worth €60 million in annual underwriting performance.  

The window is now 

Swiss Re Institute projects insured natural catastrophe losses to reach $148 billion in 2026, and up to $400 billion in a peak-loss year by 2030. Insured severe convective storm losses in Europe are growing at 10% annually while insured flood losses in Asia are rising by 12%. These trends are compounding now, across portfolios that haven’t yet recalibrated. 

The insurers best positioned to capitalise are those whose pricing, underwriting, and data infrastructure are connected tightly enough to respond at speedand that investment delivers the fastest returns precisely when conditions are most volatile. 

The three decisions outlined above will be made in your organisation this quarter either intentionally or by default. 

Find out if your organisation has a stability gap. Read our report Closing the Stability Gap: How Insurers Can Turn Pricing Strategy into Consistent Performance which includes a six-question diagnostic built for senior leadership.

We work alongside P&C insurers across EMEA and APAC to close the gap between pricing decisions and portfolio outcomes with connected platforms built around the governance and visibility your board expects.

The post The Three Invisible Pricing Decisions That Erode Your Insurance Portfolio appeared first on Sapiens.

That’s not a coincidence, and it’s not a pricing failure. It’s the natural result of an industry where nearly everyone works from the same inputs. Same data. Same pricing engines. Same claims history feeding the same models. Swiss Re’s sigma research points to the structural cause: market concentration has been declining across most major markets, with the top five insurers now holding smaller shares than in 2004 in nine of eleven large markets. More players, more competition, and broadly similar tools create the conditions for price convergence. It’s structural, not temporary.

So when almost every insurer is investing in “personalisation,” it’s worth asking what that word is really doing.

Presentation is not personalisation

The industry has largely delivered on one version of personalisation: demographic segmentation, transaction history, third-party data feeds, and increasingly sophisticated targeting. It lifts conversion rates and makes communications feel more relevant. That’s useful. But it isn’t differentiation, because every competitor has access to the same tools and the same data sources.

If everyone works from the same inputs, better messaging isn’t personalisation. It’s presentation.

Real personalisation is when data changes decisions. That goes well beyond what you say to a customer. It shapes what you actually offer them, by knowing which covers are genuinely relevant rather than serving a generic menu. It shapes how you price. And it shapes when you engage: an experience that adapts in real time as circumstances change, not a slightly better script wrapped around a standard product. This is what hyper-relevance really means: the right solution for the right customer, in the moments that matter.

The demand is there, even if customers can’t always name it. A 2025 Deloitte survey on P&C insurance found that almost 60% of customers want personalised products, yet over 70% still believe standard products meet their needs. That gap between latent demand and a market that hasn’t shown people what’s possible is a real opportunity. But it will only be captured by insurers whose operating models can support it.

The operating model is the congestion point

Here’s the part most personalisation strategies skip over. In EMEA and APAC, most P&C insurers are running personalisation initiatives on top of operating models that were never built for them. Pricing, underwriting logic, product configuration, and claims feedback sit in separate systems, updated on separate cycles, governed by separate teams. The signals exist. They just can’t flow into decisions quickly enough to matter.

We call this the Decision Gap. Pricing, underwriting, and claims signals become disconnected. Risk selection grows inconsistent. Customers end up with products that no longer reflect their actual needs or risk profile. Personalisation becomes a front-end promise the back-end can’t keep.

This is why the difference between leaders and everyone else rarely comes down to customer-facing technology. Capgemini’s World Insurance report found that only 5% of insurers globally deliver truly outstanding customer experience, and that best-in-class insurers achieve 38% higher Net Promoter Scores and 11% lower expense ratios than their mainstream peers. That difference is a function of operating model maturity, not a slicker front end. McKinsey’s research raises the stakes further: over the past five years, insurance sector leaders in AI and technology generated 6.1 times the total shareholder return of laggards, a spread wider than in most other sectors. The gap isn’t narrowing. It’s compounding.

AI accelerates whatever you already have

There’s real and justified enthusiasm for AI-driven personalisation. There’s also a risk many insurers are underestimating.

Embed AI into a fragmented operating model where decisioning logic is siloed and data doesn’t flow cleanly between functions, and you don’t fix the underlying problem. You accelerate it. You speed up flawed underwriting logic. You scale inconsistent pricing. You produce bad decisions faster.

The numbers show how wide the gap between ambition and readiness has become. A McKinsey survey of more than 50 leaders from the largest European insurer groups found that over half expect AI to deliver productivity gains of 10 to 20%. Yet only a third have initial use cases in production, and 60% describe their traditional data as merely “evolving.” The ambition is clear. The foundation that would make it real is, for many, still a work in progress.

The insurers who will make AI work for personalisation are embedding agentic AI into the core of their policy administration and decisioning building systems where every interaction generates a signal that signal informs the next decision, so product and pricing logic adapts continuously. The human stays in the loop; the underwriter’s expertise trains the system, and the system handles the scale no team can. At that point, personalisation stops being a feature you bolt on. It becomes the way the business runs.

Fix the operating system, close the gap

The personalisation gap in P&C is real. But at its core it isn’t a data problem or a technology problem. It’s an operating system problem.

The insurers who close it will stop treating personalisation as a customer experience initiative and start treating it as an operating model outcome. When data changes decisions at every point in the value chain, and those decisions adapt continuously as customer and market signals shift, personalisation stops being something you do to customers. It becomes how you run.

That takes capabilities that work together rather than in isolation: cover selection based on a customer’s real profile, adaptive product logic that evolves without cancellation and reissuance, real-time pricing that reflects current exposure, and product versioning that updates continuously without disrupting the existing book. The technical components already exist. What most insurers are missing is the operating system that connects them.

This is only the starting point. For the full picture, including where personalisation breaks and what structural readiness looks like, read the whitepaper, Pinpointing Your P&C Personalisation Problem.

The post Your P&C Personalisation Problem Isn’t a Data Problem appeared first on Sapiens.

If you’re replacing your PAS in 2026, you’re making a 10-to-15-year architecture decision at the exact moment the ground is shifting under the market. This guide shows you how to choose a platform that supports the products you haven’t launched yet, not just the ones you run today. 

What a Life and Annuity Policy Administration System Does

A PAS sits at the center of insurer operations. At its core, it handles numerous domains.

• Policy issuance and maintenance: new business processing, endorsements, conversions
• Premium billing and collections: billing cycles, payment processing, lapse management
• Claims and benefits administration: death benefit payouts, annuity income calculations, surrender processing
• Rider and product logic: living benefit riders, COLA adjustments, waiver provisions
• Regulatory and compliance reporting: state filings, NAIC compliance, reserve calculations
• Agent and distributor management: commission processing, producer portal access
• Customer and policy servicing: self-service portals, correspondence, document generation

Modern platforms extend beyond core administration to support digital distribution, API connectivity with third-party data sources and insurtech partners, and cloud-native deployment. Legacy systems, by contrast, often handle only core processing and typically require heavy middleware to integrate with peripheral systems.

Why U.S. Insurers Are Re-Evaluating Their PAS in 2026

Several forces are driving PAS modernization across the life and annuity market:

1. Product complexity has outgrown legacy platforms

Indexed universal life (IUL), registered index-linked annuities (RILAs), and hybrid products need calculation engines that most older systems weren’t built to run. Every workaround you add to launch one on a legacy engine is an operational risk you’ll carry for years. 

2. Speed to market has become  a competitive differentiator

On a modern platform, you configure new products in days instead of the months of custom development a legacy system demands. Those delays compound: every launch you can’t ship on time is ground your competitors take first. 

3. Cloud and SaaS is now mainstream

The insurance industry’s hesitation around cloud deployment has largely faded. Most insurers evaluating PAS replacements today are considering SaaS-first or cloud-ready solutions rather than on-premise systems.

4. Regulatory pressure is increasing

Changing reserve requirements, state compliance mandates, and federal reporting obligations demand systems that can adapt quickly. Legacy systems with hardcoded logic can create compliance risks.

5. Legacy vendor support is shrinking

Several long-standing PAS vendors have been acquired, consolidated, or have reduced active product development. Insurers on those platforms face escalating support costs without any clear modernization roadmap.

Key Capabilities to Look for in a Life and Annuity PAS

Not all platforms are created equal. When evaluating vendors, certain capabilities have a direct impact on long-term performance,

Product configuration engine

Can business users configure new products and riders without relying on custom code? A rules-based product configuration engine is the single most important driver of faster  time-to-market.

Calculation accuracy and auditability

Annuity crediting, reserve calculations, and benefit projections must be accurate and auditable. Vendors should be able to clearly demonstrate their actuarial validation processes and testing documentation.

API architecture

A modern PAS gives you clean APIs that connect to distribution platforms, digital applications, e-signature tools, and third-party data providers. Treat proprietary integration layers as a red flag.

Cloud deployment and scalability

SaaS or cloud-native platforms take the infrastructure burden off your team and let you upgrade faster. Ask exactly how the vendor handles upgrades, and what happens to your customizations when they do.

Regulatory update management

Insurance regulations related to filing requirements, NAIC model laws, and taxes continue to evolve. Vendors offering strong regulatory support can significantly  reduce your compliance risk.

Migration support and data conversion

Insurers replacing a PAS will face complex data migration challenges. Vendor experience with in-force block migrations rather than just greenfield implementations alone is especially important.

Reporting and analytics

You need to see your data in real time. Look for a PAS with reporting built in, or one that connects cleanly to your enterprise data warehouse.

How to Evaluate a PAS for Your Insurer’s Needs

Step 1: First define your product roadmap

A PAS evaluation with vendor demos before clearly defining your product strategy is approaching it backwards. First map your three- to five- year product roadmap, then evaluate which platforms can realistically support it.

Step 2: Separate must-haves from nice-to-haves

Build a weighted requirement matrix. Treat core capabilities like calculation engine accuracy, regulatory update support, and data migration experience as non-negotiable. Everything else should be prioritized based on business importance.

Step 3: Stress-test the product configuration process

Ask vendors to configure a specific product. This ideally should be one that represents your most complex offering. Pay close attention to whether the process  requires developer involvement or can be handled directly by business users.

Step 4: Evaluate the vendor’s financial stability and roadmap

A PAS is typically a 10 to 15 year commitment. A vendor’s funding, ownership structure, customer base, and active development roadmap matter just as much as its current set of features.

Step 5: Talk to existing clients

References provided by vendors are naturally curated. Whenever possible, connect with carriers already using the platform through industry conferences and peer networks to gain more candid unbiased feedback.

Step 6: Model the total cost of ownership (TCO)

License fees are rarely the largest expense. Be sure to account for implementation, data migration, training, ongoing support, and customization maintenance over a 5-year period.

Common Mistakes in Life and Annuity PAS Selection

Prioritizing features over architecture

A feature-rich platform built on a rigid architecture will limit you faster than a simpler platform built for flexibility. It is essential to understand the system’s data model and configurability before counting features.

Underestimating data migration complexity

In-force policy migration is consistently the highest-risk element of any PAS replacement. Insurers that underinvest in data discovery, mapping, and validation during the selection phase often face costly surprises later in the implementation.

Selecting based on current product mix

The platform you choose today must support not only your existing products but future offerings  you haven’t yet launched. Evaluating long-term flexibility is far more important than focusing just on current fit.

Ignoring the implementation partner ecosystem

Some vendors have strong products but have limited implementation support networks. Others provide access to broader partner ecosystems that can reduce risk. This factor can have a major impact on project execution.

Where Sapiens Fits

You get a platform for Life & Annuities built for U.S. and Canadian insurers running individual life and annuity products. Your business users configure complex riders, indexed crediting logic, and in-force block conversions without waiting on custom code.

That configurability is the point. When a new indexed product or rider structure comes up, you model it in the platform and take it to market in days, not the months a legacy engine would need for custom development. Crediting logic, reserve calculations, and benefit projections stay accurate and auditable, so your actuarial and compliance teams can trace every number. And because the platform is designed for real in-force complexity, you can consolidate closed blocks onto it rather than keeping legacy systems alive just to service old policies. It runs on the same foundation as more than 600 insurers across 30 countries, backed by 40+ years of insurance ontology.

If you’re weighing a PAS replacement, our platform for Life and Annuities can support the products you run today and the ones you haven’t launched yet.

The post How to Choose a Life and Annuity Policy Administration System for U.S. Insurers in 2026 appeared first on Sapiens.

What a Closed Block Is and Why It Exists

The term “closed block” has a specific regulatory meaning in demutualization. It refers to a protected portfolio of participating policies carved out during the mutual-to-stock conversion, with its own assets and liabilities managed separately. More broadly,  in industry usage, a “closed block” describes any group of policies no longer open to new business.

Closed blocks typically arise in several scenarios:

Demutualization

When a mutual insurer converts to a stock company, participating policies are often placed in a closed block with ring-fenced assets to protect policyholder dividend expectations.

Portfolio acquisitions

When one insurer acquires a block of policies from another, the acquired block is frequently managed as a closed portfolio, often on the seller’s legacy platform at first, with conversion to the buyer’s platform taking place at a later time.

Product line exits

Insurers that discontinue a product line (for example, moving out of the individual whole life market) create a de facto closed block of in-force policies that must continue to be serviced indefinitely.

Reinsurance and assumption transactions

Assumption reinsurance transactions, in which policy obligations are transferred from one insurer to another, often create closed blocks on the assuming insurer’s platform.

Across all of these scenarios, the administration challenge is the same: managing a complex but static portfolio over a long period of time often on aging or poorly documented technology while keeping costs low and service quality high.

Closed Block Administration Challenges

Closed blocks may be static from a sales perspective but they are operationally demanding. They come with several key challenges:

Legacy system dependency

Closed blocks are often left on aging legacy platforms, sometimes the original platform used when the insurer first issued the policies decades ago. These systems may have limited, if any, vendor support, undocumented processes, and feweremployees who still understand how the system works.

Product complexity and undocumented logic

Many closed blocks contain policies issued across multiple generations  of product design, each with its own unique features, riders, and administrative rules that were never formally documented. In some cases, insurers may have to depend on the institutional knowledge of employees who have since retired.

Cost management pressure

Because closed blocks don’t generate new business, administration costs directly impact profitability. Insurers are under constant pressure to reduce per-policy administrative service costs while still maintaining high customer service standards.

Long policy durations

Whole life and participating policies in a closed block may remain active for 20 to 40 years. That means that administration platforms must remain operational for extremely long periods of time.

Regulatory and dividend obligations

Participating policies often carry dividend obligations that must be accurately calculated, allocated, and paid. Demutualization closed blocks also have specific regulatory oversight of dividend scales and block asset management.

Data completeness and integrity

Policy data that originated decades ago is often inconsistent, incomplete, or coded in outdated formats. Current administration platforms must be able to handle these gaps efficiently.

Why Insurers Convert Closed Blocks to Modern Platforms

Despite the complexity involved, many insurers choose to convert closed blocks from legacy systems to modern administration platforms. 

Legacy system sunset

When a legacy policy administration system (PAS) is no longer supported by the vendor, the insurer must decide whether to convert or accept the ongoing risk of operating an unsupported system.

Cost reduction

Modern platforms, particularly cloud-based SaaS solutions, typically have lower per-policy operating costs than maintaining customized legacy systems. For large closed blocks, the savings justify a substantial upfront conversion investment.

Capability gaps

Legacy systems may struggle to meet regulatory demands. They may lack the ability to support digital self-service tools for policyholders or integrations with modern billing and payment infrastructure.

Portfolio acquisition integration

Insurers that acquire a closed block on a seller’s platform will eventually need to convert it to their own infrastructure in order to avoid ongoing platform licensing costs and operational complexity.

Risk reduction

Legacy systems with no vendor support and minimal internal expertise create significant operational risk. If critical system failures occur, insurers may have little to no path to an effective vendor-assisted resolution.

Technical Requirements for Closed Block Administration Software

A platform capable of closed block administration must meet specific requirements that differ from platforms designed for new business.

Historical product support

The platform must be configurable enough to support product designs that may be 30 to 50 years old, including dividend scales, participating policy structures, and rider features that are no longer commonly used in modern products.

Dividend calculation engine

For participating whole life blocks, the platform must support dividend scale calculations including the key components of illustrated scales such as mortality, interest, expense along with the allocation of dividends to individual policies.

Minimal new business functionality

Unlike a full PAS, a closed block administration platform can be optimized for in-force servicing functions such as claims, billing, correspondence, and policy loan processing, without the need for new business intake workflows.

Data migration and conversion tooling

Converting from a legacy system requires specialized tools for data extraction, transformation, and validation. A vendor’s experience and methodology for handling complex legacy data models can significantly impact conversion success.

Low per-policy cost at scale

The economics of closed block administration favor platforms with efficient processing architectures and cloud-based infrastructure. SaaS pricing models based on per-policy cost  may offer more predictable revenue than a capital-intensive licensed system.

Long-term vendor viability

Because some closed blocks may remain active for 20 to 40 years, vendor financial stability and ownership structure become especially important evaluation criteria.

Software Options: Approaches and Trade-offs

Insurers managing closed blocks have several strategic options.

Option 1: Keep on the existing legacy platform

This is typically the lowest cost option in the short-term but carries the highest long-term risk. It may be acceptable if the platform still has reliable vendor support and the block is small enough to wind down within a predictable time frame.

Option 2: Convert to a modern PAS

A full migration to a modern administration platform can reduce cost, deliver access to advanced capabilities, and provide ongoing vendor support. It requires the highest upfront investment, but offers the strongest long-term economics for large or long-duration blocks. 

Option 3: Outsource to a closed block administration specialist

Some insurers choose to outsource closed block administration entirely to a third-party administrator (TPA) with specific expertise in this area. This eliminates platform conversion  but creates ongoing vendor dependency and may reduce operational control.

Option 4: Convert to a cloud-based SaaS platform through phased migration

A phased migration strategy converts the block in segments by product type, issue date, or other segment criteria,  reducing the risk of any single cutover while gradually modernizing infrastructure.

The optimal strategy depends on factors such as the block size, remaining duration, the condition or status of the legacy system, and the insurer’s broader technology strategy.

What a Closed Block Conversion Project Involves

A closed block conversion has many similarities to a full PAS replacement, but it also comes with its own distinct challenges.

Product archaeology

Before a new platform can be configured, the full product inventory must be documented including every product, rider, dividend scale, and administrative rule still in force. This “product archaeology” phase is often more intensive for closed blocks than for active portfolios because documentation is frequently inconsistent or incomplete.

Data profiling and remediation

Policy data that has accumulated over decades will have gaps, inconsistencies, and legacy and outdated codes. Data profiling identifies these issues while remediation resolves them through automated transformation rules or manual review for complex cases.

Calculation parity testing

The new platform must produce outputs that match the legacy system for all policy types. This is tested through parallel testing and comparing calculations at the individual policy level.

Dividend scale migration (for participating blocks)

Dividend scales must be accurately recreated in the new platform with actuarial validation essential to confirm that future dividend allocations will be calculated correctly.

Regulatory review

For demutualization-related closed blocks, regulators may require formal review of the conversion plan. This introduces an added layer of governance not required in greenfield platform implementations.

Phased cutover and stabilization

As with any migration, phased implementation by product or policy helps reduce single-event risk. After migration, a stabilization period with greater oversight and support is standard.

Key Questions for Vendors

• Have you administered or converted a participating whole life closed block? If so, what was the block size and when were the products issued?
• How do you handle dividend scale recreation and actuarial validation?
• What is your per-policy cost model for a SaaS-administered closed block?
• How do you handle undocumented legacy product logic discovered during data profiling?

Where Sapiens Fits

In a closed block conversion, the hard part is rarely the migration. It is the product archaeology: decades of riders, dividend scales, and undocumented rules. Our platform for Life & Annuities, backed by 40+ years of insurance ontology, uses a configurable product factory and calculation engine to reconstruct legacy product structures rather than approximate them. 

You are not betting on a roadmap. More than 600 insurers in over 30 countries run on Sapiens today, including life insurers in North America and Canada, where participating whole life blocks remain significant. That production track record is what matters when a block has to be administered accurately for the next 20 to 40 years. 

The right strategy depends on your block. Our advisory team can map the economics and operational trade-offs for your specific portfolio, so you can see the full picture before you commit. Start the conversation.

The post Closed Block Life Insurance Conversion: Choosing Software for North American Insurers appeared first on Sapiens.

Why Annuity Type Matters for Software Selection

Two major annuity categories (fixed, and indexed) represent fundamentally different contractual obligations, calculation methodologies, and regulatory frameworks. Each places specific demands on the administration platform:

• Fixed annuities including multi-year guaranteed annuities (MYGAs) require reliable general account crediting, surrender charge schedule enforcement, and straightforward annuitization logic
• Indexed annuities including fixed indexed annuities (FIAs) and indexed universal life (IUL)-adjacent products require segment tracking, cap/floor/participation rate logic, crediting strategy management, and point-to-point or other index-linked calculation engines

A platform optimized for fixed products will struggle with indexed crediting complexity. Conversely, a platform built around complex indexed crediting may be overbuilt and expensive for an insurer focused primarily on fixed products.

Understanding these distinctions before evaluating software prevents costly mistakes resulting from choosing a platform that cannot grow with your product roadmap.

Fixed Annuity Administration: What the Software Must Handle

Fixed annuities including single premium immediate annuities (SPIAs), deferred income annuities (DIAs), and multi-year guaranteed annuities (MYGAs) are the most straightforward to administer. Core requirements include:

General account crediting

The platform must apply declared interest rates to policy account values on the correct crediting frequency whether annual, monthly, or daily. Rate changes need to be implemented accurately and on time with audit trails in place to support regulatory compliance.

Surrender charge schedule management

Fixed annuity contracts typically include tiered surrender charge schedules that decline over time. The system must track charge periods accurately and calculate surrender values including market value adjustments (MVAs) where applicable.

Annuitization calculation and payout management

When policies enter the payout phase, the system must calculate annuity factors, apply mortality tables, and manage scheduled payment disbursements over the payout period including joint and survivor options.

1035 exchange processing

Policies exchanged under IRS Section 1035 are common in fixed annuity sales. The system must maintain a complete history of exchanges and apply the appropriate cost basis treatment for each transaction.

Regulatory reporting

Fixed annuities are subject to state insurance department oversight and IRS reporting requirements (1099-R for distributions). The platform must generate accurate regulatory information.

Indexed Annuity Administration: What the Software Must Handle

Fixed indexed annuities (FIAs) and related indexed products introduce significantly more calculation complexity. The key requirements include:

Crediting strategy engine

FIA platforms must support multiple crediting strategies including point-to-point, monthly average, monthly cap, and others. Policyholders may select from multiple strategies when the policy is issued and may reallocate funds at segment anniversaries. The system must be able to manage concurrent strategies at the policy level.

Segment tracking and anniversary processing

Each crediting segment has its own start date, end date, index value at start, and calculation parameters. The system must track segments independently and calculate credited interest at the correct anniversary date.

Index integration

The platform must be able to receive index values (S&P 500, MSCI, custom blended indices) from external data, store them accurately, and use them in crediting calculations. Data integrity in index feeds is critical as errors directly affect policy value.

Cap, floor, and participation rate management

Product terms include caps (maximum credited interest), floors (minimum, typically 0%), and participation rates (percentage of index gain credited). These parameters may change when a policy is renewed, and the system must apply the renewal terms correctly at each segment anniversary.

Rider administration

FIA products frequently include guaranteed minimum withdrawal benefit (GMWB) riders, guaranteed lifetime withdrawal benefit (GLWB) riders, and enhanced death benefit riders. These riders have their own logic and calculations and must be tracked in parallel alongside the base contract value.

RILA: An Emerging Category 

Registered index-linked annuities (RILAs) build on the indexed model while adding structure and administration complexity. RILAs provide downside protection through defined buffers or floors while providing upside exposure to index performance. However, unlike FIAs, they don’t fully protect against downside risk.

Future administration requirements for RILAs include:

• Segment-based structure similar to FIAs, with defined crediting periods
• Buffer and floor tracking where the system must calculate policyholder gains and losses relative to defined buffer levels
• Securities registration: RILAs are registered products requiring SEC/FINRA compliance infrastructure
• Index integration with market data feeds

Most legacy platforms were never built for RILA administration. Insurers launching RILAs on older platforms frequently rely on workarounds that create operational risk. This is an area where modern, flexible platforms have a meaningful advantage.

Side-by-Side Comparison: Fixed, Indexed, and RILA Administration 

Can One Platform Handle All Annuity Types?

The honest answer: some can, but most cannot do all product types equally well. Historically, the annuity administration software market developed in product silos including fixed annuity specialists and more recently, indexed product specialists. Platforms that handle multiple products are rare, and that makes them a meaningful differentiator.

Insurers managing a portfolio that spans fixed and indexed products should be asking:

• Was the platform designed for multiple products or is one product type the primary use case while the rest are add-ons?
• Is the calculation engine unified or does each product type run on a different engine within the same platform?
• Is the policyholder servicing model consistent across product types (one portal, one statement format, one servicing workflow)?
• How quickly can the platform support new features, riders, and crediting structures as consumer and advisor demand shifts? 

Some insurers deliberately run separate platforms by product type to get best-of-breed capability in each. Others prioritize operational simplicity through a single unified platform. There is no universal right answer. It depends on your product mix, operational maturity, and integration architecture.

Key Questions to Ask Annuity Administration Software Vendors

• How many active FIA policies do you currently administer across your client base?
• How does your platform handle crediting strategy elections at segment anniversary?
• What is your approach to index data feed management and error handling?
• How does your living benefit rider engine interact with base contract values?

Where Sapiens Fits

By continually investing in our annuities product innovation, you get a platform built around the flexibility you need to respond to changing market demand. We prioritize enhancements that enable you to quickly configure new features, riders, and crediting structures so you can bring competitive products to market as consumer and advisor demand evolves.

We also work closely with your organization throughout this process, from initial product design through implementation, helping them move quickly on new offerings and stay ahead in a market that continues to shift. This partnership model is what allows insurers to remain competitive as annuity trends emerge and mature.

If your portfolio spans fixed and indexed products, you don’t need a separate system for each. The Sapiens Platform for Life & Annuities runs both on one unified policy servicing model, so you can consolidate administration rather than manage separate systems. It’s grounded in 40+ years of insurance domain expertise, with more than 600 insurers across 30 countries running on Sapiens. 

The post Annuity Administration Software Guide: Fixed and Indexed appeared first on Sapiens.

Agentic AI changes that. Not by adding another tool to ask questions of, but by acting on your workflows directly. You give it an objective, and it works autonomously towards that goal, making decisions along the way rather than waiting for the next prompt. That’s the difference between automating a workflow and running one.

The Reality Your Team Is Living With

A Deloitte survey on ceded reinsurance found that 71% of participating insurers say late or incomplete data forces manual workarounds, and that 67% of all data processing across the ceded workflow still involves some form of manual handling.

The cost lands in two places. The first is recoverable leakage: when claim data doesn’t attach to the right treaty, you miss recoveries systematically. The second is the reconciliation tax, the skilled accountant time you spend every period rebuilding bordereaux and resolving discrepancies that clean data would have prevented. As Swiss Re CEO Christian Mumenthaler put it in April 2026,

“Without reliable data, AI mainly creates additional complexity, costs and frustration.”

What Agentic AI Actually Changes

Ceded reinsurance workflows are about executing processes. Agents make the difference across three capabilities. They “understand”, reading slip letters, cover notes, and bordereaux in any format and extracting treaty-aware data without re-keying. They “reason and decide”, comparing that data against treaty terms, applying attachment and retention logic, identifying recoveries, and flagging the exceptions that need your attention. And they “act”, posting to ledgers, generating statements, and routing exceptions to the right queue without hand-off delays.

What Your Operation Looks Like Afterward

Today, ceded reinsurance runs on batch cycles: allocation overnight, reconciliation at month-end, bordereaux assembled period by period. Agents deliver continuous processing instead. Policies allocate on booking. Claims match treaty terms the day they arrive. The close becomes exception driven. Agents pre-run the numbers and flag what doesn’t balance for your accountants. Your team approves rather than rebuilds.

This keeps people in the loop, not out of it. By design, the bulk of routine ceded transactions can flow through without a human touch point. The exceptions that do reach your team arrive pre-loaded with context: treaty terms, loss history, prior decisions, and the reason they were flagged. Your senior accountants spend their time on treaty edge cases and reinsurer disputes, not on building spreadsheets.

Why Data Quality Comes First

Agentic AI amplifies what it finds. Feed it clean data, and it delivers recoveries and reliable reporting. Feed it poor data like missing fields, incorrect treaty codes, unvalidated MGA bordereaux, and it produces wrong outputs at agent speed and scale.

Fix data quality issues in MGA-sourced books upstream before agents touch it. This isn’t a footnote to a deployment plan. It’s the foundation the whole plan stands on.

Where the Industry Is Heading

Deloitte’s 2026 State of AI in the Enterprise survey found that 74% of organisations plan to deploy agentic AI moderately or extensively within two years, up from 23% at the time of the survey, a tripling of deployment intent in a single year.

The insurers who use that window to close their data governance gaps will arrive at full deployment with clean inputs. Those who skip the prerequisite work will simply move faster on a flawed foundation. Governance matters here, too. As McKinsey’s State of AI Trust in 2026: Shifting to the Agentic Era report framed it in March 2026, organisations “can no longer concern themselves only with AI systems saying the wrong thing; they must also contend with systems doing the wrong thing.” Every agent action needs a timestamped audit trail and human-in-the-loop gates designed in from the start, for Schedule F, Solvency II, and reinsurer settlement alike.

Start with Five Honest Questions

Before you deploy, ask your team five things.

1. Have reinsurance costs risen even as loss ratios stayed stable?
2. Is recoverable identification systematic, or does it depend on periodic manual review?
3. How many days per period go to bordereaux assembly?
4. Can you show reinsurers in writing what validation controls govern your cession process?
5. Have you reviewed your AI governance framework specifically for autonomous agents, not just generative tools?

Answer those, and you’ll know exactly where to start. The insurers acting on them now are the ones who’ll set the terms everyone else follows.

Sapiens Is Your Partner for Autonomous Insurance

You need AI agents that can run your ceded reinsurance desk: catching recoveries and closing the books, not just demoing well.

Agents are only as capable as the domain knowledge behind them. Sapiens’ 40+ years of insurance expertise means a working understanding of how treaties, bordereaux, and cessions actually behave, so agents can be built on that foundation, rather than generic AI you’d have to teach your business from scratch.

Learn more about our reinsurance solutions.

The post Generative AI Waits for Reinsurers to Ask. Agentic AI Acts. appeared first on Sapiens.

How Accelerated Underwriting Platforms Work

At its core, an accelerated underwriting (AUW) platform acts as a decision engine. When an application is submitted, the platform:

1. Collects applicant data through electronic application, Medical Information Bureau (MIB) inquiry, prescription history, motor vehicle records (MVR) records, and other configured data sources.
2. Applies underwriting rules through a rules engine that evaluates the data. against defined risk criteria, looking for risk indicators or contraindications
3. Routes the application based on the rules output, to one of three outcomes: instant approval, referral to traditional underwriting, or decline.
4. Generates a decision record for compliance and audit purposes, where the decision logic is documented and retained.

The speed advantage is significant. Applications that qualify for straight-through processing (STP) can be approved in minutes rather than weeks. The proportion of applications that can be automatically approved depends on the insurer’s risk appetite, face amount thresholds, and the quality of data sources integrated into the platform.

Key Data Sources That Power AUW Systems

The quality of accelerated underwriting decisions is directly linked to the data sources a platform can access. Standard data sources include:

MIB

MIB records contain coded entries related to past insurance applications, identifying conditions or risk factors that were underwritten in prior applications. MIB is a near-universal data source in life underwriting.

Prescription drug history (Rx)

Prescription databases provide prescription medication history, which is a strong indicator of chronic conditions, mental health treatment, and other underwriting-relevant factors. Rx data is now a standard component of most AUW platforms.

MVR

MVR data identifies DUI history, serious traffic violations, and driving patterns associated with mortality risk. It is particularly relevant for younger applicants.

Credit-based insurance scores

In states where permitted, credit-derived risk scores are used as indicators of the overall risk profile. Regulatory restrictions vary significantly by state.

Electronic health records (EHR)

EHR integration represents the next stage of AUW data enrichment. Where applicants consent to EHR access, platforms can review actual clinical data directly without requiring a physical exam. EHR integration is still evolving from both data availability and consent infrastructure standpoints.

Wearables and behavioral data

Some insurers have begun integrating wearable device data into underwriting decisions, particularly for wellness-based programs. This remains a niche capability and raises both data quality and privacy considerations.

Mortality and predictive models

Many AUW platforms incorporate actuarially validated mortality prediction models that are either developed internally, licensed from reinsurers (e.g., SCOR, Munich Re, RGA), or embedded in the platform vendor’s algorithm layer.

Core Capabilities of an AUW Platform

A production-ready accelerated underwriting platform must include the following core capabilities:

Rules engine with version control

Underwriting rules evolve over time. The platform should allow rules to be updated without requiring a software deployment, while maintaining full version history and audit capabilities. Business users should also be able to configure rules accordingly.

Multi-source data orchestration

The platform should be able to manage parallel data inquiries, handle API failures and timeouts gracefully, and reconcile results from multiple sources before applying rules.

Decisioning transparency and audit trail

Regulatory requirements and adverse action notice obligations mean that every decision must be fully traceable. The platform should clearly explain why a decision was made in terms that are auditable.

Referral workflow management

Applications that cannot be automatically approved must be routed cleanly to human underwriters who have full visibility of data and decision history. Poor referral workflows cause frustration among underwriters as well as a breakdown of the process.

Integration with policy administration and distribution systems

AUW platforms don’t operate in isolation. They need to integrate with the insurer’s policy administration system (PAS) to get new business and issue approved policies, as well as e-application platforms and agent portals.

Compliance controls

State regulatory variations in data use, adverse action notice requirements, and prohibited underwriting factors must be manageable within the platform’s architecture without requiring each state to have custom code.

Accelerated Underwriting vs. Simplified Issue vs. Fully Underwritten

These three approaches are often confused, but understanding the distinctions clarifies what a platform is actually being asked to do.

AUW is different from simplified issue in that it involves actual underwriting judgment based on data but without the need for physical exams. An insurer using AUW is not lowering its risk standards; it’s applying the same standards through a more efficient process.

How to Evaluate an Accelerated Underwriting Platform

1. Define the scope first

   Which face amount bands will be eligible for AUW? Which products? What risk classes? Platform requirements can differ significantly depending on whether you’re building a $500K term AUW program or a multi-product program in excess of $2M in coverage.

2. Audit your data infrastructure

   AUW platforms are only as good as the data they receive. Before evaluating providers, understand your current MIB, Rx, and MVR connectivity and any gaps in your data vendor relationships.

3. Evaluate rules engine usability

   Ask for a demonstration on how a business user rather than a developer would modify an underwriting rule. If the answer requires IT involvement, consider the operational implications of that dependency.

4. Assess decisioning transparency

   Request a sample adverse action decision record. It should be fully traceable showing which data inputs triggered which rules, with decision rationale documented in plain language.

5. Evaluate PAS integration path

   Understand how the AUW platform connects to your existing or planned PAS. API-first platforms reduce integration friction, while platforms that rely on proprietary middleware create long-term technical debt.

6. Evaluate reinsurer relationships and data obligations

   If the platform is reinsurer-backed, understand what data the reinsurer receives and whether that creates any obligations in your reinsurance arrangements.

Integration Considerations with Your Policy Administration System

The handoff between an approved AUW decision and policy issuance in the PAS is a critical integration point. Breakdowns at this stage create application-to-issue delays that undermine the speed advantage of AUW.

When evaluating AUW platforms alongside a PAS, consider the following:

• API compatibility between the AUW platform and your PAS
• Data model alignment including field mapping between underwriting decision data and policy record structure
• Continuity of new business workflow ensuring that approved decisions flow into policy setup without manual intervention
• Exception handling including how declined or referred applications are managed in both systems

Insurers replacing their PAS while also building an AUW program should carefully sequence these initiatives. 

Where Sapiens Fits

For insurers building accelerated underwriting programs, the policy-side foundation carries as much weight as the decision engine. Sapiens provides the life insurance administration platform, including new business and underwriting workflow, that approved AUW decisions flow into. 

Its API architecture and new business intake workflow connect directly to leading AUW decision engines, so an approved decision becomes an issued policy without manual handoffs. For an end-to-end digital new business process, that connection is where speed is won or lost.

Build accelerated underwriting on a policy administration platform that is already live in production, and you remove the integration risk that stalls most programs. 

That is the foundation Sapiens brings: 40+ years of life insurance domain expertise, with more than 600 insurers across 30 countries running on Sapiens today.

The post Accelerated Underwriting Platforms for U.S. Life Insurers appeared first on Sapiens.

What Variable Annuity Administration Involves

Variable annuities are investment products wrapped in an insurance contract. Unlike fixed annuities, where the insurer assumes the investment risk, variable annuity values fluctuate based on market performance of underlying sub-accounts. This creates administrative requirements that significantly differ from those of other annuity types:

Separate account management

Variable annuity values are held in separate accounts, legally segregated from the insurers’ general account and invested in sub-accounts that function similarly to mutual funds. The platform must connect with fund administration systems to receive daily valuations and apply them to policyholder account values.

Daily NAV processing

Every business day, the platform processes NAV updates from fund companies or custodians, applies them to policyholder account values, and updates unit holdings. For large VA books, this creates significant data processing demands that the platform must handle accurately within the operational windows.

Sub-account management and transfers

Policyholders allocate premiums across sub-accounts and may transfer allocations at defined intervals. The platform must process these elections, track unit holdings by sub-account, and account for any cost basis implications of transfers.

Living benefit rider calculations

Most variable annuity products sold over the past 20 years include living benefit riders such as GMIB (Guaranteed Minimum Income Benefit), GMAB (Guaranteed Minimum Accumulation Benefit), GLWB (Guaranteed Lifetime Withdrawal Benefit), or combinations of these. These riders maintain separate calculation tracks including benefit bases, roll-up rates, and withdrawal histories that must be calculated in parallel with actual account values.

Fund company integration

The platform must support fund company data feeds including the receipt of NAV data, processing of fund substitutions, and management of share class changes in compliance with prospectus requirements.

FINRA and SEC compliance

Variable annuities are registered securities. The platform must support prospectus-based disclosures, FINRA suitability documentation, and SEC reporting obligations that do not apply to indexed annuities.

Cloud vs. On-Premise: The Real Operational Differences

The “cloud vs. on-premise” question for variable annuity administration is not simply a deployment preference. It has direct implications for processing architecture, upgrade cycles, integration, and total cost of ownership.

For variable annuity administration specifically, cloud deployment offers meaningful advantages in two areas: daily processing scalability (cloud infrastructure can scale for end-of-day NAV processing peaks without over-provisioned hardware) and integration with external data sources. A variable annuity’s money is invested in underlying funds run by asset managers, and the platform needs each fund’s daily prices to value policyholder accounts correctly. Those fund company APIs and market data feeds integrate more cleanly in a cloud-native environment.

The primary concern about cloud deployment is data residency and security, particularly for insurers with strict data governance requirements or those operating in jurisdictions with specific data localization rules. 

Core Capabilities of a Cloud VA Administration Platform

A cloud-based variable annuity administration platform must include: 

Daily valuation engine

Automated NAV ingestion from fund data providers, application to policyholder unit holdings, and account value recalculation, completed within a defined daily processing cycle with full reconciliation and exception management.

Sub-account and unit holding management

Accurate tracking of unit purchases, redemptions, and transfers. Calculation of cost basis for tax reporting. Fund substitution processing in compliance with SEC requirements.

Living benefit rider engine

Separate calculation tracks for each rider type. Benefit base maintenance, roll-up processing, withdrawal tracking, and step-up elections, all calculated alongside actual account values and updated with each transaction.

Systematic transactions processing

Dollar cost averaging, automatic portfolio rebalancing, systematic withdrawals, and required minimum distribution (RMD) processing, executed automatically based on policyholder elections.

FINRA/SEC compliance infrastructure

Support for prospectus-based disclosures, suitability documentation, FINRA reporting, and SEC annual filing requirements.

1099-R and tax reporting

Accurate tax reporting for annuity distributions, including correct treatment of qualified and non-qualified contracts, Roth conversions, and 1035 exchanges.

Policyholder self-service portal

Online access to account values, sub-account performance, transaction history, beneficiary management, and electronic statement delivery.

Integration Architecture for Cloud VA Platforms

Cloud variable annuity platforms operate within a range of external data sources and internal systems. Key integrations include:

Fund company data feeds

NAV data from fund companies or custodians (for example, through DTCC/NSCC infrastructure) is the most critical integration. The platform must handle feed failures, late NAVs, and fund substitution events without compromising data integrity.

Policy administration system (if separate)

Some insurers run a separate PAS alongside a specialized VA platform. Integration between the two systems must handle new business intake, policy servicing events, and commission processing without creating data duplication or gaps.

Custodian and clearing infrastructure

Separate account assets are held at custodians or cleared through NSCC. The platform must connect with this infrastructure to support trade settlement, asset reconciliation, and corporate action processing.

General ledger and financial reporting

Separate account and general account accounting must flow correctly to the insurer’s general ledger. Cloud platforms should provide clean data exports or direct integration with ERP and actuarial systems.

Reinsurance administration

Living benefit riders are often reinsured. Cession data including rider charges, benefit payments, and experience reports must flow accurately to the reinsurer’s systems.

Living Benefit Rider Administration in the Cloud

Living benefit riders deserve special attention in any VA platform evaluation because they are where most platform capabilities are tested.

Benefit base tracking

The benefit base (the amount used to calculate guaranteed income or withdrawal amounts) typically grows through a roll-up rate or step-up mechanism. It must be maintained separately from the actual account value and updated correctly through every transaction.

Step-up elections

Many GLWB riders allow the policyholder to “step up” the benefit base to the current account value on policy anniversaries if the account value has exceeded the benefit base. These elections must be tracked and applied correctly.

Withdrawal tracking

Guaranteed withdrawal amounts depend on cumulative withdrawal history. Excess withdrawals reduce the benefit base proportionally. The platform must track withdrawals accurately because errors compound over time.

Income phase administration

When policyholders begin taking guaranteed income, the platform must calculate the correct payment amount based on the benefit base, the withdrawal percentage for the policyholder’s age at election, and any applicable joint life adjustments.

Rider charge processing

Living benefit rider charges are typically deducted from account value as a percentage of the benefit base. These charges must be deducted correctly each quarter without affecting the benefit base calculation.

Regulatory and Compliance Considerations

Variable annuities sold in the U.S. require compliance capabilities beyond standard insurance administration platforms:

Prospectus and disclosure management

The platform must support delivery of current prospectus documents and Statements of Additional Information (SAIs) to policyholders in electronic or paper form.

FINRA Rule 2111 / Reg BI suitability documentation

Suitability documentation collected at the point of sale must be stored and accessible. The platform should support integration with distribution management systems where this data is captured.

SEC annual reporting

Separate account annual and semi-annual reports (Form N-4) require data from the administration platform in prescribed formats.

State regulatory variation

Annuity suitability regulations vary by state. The platform’s compliance configuration must accommodate state-level rule variations without requiring custom code for each jurisdiction.

What Questions to Ask Your Vendor

Every platform in this market calls itself “cloud.” They don’t all deliver the same thing. Some are legacy on-premise systems lifted into a hosted environment; others are built and run as true SaaS. That gap shows up in the work you do every day: end-of-day valuations, upgrades, and integrations. Before you shortlist any vendor, ask them these questions:

Cloud-native vs. Cloud-hosted

A cloud-native platform is built from the ground up for cloud architecture using microservices, scalable infrastructure, and API-first design principles. A cloud-hosted platform is typically a legacy application running on cloud infrastructure. The operational benefits of cloud are significantly greater with a true cloud-native architecture.

Daily processing SLA

What processing window does the vendor commit to for daily NAV processing? What service level commitments apply to processing completion time, and what recourse is available if those are missed?

Fund data integration network

How many fund companies and data providers are already connected to the platform? A broad set of existing integrations can reduce implementation effort and risk.

Rider capabilities

Test the rider engine against your most complex products. Ask for a live demonstration of step-up election processing, excess withdrawal calculations, and income phase initiation.

Regulatory update management

How does the vendor manage FINRA, SEC, and state regulatory updates? Is there a dedicated compliance team, and what is the service level agreement for implementing regulatory changes?

How to Migrate or Deploy a VA Platform 

For insurers migrating from on-premise:

• Profile your data and document your legacy system first
• Separate account reconciliation between the legacy system and new platform to reduce risk
• Validate benefit base data with actuarial signoff for in-force living benefit riders n.
• Plan regulatory reporting continuity across the migration date explicitly

For insurers building a new VA platform:

• Start fund company integration early. NSCC connectivity and fund company data agreements can take 6 to12 months
• Run living benefit rider configuration and actuarial testing in parallel with platform buildout
• Build FINRA/SEC compliance review into the testing program, rather than post-launch

Where Sapiens Fits

If you’re running variable annuities alongside fixed and indexed products, you know the cost of fragmentation: separate systems, duplicated integrations, and reconciliation work that grows with every product line. Sapiens administers your full annuity portfolio on one cloud-based SaaS platform, so your VA book and your FIA book share the same architecture.

Our platform scales, upgrades continuously, and carries none of the infrastructure burden of on-premise systems. One platform gives you one set of fund integrations, compliance configuration, and data model feeding your general ledger and actuarial systems. Our North American support covers U.S. and Canadian regulatory requirements.

We bring 40+ years of life & annuities domain expertise, with more than 600 insurers across 30 countries running in production today.

The post Variable Annuity Administration in the Cloud: Buyer Guide for 2026 appeared first on Sapiens.

Why Legacy Replacement Is Uniquely Risky in Life Insurance

Life insurance is unlike most other industries. Policy contracts can span 20, 30, or 40 years. In-force data is complex, often inconsistently structured across decades of product launches, platform changes and system patches, and highly sensitive to calculation accuracy.  Billing errors may be fixable, but errors in a benefit calculation or a lapse notice can trigger regulatory action and legal liability.

This is why replacing a legacy system in life insurance carries risk that other sectors do not face at the same scale:

• Calculation parity risk: The new system must produce outputs identical  to the legacy system for all in-force policies. Even minor rounding differences in annuity crediting or cash value calculations must be identified and resolved before cutover 
• Data completeness risk: Decades of policy data may be incomplete, inconsistent, or stored in formats that no longer have active documentation. Missing fields must be reconstructed, defaulted, or flagged 
• Regulatory risk: Mid-migration periods often involve parallel systems, creating compliance complexity around which system of record governs 
• Customer experience risk: Any billing interruption, delayed claim, or incorrect statement during migration can lead to complaints from policyholders and potential regulatory scrutiny 

Understanding these risks before the project begins is the first step toward  managing them.

The Five Phases of a Life Insurance Platform Migration

Phase 1: Discovery and Data Profiling

Before you configure any new system,  you need a  thorough understanding of the existing data. During data profiling, you typically:

• Extract all in-force policy records and map fields to a canonical data model
• Identify data quality issues such as nulls, outliers, inconsistent product codes, and legacy flags with unclear meaning
• Document the calculation logic in the legacy system, including edge cases and workarounds
• Build a complete product inventory covering every product, rider, and endorsement currently in force

This phase is consistently underinvested. Projects that cut corners here pay for it later in validation. 

Phase 2: Platform Configuration and Product Setup

Once data is profiled, you configure the new platform to support all products and riders identified in Phase 1. This means you:

• Build product structures in the new system’s product factory
• Configure calculation rules, crediting logic, and benefit parameters
• Set up billing cycles, lapse logic, grace periods, and reinstatement rules
• Establish regulatory outputs including reserves, filings, tax reporting

Do the Configuration in close collaboration between your business stakeholders and implementation team. Run actuarial validation of calculation logic in parallel.

Phase 3: Data Conversion and Validation

This is the highest-risk phase. To convert your data, you:

• Transform legacy data into the new system’s data format
• Load converted policies into a test environment
• Run parallel calculations, generating the same outputs from both old and new systems and reconciling differences
• Resolve discrepancies at the record level before advancing to cutover

Expect  to run multiple conversion cycles, each revealing new edge cases that require resolution.  Budget for this, because a single conversion cycle is almost never enough.

Phase 4: Parallel Run and User Acceptance Testing (UAT)

Before cutover, validate the new system against live operations through a parallel run period where: 

• Both systems process the same transactions simultaneously
• Outputs are compared and reconciled
• Business users validate workflows, statements, and servicing processes
• Agent and distributor portal access is tested under realistic conditions
• Compliance teams review regulatory outputs

The parallel run period typically runs for one to three full billing cycles. Shorter parallel runs introduce cutover risk.

Phase 5: Phased Cutover and Stabilization

Few insurers attempt a single “big bang” cutover of all in-force policies. Phased migration by product line, issue date range, or distribution channel reduces risk by limiting the scope of each cutover event.

Post-cutover stabilization involves:

• Hypercare support during the first 60 to 90 days including elevated support staffing and rapid issue resolution
• Daily monitoring of billing runs, claims outputs, and system performance
• Review of policyholder communication and statement accuracy  
• Monitoring of agent and distributor feedback to identify servicing gaps early

Managing In-Force Policy Data Migration

The in-force block is the most sensitive part of any migration. Key practices include:

Document legacy business rules before you begin configuration

Legacy systems often contain undocumented logic, essentially workarounds built for specific products or edge cases that exist only in the code. Your business analysts must work with legacy system developers, or with available documentation if developers are gone, to identify and document this logic before migration begins.

Build a conversion specification document

Every legacy field should map to a  corresponding field in the new system, and all transformation rules should be documented. This specification becomes the governing document for the technical conversion team.

Use automated reconciliation tooling

Manual reconciliation of large in-force blocks is prone to error. Automated comparison tools are essential for identifying record-level discrepancies between legacy and new system outputs, particularly in migrations involving more than a few thousand policies.

Establish a data freeze protocol

Define a clear data freeze date, after which changes in the legacy system must be tracked and manually applied to the new system to maintain synchronization during conversion.

Protecting Customer Experience During a Migration

The migration should feel invisible to your policyholders. If they notice it, something has gone wrong. To protect the positive customer experience, make sure that you:

Maintain uninterrupted billing

Billing accuracy is the clearest sign of platform stability to policyholders. Validate the billing logic thoroughly before cutover, and keep legacy system processing available  as a fallback for the first billing cycle  after cutover.

Preserve statement continuity

Review policy statements, annual notices, and correspondence for format and content accuracy–after migration. Changes in statement layout or wording can confuse policyholders and drive up contact center calls.

Communicate proactively with agents and distributors

Agents are the first point of contact when policyholders have questions. Brief them on changes to their portal access and servicing workflows before cutover, not after.

Plan for a rise in contact volume

Even well-executed migrations generate a temporary increase in policyholder and agent inquiries. Adjust contact center staffing accordingly for the first 30 to60 days  after cutover.

Common Pitfalls and How to Avoid Them

Underestimating data quality

Almost every insurer believes its data is cleaner than it actually is. Build a 30 to50% buffer into your data remediation timelines based on findings from the data profiling phase.

Compressing the parallel run

Pressure from management to hit go-live dates often results in shortened parallel runs. This is one of the most reliable predictors of incidents after cutover.  Protect the parallel run timeline.

Treating migration as a pure IT project

Migrations managed solely by IT fail at the business requirements level., Your actuarial, operations, compliance and distribution teams must be active participants from the start. 

Selecting a partner without migration experience

A platform may be powerful, but without a proven track record of in-force block migrations, the risk climbs sharply. Ask specifically for client references on migrations of comparable complexity, not just greenfield implementations.

Skipping stabilization planning

Hypercare is not optional. Post-cutover incidents require a rapid, coordinated response. Without a clearly defined stabilization protocol, that response becomes– reactive and slow.

What to Ask About Migration Support

When you evaluate PAS partners, these are the migration questions that matter:

• How many in-force block migrations have you completed for U.S. life insurers in the past five years?
• What is your data conversion methodology, and do you provide automated reconciliation tooling?
• How do you handle undocumented legacy business rules discovered during data profiling?
• What does your parallel run process look like, and what is the minimum parallel run period you recommend?
• Do you provide hypercare support post-cutover, and for how long?
• Can you provide references from insurers who have migrated a comparable in-force block?

Where This Leaves You 

You are not just choosing a platform. You are choosing a foundation that either makes intelligence possible across your business or keeps it stuck between the board deck and operations. The two risks that decide the outcome are the same two every migration program relies on: choosing the right platform fit, and executing the transformation without disrupting the people who depend on you.

Sapiens delivers life and annuity platform migrations for insurers across North America, including programs involving significant in-force blocks. The Sapiens life platform includes conversion tooling and a structured migration methodology built around the data profiling, calculation parity, and parallel run challenges described in this guide. Behind it sits 40+ years of insurance ontology and more than 600 insurers across 30 countries running on Sapiens. A modern platform, cloud deployment, and a proven migration track record are what reduce the two risks that matter most.

The post How to Replace a Legacy Life Insurance Admin Platform Without Customer Disruption appeared first on Sapiens.

The MGA market grew 16% in 2024 to $114 billion in US premiums, with fronting companies supporting more than 20% of it. But the data flowing in from each MGA decides how much of that growth reaches your bottom line. That’s the ground fronting teams lose without ever knowing it and we work with you to win back.

What Poor MGA Data Quality Really Costs

The cost hides in three places, most teams never call data problems.

The recoverable you never find. Your reinsurance system matches gross claim data to treaty terms, so one wrong field, a misread date of loss or catastrophe code, can attach a claim to the wrong treaty or none at all. The money’s owed, but the process to find it never runs.

The reconciliation tax. Every MGA sends bordereaux its own way, and your team cleans, reformats, and reconciles each one every quarter, work that buys no strategic value.

The treaty pricing premium. In a discussion with They now penalise poor or untimely data with higher pricing – a global trend that already reaches Lloyd’s coverholders. If your reinsurance costs have climbed while your loss ratios held steady, your data is the likely reason.

The Problem Compounds as You Grow

Every new MGA adds another format, another reconciliation cycle, and another source of error, and one bad field can distort recoveries across the whole portfolio. The market’s consolidating too: , forcing you to manage MGAs at very different stages of data maturity. Your reinsurers won’t wait for it to settle, and neither should you.

Fixing MGA Data Quality at the Source

A stronger reinsurance system alone can’t fix data that arrives broken, so the fix spans two layers. The upstream work, standardizing and validating your MGA data, applying governance and exception handling, and turning the unstructured documents alongside bordereaux into structured inputs, is typically led by a systems integrator, and we partner closely with them so nothing reaches your treaty calculations unchecked.

Downstream, in the system of record, we work directly with you: we validate the fields most often behind shortfalls, control every change with a full audit trail, assign treaties automatically, and identify recoverables across all open treaties continuously, where manual review only works periodically. Together, that’s how data quality becomes built in, not corrected after the fact.

A stronger reinsurance system alone can’t fix data that arrives broken, so we work both ends with you. On the way in, we standardise and validate your MGA data. We apply governance and exception handling before it reaches treaty calculations and turn unstructured documents alongside bordereaux into structured inputs. In the system of record, we validate the fields most often behind shortfalls and control every change with a full audit trail. We assign treaties automatically and identify recoverables across all open treaties continuously, where manual review only works periodically. That’s how data quality becomes built in, not corrected after the fact.

See What Poor Data Is Costing You

Most fronting teams underestimate the gap, and it rarely shows until the system is doing the work. The numbers are rarely small. At three MGA relationships running 70% manual processing, a fronting company with $50 million in MGA-sourced premium can face roughly $2.5 million a year in recoverable leakage and reconciliation cost, and it climbs with every MGA added.

The recoveries are real, documented in our implementations. Automated reviews have found $1.9 million in aggregate claims invisible to manual processes at one insurer, $1.3 million at a personal lines mutual, more than $10 million in unidentified facultative recoveries at a large multi-line insurer, and over $3 million in old losses at a workers’ compensation carrier. The money was owed but never collected, because the process to find it didn’t exist, and nearly every organization that automates recoverable identification finds amounts it had written off.

We won’t hand you a leakage estimate before we start, because that number isn’t real until the system is live. What we can tell you from experience is that once the platform is deployed and recoverable identification runs across your book, we almost always find claims leakage opportunity, amounts that manual processes had quietly written off. Insurers that automate processing and validation have reported up to 30% gains in operational efficiency, and those that act reach payback within the first year, through recovered dollars, lower reconciliation cost, and a stronger renewal position.

Experience That Closes the Gap, Together

This isn’t theoretical. More than 600 insurers across 30 countries run on Sapiens, and our reinsurance clients stay with us, a 10-plus year average tenure and a 95%+ retention rate. We’re recognised in the Datos Insights Reinsurance Administration Solutions Market Navigator and named a Celent “Luminary” in ceded reinsurance systems. With our implementation and data partners, we’ve done this at scale. When your recoveries grow and your renewals get easier, that’s the outcome we’re working toward, side by side with your team.

The post MGA Data Quality: The Reinsurance Premium You Pay Without Knowing appeared first on Sapiens.

The Proposition Agility Index

The Proposition Agility Index maps five stages of proposition maturity across seven capability dimensions and can be explored in more detail in our whitepaper, The Life & Pensions Proposition that Delivers Its Promise.

The following three scenarios describe insurers at different points on the Proposition Agility Index. Most senior leaders will recognise elements of at least one scenario, and many will recognise all three.

Scenario A: The Slow Updater

A CI product was market-leading when it launched three years ago, covering 35 conditions at a competitive premium. Today, the top three competitors cover 60 to 80 conditions at a comparable price point. This gap was flagged by your product team 18 months ago. The update is still working through actuarial repricing, regulatory review, system configuration, and distribution rollout. By the time it reaches the market, the benchmark will have moved again.

Meanwhile, your adviser panel has noticed. New business volumes are down 12% year-on-year, and your retention team is fielding more cancellation enquiries than at any point in the last five years.

The root cause is not the product team who identified the problem early. The root cause is an operating model where changing a product definition requires a chain of sequential dependencies across actuarial, compliance, IT configuration, testing, and distribution, each operating on its own timeline. The total cycle time from concept to customer can exceed 18 months. In a market that moves quarterly, that is a structural disadvantage.

The Slow Updater typically sits at Stages 1–2 on the Proposition Agility Index (Static or Modular): product changes are project-led, configuration is monolithic, and distribution updates are manual.

Scenario B: The Compliance Firefighter

The product development team spends more time responding to regulatory change than evolving the proposition. IFRS 17 reporting consumed most actuarial and IT bandwidth for the past two years. The FCA’s Consumer Duty requirements forced a product value review across your entire book. Each change is handled as a bespoke project, with its own timeline, budget and system workarounds. The cumulative effect is that the product roadmap is 18 months behind plan, and cost-to-serve has risen 20% in three years while your proposition has barely changed.

The deeper issue here is not regulatory burden per se. Every insurer faces the same requirements. The issue is that the operating model treats each regulatory change as a bespoke exercise rather than absorbing it through configuration. The capacity consumed by compliance is capacity that cannot be directed at proposition evolution. The two compete for the same resources, and compliance always wins because it has a deadline.

The Compliance Firefighter has often invested significantly in modernisation but remains at Stage 2 (Modular) on the Proposition Agility Index: change velocity is quarterly at best, and regulatory response is still project-driven rather than configuration-led.

Scenario C: The Closed Book Drag

Four legacy policy admin systems support products that are no longer sold, but still carrying more than 200,000 policyholders, consuming 35% of the IT run budget. Every regulatory change must be applied four times, on four different platforms, with four different testing cycles.

Closed book customers receive a visibly lower standard of service: no digital self-service, slower response times, limited ability to adjust their cover. When advisers review these clients, they recommend switching. You are losing customers you have already paid to acquire, and your operations team is stretched so thin maintaining legacy that they have no capacity for transformation.

What makes this scenario particularly costly is the hidden opportunity cost. Those 200,000+ policyholders represent accumulated trust, established premium flows, and the lowest-cost growth opportunity in your portfolio. Those policyholders on a modern platform would become cross-sell candidates, retention assets, and advocates. Without it, they are a drain on resources and avoidable attrition.

The Closed Book Drag often coexists with pockets of modernity elsewhere in the business. The insurer may be at Stage 3 (Personalised) for new business but remains at Stage 1 (Static) for closed books  creating an internal agility gap that compounds over time.

Which scenario dominates your operating reality?

Most L&P insurers will recognise elements of all three scenarios. Which one is your primary constraint right now? This determines where your performance gap is most acute, where the protection gap for your customers is growing fastest, and where structural investment will deliver the greatest return.

These scenarios sit within a broader Proposition Agility Index strategic framework which maps five stages of proposition maturity – from static and product-led through to anticipatory and customer-led – across seven capability dimensions.

Learn where you sit on the Index, and what it takes to progress, in our whitepaper, The Life & Pensions Proposition that Delivers Its Promise.

Download it now to identify your path forward.

The post Three Scenarios That Reveal Where Your Life & Pensions Performance Gap Is Most Acute appeared first on Sapiens.

Digital transformation in life insurance isn’t a single initiative. It’s a coordinated modernization across six operational pillars, executed over a multi-year period. Insurers that treat it as a technology tend to fall short. Those that treat it as a business strategy enabled by technology are far more likely to succeed. This roadmap explores the six pillars that matter, how to sequence them, where U.S. and Canadian insurers diverge in their challenges, and what separates transformation programs that deliver from those that stall.

Why Digital Transformation in Life Insurance Is Different

Life insurance is a conservative industry, and for good reason. Policies span decades. Mistakes in calculation or administration have legal and regulatory consequences. The regulatory environment is fragmented: in the U.S., across 50 state regulators; in Canada, across federal and provincial frameworks. Legacy systems remain deeply embedded because they represent decades of operational logic that’s difficult and expensive to reconstruct.

This creates a transformation environment unlike retail, banking, or even property and casualty insurance:

• The cost of disruption is extremely high. A billing failure or lapse error isn’t just an operational inconvenience but a regulatory event and a liability
• The technology estate is complex. Core systems may be 30–40 years old, with layers of customization, integration, and workarounds built up over decades
• The pace of change is inherently constrained. State filing requirements, actuarial signoffs, and compliance reviews mean new products and processes take longer to deploy than in other industries
• The customer relationship is long-term and trust dependent. Policyholders aren’t transactional customers. A poor experience during a transformation can damage relationships built over decades.

Understanding these constraints is the first step toward building a realistic, achievable transformation roadmap.

The Six Pillars of Life Insurance Digital Transformation

Digital transformation in life insurance spans six operational pillars. Each has its own timeline, dependencies, and risk profile. Established insurers are typically working across multiple pillars simultaneously, but with clear sequencing and prioritization.

Pillar 1: Core System Modernization

The policy administration system is the foundation of all other transformation. Modernizing the PAS, whether through replacement, migration to cloud, or API enablement of a legacy system, unlocks speed in every other pillar. Insurers that attempt to build digital capabilities on top of an unmodernized core consistently face limitations.

Pillar 2: Digital Distribution and New Business

Application intake, agent portals, consumer-facing quote and apply platforms, and straight-through processing (STP) from application to policy issue. This is often the highest-visibility pillar because it affects sales volume and distributor satisfaction directly.

Pillar 3: Underwriting Automation

Accelerated underwriting, automated rules engines, third-party data integration, and referral workflow optimization. The goal is faster decision-making, lower unit cost per application processed, and more consistent risk selection.

Pillar 4: Customer Experience and Self-Service

Policyholder portals, digital statements, beneficiary management, loan requests, and surrender processing without agent intermediation. This pillar improves retention and reduces servicing cost simultaneously.

Pillar 5: Data and Analytics

Building a modern data infrastructure including data warehouse, real-time reporting, experience analysis, and predictive modeling. Without modern data capabilities, every other pillar is operating partially blind.

Pillar 6: Finance, Actuarial, and Regulatory Technology

Modernizing finance and actuarial processes: reserving systems, financial close, regulatory reporting automation, and actuarial modeling environments. This pillar is often underinvested relative to its risk and cost implications.

A Phased Transformation Roadmap

Not all domains can be addressed simultaneously without risking program overload. The following sequencing reflects a practical, risk-managed approach for a mid-sized life insurer:

Phase 1: Foundation

• Assess and document the current technology landscape
• Define target architecture and platform strategy
• Launch core system modernization (PAS selection or migration initiation)
• Implement API layer to enable integration while legacy system is still in place
• Begin data foundation work: data catalog, governance framework, initial data warehouse

Phase 2: Distribution and New Business

• Deploy modern e-application and agent portal capabilities
• Implement accelerated underwriting for eligible products
• Build straight-through processing from application to policy issue for simple products
• Improve new business cycle time measurement and management

Phase 3: Customer Experience

• Launch policyholder self-service portal
• Implement digital statement delivery and e-consent
• Automate routine servicing transactions (loans, beneficiary changes, address updates)
• Build omnichannel service capability (phone, digital, email, chat)

Phase 4: Data and Analytics

• Build real-time reporting on new business, lapse, and claims
• Implement experience analysis capabilities
• Develop predictive models for lapse propensity and customer lifetime value
• Build executive and operational dashboards

Phase 5: Finance and Actuarial

• Modernize reserving and financial close processes
• Implement automated regulatory reporting
• Upgrade actuarial modeling environment

U.S. vs. Canadian Insurers: Where the Journeys Differ

U.S. and Canadian life insurers share many transformation challenges but there are differences between the two markets:

Regulatory fragmentation

U.S. insurers face regulatory variation across the 50 states in product approval, data use, and compliance requirements. This makes regulatory technology such as automated state filing, compliance tracking, and adverse action management a more complex investment in the U.S. than in Canada’s federal-provincial regulatory environment.

Product mix

Canadian insurers have historically maintained larger participating whole life portfolios relative to the U.S. market’s stronger focus on term, IUL, and annuities. This affects both core system requirements and digital distribution strategy.

Distribution model

The Canadian market has a higher concentration of captive agent and bank distribution. U.S. insurers face more complexity in managing independent agent relationships, IMO/BGAs, and broker-dealer distribution, each requiring different portal and compensation management capabilities.

Market scale

The U.S. market is roughly 10 times the size of Canada’s. This affects the depth of the vendor market: more vendors target U.S. insurers, and some SaaS platforms have stronger U.S. regulatory support than Canadian ones. Insurers operating in both markets should evaluate platform capabilities across both regulatory environments.

Common Transformation Failures and How to Avoid Them

Starting with the shiny, not the foundation

Many insurers launch consumer-facing digital initiatives before modernizing core systems. The result is a digital front end that can’t deliver on its promises because the underlying systems can’t support it. Core system readiness isn’t optional.

Underestimating program governance requirements

Digital transformation isn’t an IT program. It requires business sponsorship, executive alignment, cross-functional governance, and active change management. Programs governed solely by IT consistently fail to achieve adoption.

Trying to do everything simultaneously

Transformation fatigue is real. Organizations can only absorb a limited amount of change. A roadmap that launches five major initiatives in parallel creates execution failure across all of them.

Treating vendor selection as a one-time decision

The vendor landscape evolves. A platform selected for digital transformation in Year 1 must be evaluated annually to ensure that it remains the right fit. Build vendor review checkpoints into the governance process.

Ignoring the people dimension

Technology is the easiest part of transformation. Changing how underwriters, actuaries, and service teams work is harder. Investment in change management consistently correlates with successful outcomes.

Measuring only technology delivery

Successful transformation is measured in business outcomes: new business cycle time, STP rate, lapse rate, per-policy operating cost, and Net Promoter Score. Programs that measure only technology delivery lose sight of the real objective.

How to Build and Govern a Transformation Program

Establish a Transformation Management Office (TMO)

A dedicated governance function that isn’t the IT PMO is needed for programs that span multiple domains over several years. The TMO manages interdependencies, resource allocation, and program-level risk.

Define the Target Architecture

Before launching individual initiatives, define what the technology environment should look like when transformation is complete. This prevents initiatives from moving in conflicting directions.

Create a Benefits Tracking Framework

Map each initiative to measurable business benefits and track progress regularly. This keeps transformation efforts focused on results rather than project delivery milestones.

Build in Learning Cycles

Phase gates between transformation stages allow the organization to learn from early efforts before committing to later ones. Rigid multi-year plans that ignore feedback systematically fail.

Where Sapiens Fits

Sapiens serves as a core platform partner for life and annuity carriers pursuing digital transformation in North America. Our solutions support policy administration, new business processing, billing, claims, and distribution management covering multiple operational areas within a single vendor relationship.

For insurers at the beginning of their transformation program, Sapiens provides the PAS foundation needed to support digital distribution, underwriting automation, and customer experience. Our cloud-native deployment model reduces burden on infrastructure, and our API architecture supports integration with third-party solutions across the technology landscape.

For insurers further along in their transformation journey, Sapiens can be evaluated as a targeted replacement for specific system components not necessarily requiring a full-stack commitment.

Sapiens’ North American-specific regulatory capabilities, including support for U.S. state filing requirements and Canadian provincial compliance, are relevant for carriers operating in both markets.

The post Life Insurance Digital Transformation Roadmap for U.S. and Canadian Insurers appeared first on Sapiens.

This series has been an invitation into conversations we’ve been having with industry leaders: what the future looks like, what works, what doesn’t, and what it takes to build the next 30 years of our industry.

Whether you’re an insurance company or a firm that serves them, the time is now. The change that disrupted retail, finance, and entertainment has already happened. Now it’s our turn.

The vision is to build a platform that autonomously pursues business outcomes. This is the distinction that matters: autonomous agentic AI is the goal, not an endless stream of automated processes that only accelerate aspects of your business without transforming the whole.

Step 1: Redefine How You Measure Technology Value

The most immediate shift is in the questions you ask of your technology investments, and it’s one most insurers can make right now. Stop evaluating platforms on feature counts and start holding them accountable for the business metrics that matter to your organisation. For claims, that means measuring settlement speed, processing cost per claim, fraud detection accuracy, and customer satisfaction scores, not system availability or transaction throughput. For underwriting, it means loss ratio improvement, risk selection accuracy, and portfolio profitability, not the number of rating factors supported or API connections available. Every core platform in your technology stack should have a defined set of business outcomes it is expected to move, and structure your vendor relationships to show progress against those outcomes. The shift requires insurers to build stronger feedback loops between their technology teams and their business operations. The data that measures loss ratios and claims cycle time needs to flow back into the evaluation of the platforms driving those results.

Step 2: Prepare Your Data Foundation

Outcome-as-a-Service is only possible if AI has the data it needs to optimise intelligently. Before an insurer can benefit from outcome-driven platforms, it must address the data fragmentation that still plagues most insurance organisations. Siloed policy, claims, billing, and customer data prevents AI from seeing the full operational picture it needs to make meaningful decisions. The practical implication is that data unification is not an IT project, it is a strategic prerequisite for competing in the next era of insurance. Insurers should prioritise building unified data environments that give their platforms continuous access to clean, connected operational data across the business. Without this foundation, AI-driven optimisation remains theoretical.

Step 3: Shift Your Vendor Strategy Towards Platform Thinking

The autonomous insurance platform works at the centre of a network where specialised partners contribute fraud detection, telematics, parametric triggers, embedded insurance distribution, and capabilities the platform orchestrates dynamically. Evaluating vendors in isolation no longer works. We recommend platforms with open architectures, strong APIs, and true insurance ontology the platform can train on. For instance, at Sapiens, we have 40+ years of insurance domain logic, with more than 2,000 workflows modelled, unified across P&C, life and reinsurance.

Step 4: Train Your Team to be the Training Signal

The autonomous platform is impossible without human beings. AI capability and human judgement are inseparable. Each makes the other stronger. As the platform takes on more of the execution, your people’s role shifts. They move from processing decisions to shaping them: setting objectives, applying expertise, and making sure the platform keeps learning. That means training your teams not just to use the platform, but to continuously direct it so it gets better at delivering the outcomes that matter to your business.

Step 5: Act Today

Technology alone will not complete this transition. Insurers must prepare their organisations to operate in an outcome-driven model. That means underwriting, claims, and operations leaders need to understand what an autonomous agentic AI platform is optimising for and actively participate in setting those objectives. It also means actuaries and data scientists need to work alongside platform teams to define the outcome metrics that matter. And it means executives need to hold both their internal teams and their technology partners accountable to business performance, not just implementation milestones. Those who continue to measure technology by features and to absorb all the risk of translating capability into performance will find the gap increasingly difficult to close. We’re building compounding advantages in operational efficiency, risk intelligence, and customer experience to work for your successful business outcomes.

Read the other blogs in the series:

5 Insurtech Capabilities the C-Suite is Discussing: Insider Perpective

How a Platform Approach Drives Autonomous Insurance Outcomes

Outcome-Seeking AI: Insurance’s Next Platform Evolution

The Case for Building Insurance Platforms That Actually Understand Insurance

When AI Begins Running the Insurance Workflow

The post Getting Started with Outcome-as-a-Service appeared first on Sapiens.

I’ve spent a long time working with IBM Z, long enough to have a healthy sense of what’s genuinely new versus what’s incremental. IBM z17 feels genuinely new.

Not because of any single capability, but because of what the combination means: AI inference, built directly into the processor, running at mainframe speed and scale, on the most trusted enterprise platform in the industry. That changes what’s possible for you every day. 

As you advance AI adoption strategies, the need for a trusted, secure, and high-performance platform has never been greater. IBM z17 is purpose-built to unlock the full value of AI while maintaining the reliability, security, and scalability essential for mission-critical operations. 

Accelerate Growth with AI at the Core 

IBM z17 helps you bring AI closer to mission-critical data and business processes, enabling: 

• Faster business outcomes and fuel business growth 

• Innovate with new agentic AI services and speed  

• Modernize faster to free up development resources 

By integrating AI directly into the core of the enterprise, you can move from experimentation to execution, transforming insights into measurable business value. 

From where I sit, the most significant shift is the move from AI as a separate layer to AI as infrastructure. When inference runs on-chip at transaction speed, the architecture question changes entirely. You stop asking “how do we connect AI to our systems?” and start asking “what do we do with AI that’s already there?” 

Transform Operations with End-to-End Automation 

Operational efficiency remains a top priority if you’re seeking to maximize performance and agility. IBM z17 helps you: 

• Slash operational complexity 

• Expedite insights and eliminate blind spots 

• Get the most out of your investment 

These capabilities enable you to improve productivity, increase responsiveness, and create a more resilient operational foundation for the future. 

Automation on IBM Z has always been powerful. IBM z17 takes it further by making it possible to automate decisions that previously required human review, not because the data wasn’t there, but because the processing speed wasn’t. That fundamentally changes the economics of back-office operations. 

Protect Critical Data with Uncompromising Security 

As cyber threats continue to evolve, organizations require a platform designed to safeguard sensitive data and maintain trust. IBM z17 helps you: 

• Strengthen cyber resilience 

• Increase transparency to reduce risk 

• Simplify security and compliance operations 

By embedding security throughout the platform, IBM z17 helps you protect critical workloads while supporting regulatory and business requirements. 

Security on a mainframe has always been a strength. What IBM z17 adds is transparency with the ability to demonstrate compliance, trace decisions, and surface risk in real time. For the customers Sapiens works with, many operating under strict regulation, and other demanding frameworks, that auditability is as important as the protection itself. 

How Sapiens Helps Clients Innovate on IBM Z 

At Sapiens, our relationship with IBM Z runs deep and it’s an active one. We don’t just support IBM Z; we build on it, continuously, alongside our customers. 

Our eMerge platform is designed around one core idea: give development teams the speed and flexibility of modern tools, without abandoning the reliability and security of the mainframe. In practice, that means you can deliver new applications and regulatory frameworks faster than they ever could with traditional approaches. 

A recent example illustrates this well. A government agency needed to implement a new regulatory framework quickly with a hard legislative deadline. The requirements were complex: registration and licensing workflows, integration with legacy systems and external data sources, digital citizen access, and full compliance with regulatory rules. Using eMerge on IBM Z, the team delivered the system ahead of schedule. The key factors: 

• Rapid development: speed meant the legislative deadline wasn’t a constraint; the team had time to spare  

• Embedded solution modeling: streamlining registration and licensing without building everything from scratch  

• Automated technical complexity: the development team stayed focused on business logic, not infrastructure 

• Integration without disruption: connecting to legacy systems and external data sources without rebuilding them 

• Platform independence: creating and deploying across environments efficiently, with IBM Z as the reliable core 

That’s what IBM Z and eMerge together make possible. IBM z17 takes those capabilities and makes them faster, more efficient, and AI-ready. 

“We don’t ask you to choose between innovation speed and operational stability. eMerge on IBM Z gives you both.”

Integrating IBM z17 Capabilities into What We Build 

We’re actively working with our customers to integrate, migrate, and implement solutions that take full advantage of what IBM z17 brings. That work is happening now, not on a future roadmap. 

The focus is on making IBM z17’s AI inferencing capabilities available to eMerge-based applications without requiring our customers to manage the integration themselves. When you run on eMerge with IBM z17, the platform does that work so your teams can focus on what the AI actually does for your business. 

How AI Is Reshaping What Our Customers Need 

Sapiens recently unveiled a new direction for how we deliver AI-powered solutions to insurers: built around the concept of hyper-relevance: the ability to put the right decision, for the right customer, in the right moment, every time. 

That vision requires infrastructure that can act at the speed of a decision. It requires data that’s live, not stale. And it requires AI that’s embedded in the transaction, not sitting alongside it waiting to be consulted. 

IBM z17 is the infrastructure answer to that challenge. The Telum II processor’s integrated AI accelerators mean inference happens where the data lives – on the mainframe, inside the transaction – rather than in a separate AI layer with its own latency, its own data movement costs, and its own failure points. 

Our customers are already asking how to get AI into their core operations without compromising the reliability they depend on. IBM z17 and eMerge together give us a clear answer. 

What Sapiens Brings to IBM Z Customers 

There are a few things we’ve built over decades of IBM Z work that are genuinely hard to replicate: 

• Deep platform knowledge: we know how IBM Z behaves under load, in production, with real regulatory constraints. That knowledge is embedded in how we design eMerge. 

• Continuous investment: we track IBM Z releases and update eMerge accordingly. Our customers don’t need to manage the gap between IBM Z capability and application readiness. 

• A proven delivery approach: from rapid implementation to full modernization, we’ve done this across government, insurance, and financial services in more than 30 countries. 

Why IBM z17 Matters to Our Shared Customers 

IBM z17 is the platform that makes AI-native enterprise operations practical at scale. If you’re already running on IBM Z, it means the infrastructure you trust becomes meaningfully more capable without disruption, without migration, without risk. 

Together with IBM Z, Sapiens remains focused on helping you fully capitalize on the AI era while preserving the reliability, scalability, and trust that define the mainframe. We’re committed to helping you turn today’s investments into a sustained competitive advantage, driving long-term value in an increasingly digital world. 

“The organizations that will lead in the AI era aren’t the ones that bolt AI onto existing systems. They’re the ones that build it into their core. IBM z17 and eMerge make that possible today.”  

Performance Disclaimers 

* ** 5.6x fewer cores / 450 billion inference operations: Performance results are based on IBM® internal tests running on IBM Systems Hardware of machine type 9175. The MegaCard benchmark was containerized and deployed on Red Hat® OpenShift® Container Platform (RHOCP) 4.17 on Red Hat Enterprise Linux® (RHEL) 9.4 with KVM. Compared x86 Server Hardware configuration: 28 Sapphire Rapids Intel® Xeon® Gold CPUs @ 2.20 GHz with Hyper-Threading turned on, 1 TB memory. The 450 billion inference operations benchmark was executed with 1 thread performing local inference operations using a LSTM based synthetic credit card fraud detection model to exploit the Integrated Accelerator for AI. A batch size of 160 was used. IBM Systems Hardware configuration: 1 LPAR running Red Hat® Enterprise Linux® 9.4 with 6 IFLs (SMT), 128 GB memory. 1 LPAR with 2 CPs, 4 zIIPs and 256 GB memory running IBM z/OS® 3.1 with IBM z/OS Container Extensions (zCX) feature. Results may vary. 

The post Advancing Mainframe Momentum in the AI Era with IBM z17™ appeared first on Sapiens.

Together, with the infrastructure of eMerge, you can use the AI of IBM 17 to process documents and data.

The Pressure Insurers Are Under

The customers we work with aren’t struggling with a shortage of data. They’re struggling with latency as the lag between what the data knows and when a decision gets made.

At the same time, regulatory scrutiny is tightening. Fraud is getting faster and more sophisticated. And customers expect responses that feel instant and personal. The margin for delay has shrunk to near zero.

Most platforms weren’t designed for this. They were designed for batch processing, overnight runs, and T+1 reporting. That world is gone.

What IBM z17 Changes

IBM z17 is purpose-built to bring AI into the core of mission-critical workloads not as a bolt-on, but as infrastructure. Three capabilities matter most to insurance:

Real-time AI inferencing at transaction speed

IBM z17 processes up to 450 billion inference operations per day with 1 millisecond response time using a credit card fraud detection deep learning model.* That means AI decisions happen inside the transaction, not after it.

AI efficiency that changes the infrastructure equation

Running an AI-infused OLTP workload on IBM z17 with OpenShift Container Platform requires up to 5.6x fewer cores than a comparable x86 platform.** If you’re managing cost pressure alongside growth, that’s a meaningful shift.

Security and compliance built in, not bolted on

IBM z17 embeds security throughout the platform strengthening cyber resilience, increasing transparency, and simplifying compliance operations. That matters if you’re operating under Solvency II, DORA, and a growing patchwork of data regulations.

Where eMerge Fits

eMerge is Sapiens’ development platform for building and running insurance applications on IBM Z. It gives insurers’ development teams a modern, agile environment without walking away from the reliability, security, and performance that mainframe infrastructure delivers.

With IBM z17, eMerge-powered applications can use the Telum II processor’s integrated AI accelerators. That means:

• Fraud models run inside the payment transaction, not a downstream batch
• Underwriting rules execute against live data in real time
• Claims triage happens in real time flagging complexity before it becomes cost
• Agentic AI workflows can trigger and resolve autonomously cutting the manual steps that slow insurers down

The result isn’t just faster processing. It’s a fundamentally different ability to serve customers, manage risk, and respond to market conditions.

Helping Clients Innovate and Get There Faster

Working closely with our IBM Z customers, we continuously evolves eMerge to support the latest IBM Z releases and functionality. That’s not a passive commitment. It means our customers don’t have to choose between staying current on IBM Z capabilities and maintaining the stability of their core workloads. We do that work for them.

The result is rapid time to market. When IBM Z introduces new capabilities, like the AI inferencing advances in IBM z17, our customers can act on them quickly, without waiting for lengthy integration cycles. eMerge is already there.

We’ve seen this in practice. A government office managing a wide range of citizen-facing and back-office applications entirely on IBM Z used eMerge to replace paper-based, manual processes with fully digital, automated workflows. Citizens can now submit requests, receive real-time validation against business rules, track their case status, and receive PDF confirmations without leaving the government’s secure digital environment. What took days of manual handling now happens in minutes.

That’s the kind of transformation eMerge makes possible and IBM z17 takes it further.

“eMerge took manual processes and turned them into automated, easy–to–access digital services, increasing efficiency for both government workers and the citizens they serve.”

How We’re Integrating IBM z17 Capabilities into eMerge

Our R&D teams are actively working on integrating IBM z17 capabilities directly into the eMerge platform. The focus is on making the Telum II processor’s AI acceleration – and the broader performance gains of IBM z17 – available to eMerge customers without requiring custom integration work on their end.

AI and Automation Are Reshaping What Customers Expect

Insurance is changing faster than most platforms were designed to handle. Customers are asking for decisions in seconds, not days. Regulators are demanding greater transparency and auditability. Fraud is evolving in real time. If you can respond to all three simultaneously with speed, accuracy, and a clear audit trail you’ll be the one pulling ahead.

Our platform drives better business outcomes and delivers sustained performance across the entire insurance lifecycle from policy and underwriting through claims, reinsurance, and compliance. On IBM z17, those capabilities run on infrastructure built for exactly this kind of demand.

What Sapiens Brings to IBM Z Customers

Our relationship with IBM Z customers runs deep. We understand how mission-critical workloads actually behave in production: the edge cases, the compliance requirements, the performance characteristics that matter under load. That depth of knowledge is what lets us move quickly when new capabilities arrive.

The key benefits eMerge on IBM Z delivers:

• Consistency: unified back-office applications with a shared look, feel, and service layer across the organization
• Centralized data: all application data held in IBM Z, eliminating redundant web service calls and reducing integration complexity
• Compliance by design: business rules and regulatory validations built into every workflow
• Full digital services: no paperwork, no duplicate data, no manual errors; everything validated at the point of entry
• Secure access: applications accessible to internal teams (with appropriate permissions) and external users, with proper authentication and audit trails

Why IBM z17 Matters to Our Shared Customers

IBM z17 is the platform that makes AI-native insurance operations practical at enterprise scale. If you and your organization are running eMerge on IBM Z today, it means your existing investment gets meaningfully more powerful without migration, rearchitecture, or disruption.

For those evaluating IBM Z, the performance and efficiency data clearly make the case. Process 50% more AI inference operations per day than IBM z16, using the same credit card fraud detection model. Run AI-infused OLTP workloads with up to 5.6x fewer cores than x86. All while maintaining the security and reliability that mission-critical operations depend on.

Together, Sapiens and IBM Z are focused on one thing: helping you and insurance organizations get more from the infrastructure they already trust and use it to deliver what their customers expect next.

“IBM z17 creates new opportunities for our mutual customers to advance their business transformation strategies while enabling Sapiens to further enhance our application capabilities. Working alongside IBM Z, Sapiens is committed to helping organizations use IBM z17 to accelerate AI adoption, modernize mission-critical workloads, and strengthen operational resilience.” – Mordi Tzamit, VP and Head of eMerge Unit, Sapiens

What’s Next

IBM z17 launched on July 7, 2026. Sapiens eMerge is ready.

We’re working with insurers like you now to map out how IBM z17 capabilities connect to their specific AI priorities whether that’s fraud, underwriting, claims automation, or regulatory compliance.

If you want to understand what this means for your operations, talk to our team.

Performance Disclaimers

* 450 billion inference operations: Performance result is extrapolated from IBM® internal tests running on IBM Systems Hardware of machine type 9175. The benchmark was executed with 1 thread performing local inference operations using a LSTM based synthetic credit card fraud detection model to exploit the Integrated Accelerator for AI. A batch size of 160 was used. IBM Systems Hardware configuration: 1 LPAR running Red Hat® Enterprise Linux® 9.4 with 6 IFLs (SMT), 128 GB memory. 1 LPAR with 2 CPs, 4 zIIPs and 256 GB memory running IBM z/OS® 3.1 with IBM z/OS Container Extensions (zCX) feature. Results may vary.

** 5.6x fewer cores: Performance result is based on IBM® internal tests running on IBM Systems Hardware of machine type 9175. The MegaCard benchmark was containerized and deployed on Red Hat® OpenShift® Container Platform (RHOCP) 4.17 on Red Hat Enterprise Linux® (RHEL) 9.4 with KVM. Compared x86 Server Hardware configuration: 28 Sapphire Rapids Intel® Xeon® Gold CPUs @ 2.20 GHz with Hyper–Threading turned on, 1 TB memory. Results may vary.

The post AI That Acts at the Speed of Insurance: Sapiens eMerge and IBM z17™ appeared first on Sapiens.

Forget the productivity gains. Forget the checkbox mentality about AI. Autonomous agentic operations mark the point where AI systems do not simply analyse data or recommend action. Working seamlessly across an ecosystem of disparate platforms, they access the full operational environment, which allows them to coordinate processes across underwriting, claims, customer engagement, and risk management towards a complete and radical transformation of how we serve customers.  

Tangible Results, Not Hype 

For claims, the agentic workflow delivers faster claims resolution and improved customer experience. Instead of multiple manual steps, the entire process of submit, review, assess, and approve is dynamically orchestrated by multiple AI agents swarming to solve complex cases quickly and easily. They analyse claim documentation and images, assess damage using computer vision, check historical patterns for fraud indicators, retrieve policy details, estimate repair costs, and initiate settlement recommendations.  

What human teams once took days, weeks and months to complete now takes seconds, minutes and hours at most. Agentic seeks to eliminate process inefficiencies, overpayments, and missed subrogation, which drain 5-10% of total claims cost, resulting in avoidable material losses. It also tackles bottlenecks which often unnecessarily delay claims for weeks and months and cost companies millions in lost opportunities and inaccurate filings. Underwriting is another area where dramatic changes are happening.  

Where human underwriters used to spend 40% of their time on administrative tasks, AI agents now continuously analyse risk signals from historical claims data, environmental risk models, economic indicators, behavioral data, IoT and telematics inputs.  

In terms of customers, relationships have traditionally been reactive. Policyholders contact insurers when something goes wrong or when policies need to be renewed. Customers, who are spammed in all other areas of their life, are aware of the transactional nature of such communications and act (or don’t act) accordingly.  

In the new operational model, agentic AI enables a proactive, fast and affirmative way of doing business. AI agents are configured to monitor behavioral signals, life events, and risk indicators to anticipate customer needs. When AI approaches customers who might require additional coverage, or at risk they might not be aware of, or are going through life changes that might require policy adjustments, they are viewed as more human and empathetic, according to a survey by Lemonade.  

With personalised communication across digital channels, insurance relationships become continuous rather than episodic, which makes a profound change in customer engagement. 

I am not talking about simple automation or process efficiency. I mean that AI is doing something that human beings cannot: see across the entire organisation, analysing millions of data points and making decisions based on that data. There is nothing speculative in these results, nor do I intend to induce any of the usual paranoia about AI.  

What Agentic AI Actually Means 

Agentic AI refers to intelligent systems that understand goals and objectives, analyse complex operational data, execute actions across multiple systems, and adapt decisions based on new information.  

Instead of waiting for humans to initiate every step in a workflow, AI agents coordinate activities dynamically to achieve a defined business outcome. In the insurance context, this shift means AI systems manage large portions of the operational lifecycle, not by replacing human expertise, but by orchestrating the workflow in which that expertise operates. 

This point cannot be overemphasized. Against recent advertisements presenting AI as a replacement for people, AI and human aren’t opposites — they’re interdependent. The underwriter’s expertise becomes the training signal. Without human judgment, strategy, and relationships, agentic operations don’t function. Without AI, human expertise can’t scale. 

As I discussed recently in an interview with Life Insurance International, human capital (judgment, relationships, ingenuity) and token capital (the AI capability the firm owns and controls) are the same. Human beings spend token capital, and token capital relies on human beings. Without them, agentic operations depend on humans for complex judgment, strategy, and relationships. The underwriter’s expertise becomes the training signal.  

Insurance Reaches that “Netflix Moment”  

Every major industry eventually experiences a moment when technology stops simply supporting operations and begins actively driving them. We have seen retail experience this shift with digital commerce platforms. Finance experienced it with algorithmic trading systems. Now it’s insurance’s turn – that “Netflix moment” when the industry either completely transforms its business and dominates the future or goes the way of Blockbuster.  

While many insurance companies are still deeply invested in the previous generation of automation for document processing, routing claims to adjusters, triggering underwriting reviews and processing payments and billing events, there are many insurers using AI agents to coordinate workflows, platforms expand capabilities, and operations continuously optimize for measurable outcomes. 

Agentic AI can only function effectively within unified platform environments. If underwriting systems, claims systems, customer platforms, and data environments remain fragmented, AI agents cannot coordinate workflows across the enterprise, which is why the shift toward platform architecture is so critical. 

The insurers already operating this way aren’t running pilot programmes. They’re running the business. The question isn’t whether agentic AI will define insurance operations, it’s whether your organisation will define it or react to it.

Read the other blogs in the series:

5 Insurtech Capabilities the C-Suite is Discussing: Insider Perpective

How a Platform Approach Drives Autonomous Insurance Outcomes

Outcome-Seeking AI: Insurance’s Next Platform Evolution

The Case for Building Insurance Platforms That Actually Understand Insurance

The post When AI Begins Running the Insurance Workflow appeared first on Sapiens.

Why IUL Is Administratively Complex

Universal life products are inherently more complex than whole life or term because of their flexible structure: policyholders can adjust premium payments, change death benefits, and take policy loans at any time. IUL adds an indexed crediting layer on top of that flexibility.

The result is a product with multiple interacting calculation tracks:

• Account value: cash value of the policy, updated as premiums are received, credits are applied, and charges are deducted
• Cost of insurance (COI): monthly charge for the death benefit coverage, which varies by age, face amount, and mortality table
• Indexed crediting: interest applied based on the performance of a linked index, subject to caps, floors and participation rates tracked in segments
• Rider calculations: separate calculation tracks for each rider which may include guaranteed income benefits, chronic illness benefits, overloan protection, and return of premium features
• Policy loan and net amount at risk: loan balances reduce account value and introduce interest accumulation that must be tracked separately

Each of these calculations is manageable on its own.The challenge is that they continuously interact with one another. A policy loan may reduce the account value that is subject to indexed crediting. A chronic illness acceleration may reduce both the death benefit and cash value. A no-lapse guarantee may require a separate shadow account calculation alongside the actual account value. Platforms that manage these interactions correctly are rare.

The Indexed Crediting Engine: What It Needs to Do

The indexed crediting engine is the computational core of an IUL platform and has several specific requirements:

Multiple concurrent crediting strategies

Policyholders often allocate premiums across multiple indexed strategies. For example, 50% may go towards a point-to-point S&P 500 strategy while another 50% may be allocated to a two-year indexed strategy. The platform must track each allocation separately with its own start value, index reference, and crediting parameters.

Segment tracking at the policy level

Each segment must maintain its own:

• Start date and anniversary date
• Start index value
• Applicable cap, floor, and participation rate (which may change at renewal)
• Current credited interest (before the segment anniversary, this may be zero for point-to-point strategies)

Renewal and reallocation processing

At segment anniversaries, the credited interest is locked in and a new segment begins. At that point, policyholders may choose to reallocate funds across strategies at each anniversary. The platform must process these reallocations and start new segments correctly.

Index data feed management

The platform must receive index values from an external data provider, store them accurately by date, and apply the correct value in crediting calculations. Mistakes in data provided or incorrect index values can corrupt every calculation downstream.It is a critical operational risk that must be managed with redundancy controls and reconciliation mechanisms.

Partial surrender and premium allocation impact

Partial surrenders and additional premium payments can affect the segment structure. The platform must be able to process mid-segment transactions without disrupting the integrity of the calculation for the affected segment.

Rider Administration: The Hardest Part

Riders are where most IUL platforms struggle. The problem is not the existence of riders because most platforms can handle basic term riders or waiver of premium rider calculations. The real challenge is the interaction between rider calculations and the base contract, as well as the complexity of modern IUL rider designs.

A platform that handles riders “cleanly” should be able to:

• Run rider calculations on separate tracks without affecting base contract calculations
• Process interactions between the rider and base contract in the correct sequence (for example, when a chronic illness acceleration reduces the death benefit and account value)
• Handle rider elections, changes, and terminations without requiring manual intervention or workarounds
• Support rider-specific reporting and disclosure requirements without requiring custom development

Common IUL Riders and Their Administrative Requirements

Guaranteed Lifetime Withdrawal Benefit (GLWB) / Chronic Income Rider

These riders provide guaranteed income based on a benefit base that may grow independently of the account value. Administrative requirements include:

• Maintain a separate benefit base calculation (often with a roll-up rate)
• Track cumulative withdrawals against the guaranteed withdrawal amount
• Apply benefit base reductions for excess withdrawals
• Process income payments separately from base contract transactions

Chronic Illness Accelerated Death Benefit

This rider allows partial acceleration of death benefits upon diagnosis of a qualifying chronic condition. Administrative requirements include:

• Claim adjudication workflow for chronic illness certification
• Reduction of death benefit face amount and account value upon acceleration
• Impact on any remaining death benefit and future COI charges

Overloan Protection Rider

This rider prevents policy lapse when policy loans have reduced the account value to a critically low level. Administrative requirements include:

• Continuous monitoring of account value relative to outstanding loans
• Trigger calculations when the overloan protection conditions are met
• Conversion logic to a paid-up policy structure upon trigger

No-Lapse Guarantee (NLG)

This rider maintains the death benefit even if the account value drops to zero, provided the shadow premium test is met. Administrative requirements include:

• A separate “shadow account” calculation running in parallel with the actual account value
• Shadow COI charges applied at guaranteed rates rather than current rates
• Lapse testing based on the shadow account instead of the actual account value

Return of Premium Rider

This rider guarantees the return of premiums paid if the policy lapses or the insured dies within a specified period. Administrative requirements include:

• Tracking cumulative premiums paid (net of withdrawals and loans in some designs)
• Benefit calculation on trigger events

Policy Loan and Lapse Logic in IUL

IUL policy loans are more complex than loans on traditional whole life products because the indexed crediting creates interaction between the loan balance and segment values.

Indexed versus fixed loan strategies

Most IUL products offer both indexed and fixed loan options. With an indexed loan, the loaned amount continues to participate in indexed crediting while being charged a fixed loan interest rate. With a fixed loan, the loaned amount is moved to a fixed interest account.The platform must track which segments are subject to which loan strategy.

Lapse logic

IUL policies lapse when the account value cannot support monthly COI charges, assuming no no-lapse guarantee is in force.The platform should:

• Calculate monthly COI charges based on current face amount, age, and the net amount at risk (difference between death benefit and account value)
• Project whether the account value will sustain the policy through the grace period
• Generate lapse warnings and grace period notices accurately

Errors in lapse logic are among the most serious operational failures in life insurance administration. They can trigger inappropriate lapse of in-force policies and create regulatory and legal exposure.

How to Evaluate a Platform’s IUL Capability

Test rider interaction scenarios

Ask vendors to demonstrate a specific scenario such as a policyholder taking an indexed loan, the segment anniversary arriving and the chronic illness rider is triggered in the same month. Walk through exactly how each calculation is sequenced and how each track is updated. A platform that can walk through this clearly is a platform that handles it correctly.

Ask about COI change processing

COI rates change every year as the policyholder reaches a new attained age. Ask how the platform handles the annual COI update for a large in-force block. Is it automated or does it require manual processing?

Request an actuarial validation package

Any serious IUL platform should have an actuarial testing package that documents calculation accuracy for standard and edge case scenarios. Request it and have your actuaries review it.

Evaluate the product factory for rider configuration

Ask how a new rider is configured in the platform. Does it require developer involvement, or can a business analyst configure it? The answer reveals how quickly you can launch new product variations.

Ask about their existing IUL client base

How many IUL policies are currently in force on their platform? What is the range of rider complexity in that book? A vendor with a large, complex IUL client base has a clear capability that a vendor with minimal IUL experience cannot match.

Where Sapiens Fits

If you are launching or migrating an IUL book, the platform underneath it decides which riders you can offer and how cleanly you can service them. That makes it an architecture question before it is a product question. Sapiens Insurance Platform for Life & Annuities is built for the indexed and universal life products sold in the North American market: a calculation engine that runs concurrent crediting strategies, a rider administration framework that keeps living benefits, chronic illness acceleration, and overloan protection on separate tracks, and a product factory that lets you configure new rider designs without custom development. 

More than 600 insurers in over 30 countries run on Sapiens, backed by 40+ years of insurance domain expertise and ontology. You get a partner with specific experience in this annuities, not a general life system adapting under pressure.

The post Indexed Universal Life Administration: Choosing a Platform That Handles Riders Cleanly – In Depth Guide appeared first on Sapiens.

They’re not wrong. There’s no shortage of ambition – but almost nothing that’s actually changing how underwriting works, how risk is priced, or how customers are served. (Sapiens is changing this, but that’s a story for another day.)

What’s becoming clear is that the step from AI on slides to AI in production isn’t primarily a technology problem. It’s a human one. It’s the years of experience built over time – the deep functional knowledge of how insurers underwrite, price, and serve their customers – that makes the difference in building a Platform Insurance Company.

The Myth of the “One Platform Fits All” Approach

Across many industries, platform architectures have let companies standardise technology and simplify operations. Insurance is different. The operational models across insurance segments aren’t just varied – they’re profoundly different, with entirely distinct complexities between sectors.

A reinsurer and a workers’ compensation carrier might both call themselves insurers. But they’re running almost entirely different businesses with different timescales, different data, different regulatory pressures. They’re not playing the same game. And the technology that serves them shouldn’t pretend they are. These are entirely different definitions of what risk means. Technology needs to start there.

From Horizontal Platforms to Micro-Vertical Platforms

The insurers making the most progress right now aren’t the ones with the biggest platforms. They’re the ones with the most focused ones.

The next generation of insurance technology will shift away from generic horizontal software towards line-of-business and micro-vertical capabilities built specifically for distinct insurance domains and designed to handle the complexities that horizontal platforms paper over.

The difference is in what comes pre-built. A micro-vertical platform arrives already shaped around the regulatory environment, the data structures, and the risk logic of a specific insurance domain. There’s far less to configure because far more is already understood.

Think of it this way: instead of buying a platform and spending eighteen months explaining insurance to it, you start with one that already speaks the language. The workflows reflect how the business actually runs. The data models reflect how risk is actually structured. That changes what implementation even looks like.

Why AI Requires Domain Intelligence

This matters even more now that AI is at the centre of every platform conversation.

AI systems are only as effective as the context they operate in. A model trained on generic data will struggle with the nuances of insurance risk and in this industry, nuance is everything.

Ask any experienced underwriter. A commercial property model needs to understand exposure aggregation, catastrophe risk, and building characteristics. A life insurance model needs to interpret medical data, mortality risk, and long-term actuarial projections. A reinsurance model needs to understand treaty structures, capital allocation, and portfolio diversification. When I show underwriters what a model trained on their actual risk data can do versus a generic one, the difference isn’t marginal. It’s a different tool entirely.

This intelligence can’t be bolted on using best of breed solutions. It has to be integrated  to the platform itself through industrialised interfaces such as MCP, trained on the operational reality of each insurance vertical, not on data that happens to mention insurance.

The Power of Domain-Aware AI Platforms

What this means practically for CEOs: faster time to value from a platform that already understands your business. More accurate predictions because the model was trained on your operational reality, not generic data. Faster automation of complex workflows because the system understands the context in which decisions have to be made. And regulatory requirements built into the workflow from day one, not retrofitted at the end of the project.

That last one tends to get the most attention in the room.

The Strategic Advantage of Micro-Vertical Platforms

Insurance buyers are getting more sophisticated. They’re asking harder questions: not what the software can theoretically be configured to do, but what it actually knows about their business. That’s changing the competitive dynamics to outcomes.

The challenge for insurance is much more than technology. It’s risk intelligence. Platforms that understand the nuances of risk across different insurance verticals will have a real advantage in delivering meaningful business outcomes. And the gap between those that do and those that don’t is only going to widen.

The Role of Partner Networks in Micro-Vertical Innovation

Micro-vertical platforms won’t evolve in isolation. They’ll increasingly connect with specialised networks of partners – insurtech innovators, data providers, AI model developers, risk analytics firms, and regulatory technology providers.

These partner networks will continuously expand what the platform can do, letting insurers incorporate new data sources, analytics models, and operational tools without rebuilding or customising core systems. The platform becomes a living body of insurance intelligence that gets sharper over time.

The Next Evolution of the Platform Insurance Company

We’re moving towards platforms with more intelligence.

Insurance leaders want a platform that understands the difference between a treaty and a binder, between a long-tail liability and a short-cycle property claim and builds that knowledge into every workflow, every model, every decision. This is what separates technology that processes insurance from technology that understands it.

The next generation of insurers won’t operate on unified platforms. They’ll operate on intelligent ones, deeply aligned to the specific risk environments they serve, combining operational architecture, AI-driven intelligence, domain expertise, and connected partner innovation.

Together, these capabilities form the foundation of the Autonomous Insurance Platform and they open up a genuinely new possibility: technology that can continuously optimise risk, operations, and customer outcomes at the same time.

What Comes Next

In the next article in this series, I’ll explore another critical dimension of the Platform Insurance Company: partner network innovation.

The future of insurance technology won’t be built by a single organisation. It will emerge from connected networks of insurtech startups, data providers, AI innovators, cloud platforms, and system integrators. The platforms that best enable those connections will define the next era of insurance.

Read James’ previous posts here:

5 Insurtech Capabilities the C-Suite is Discussing: Insider Perspective

How a Platform Approach creates Autonomous Insurance Outcomes

Outcome-Seeking AI: Insurance’s Next Platform Evolution

The post The Case for Building Insurance Platforms That Actually Understand Insurance appeared first on Sapiens.

We introduced a new brand and a new London home. We launched a mission, vision, and employee value proposition with a refreshed set of company values that share our connection with customers.

I am honoured and proud to be part of this company of humans (Sapiens) who are moving with a shared purpose.

The insurance industry is (as many industries are) facing profound changes with AI – and our new brand reflects the priorities of customers. Our 40+ years of industry experience has brought us to this moment, along with months of hard work by our team (thank you all!).

Our message is informed by a simple but powerful belief: insurance works best when it understands people, not just policies. That idea shapes how we talk about our technology, our strategy, and how we view ourselves.

Brands are not built through logos or taglines alone. They are built through experiences and culture. At Sapiens, we are not changing who we are. We are becoming a more clear and bold Sapiens.

Putting People at the Centre

At the centre of our brand is a simple but powerful belief: insurance works best when it understands people, not just policies.

That idea shapes how we talk about our technology, our strategy, and how we view ourselves.
Because brands are not built through logos or taglines alone. They are built through experiences and culture. Through the behaviours people see and feel every day.

That is why this week has been about much more than marketing. Our new Employee Value Proposition and refreshed values play a critical role in the Sapiens story and reflect the kind of organisation we are building together.

An organisation that is connected globally. Invested in accountability. Engaged with people.

Supporting Insurers with New Momentum

The industry is entering a period of profound change. AI is reshaping expectations. Customers expect relevance in every interaction. Speed and personalisation are no longer differentiators; they are the baseline. Insurers are looking for partners who can help them navigate that complexity with confidence.

Our role is to give them the tools to navigate it with confidence.

For marketing teams, brand is often described as the promise you make to the market. Culture is how you keep it.

This week matters because these two principles are now fully aligned.

And what excites me most is that this is only the beginning.

A new brand does not transform a company overnight. But it can create clarity. Momentum. Shared belief. It can help people see where we are heading and why.

This is an exciting moment for Sapiens.

The post Sapiens New Brand: Who We Are and Where We’re Going  appeared first on Sapiens.