The Insurtech Reality Check: Myths, Tropes, and What Easy Money Actually Bought
Let me say what most people in this industry are thinking but few will post on LinkedIn. Insurtech has a credibility problem, and it's not because the technology is bad. The technology is genuinely getting better. The problem is that an entire generation of vendors has spent a decade selling a story that does not match the lived experience of the people writing the checks. And the people writing the checks have noticed.
A recent study found that two out of three brokers believe vendor claims of time savings, efficiency, and ROI are overstated. Only 23 percent of broker respondents feel that vendors can be counted on to do what is right. Only 9 percent agree that vendors have their best interests at heart. These are not numbers from a hostile industry survey. These are numbers from the actual customers that insurtech vendors are trying to sell to, telling researchers that they don't trust the marketing they're hearing. That's not a category in trouble. That's a category that has trained its own customers to discount everything it says.
This piece is going to be uncomfortable. I'm going to name the tropes, name some of the value props that get repeated until they lose meaning, and call out the patterns that have allowed bad selling to flourish for too long. I'm also going to be specific about what easy money has actually bought the industry, because the bill is coming due, and most of the people inside the bubble haven't figured that out yet.
I've been on both sides of this table. I've sold insurtech and I've bought it. I've sat in pitch meetings as a vendor trying to convince a carrier to give us a chance, and I've sat in evaluation meetings as a buyer listening to vendors who clearly didn't understand my business. The pattern I'm describing isn't theory. It's the lived experience of every operator in this category.
The Foundational Tropes: “Insurance Is Broken, Boring, Slow, and Confusing”
Before we get into the specific value prop tropes, we have to deal with the foundational ones. The rhetorical frames that show up in almost every insurtech pitch deck, founder LinkedIn post, and venture investment memo. Insurance is broken. Insurance is boring and confusing. Insurance is slow. Each of these frames is so common that the people repeating them have stopped noticing they are saying them, and the people on the receiving end have stopped pretending they take them seriously. These frames are the original sin of insurtech marketing, and they have done more long-term damage to vendor-buyer relationships than any specific product overpromise.
“Insurance Is Broken”
Insurance is not broken. Insurance is a $1.7 trillion US industry that has paid out claims through pandemics, wildfires, hurricanes, financial crises, two world wars, and the worst catastrophe years in recorded history, while remaining solvent and continuing to underwrite new risk. The system has problems, real ones, but a vendor walking into a carrier or MGA and telling them the business they have spent their careers building is broken is making the worst possible opening move. The buyer hears: I do not understand what you actually do, I have not bothered to learn, and I am here to fix you anyway.
What insurance actually is, when you spend time inside it, is a stunningly complex risk-transfer mechanism that prices and pools billions of individual risks while navigating fifty-state regulatory frameworks, capital adequacy requirements, reinsurance treaty mechanics, claims adjudication law, fiduciary obligations to policyholders, and ratemaking discipline that has to hold up across multi-decade cycles. The fact that this system works at all is one of the more impressive achievements in modern finance. The fact that it works imperfectly is not the same as it being broken. The pieces of the system that genuinely need improvement are specific, technical, and well-known to the people inside it. They do not need to be discovered by a 28-year-old founder reading McKinsey reports.
The vendors who do well in this market open with the opposite framing. Insurance is hard, the system mostly works, here is the specific friction point we have decided to focus on, here is why we think we can move the needle on it. That kind of framing tells the buyer that the vendor respects the industry, has chosen their problem carefully, and is not pitching salvation. The vendors who lead with "insurance is broken" are pitching salvation, and the buyers have heard that pitch a thousand times from people who turned out to be unable to deliver it.
“Insurance Is Boring and Confusing”
This is the most insulting of the foundational tropes, and the one that founders use without realizing how it lands. Insurance is not boring. Insurance is one of the most intellectually demanding industries that exists. It requires mastery of probability, capital management, regulatory law, contract construction, claims adjudication, distribution economics, capacity dynamics, and human behavior under stress. The people who run this industry are not bored. They are deeply engaged with problems that are far harder than anything most consumer tech companies have ever had to solve. The fact that the surface aesthetic is more conservative than a Silicon Valley startup says nothing about the underlying complexity.
The "confusing" version of this trope is similarly off-base. Yes, insurance contracts are dense. Yes, the average consumer does not understand the difference between actual cash value and replacement cost. Yes, the regulatory language can be opaque. None of this means the industry is failing at communication. It means the underlying product is genuinely complex, the legal obligations require precise language, and consumer education is a real and ongoing challenge that the industry has been working on for decades. A vendor who shows up with a colorful UI and declares that they are going to make insurance simple is volunteering to learn, the hard way, why the existing language exists. The lawyers in the room will explain it to them eventually. The buyers in the room have stopped waiting for that explanation to land.
The Lemonade campaign that built its brand on "insurance feels broken, we are going to make it feel like a tech product" is a useful case study here. The marketing was brilliant. The aesthetic worked. The brand resonated with a generation of younger consumers. Then the underlying economics caught up with the brand, and the combined ratios told a different story than the marketing did. Treating insurance as if it is fundamentally a UX problem produces companies that are very good at customer acquisition and very bad at underwriting discipline. The industry is not boring or confusing because it has failed to hire good designers. It is the way it is because the underlying problem is hard, and pretending otherwise is the fastest way to find out exactly how hard it actually is.
“Insurance Is Slow”
Insurance is slow when it needs to be slow, and fast when it can be fast. The trope that insurance is fundamentally a slow industry being held back by old technology is a misread of why specific processes take the time they take. A specialty submission with a $50 million property schedule, multiple coverage parts, and a complex loss history takes time to underwrite because there is real analysis happening, not because the carrier is technologically backward. A reinsurance treaty renewal takes time because billions of dollars of capacity are being committed and the modeling needs to be defensible. A complex commercial claim takes time because liability allocation, coverage interpretation, and reserve setting are genuinely hard problems with high financial and legal stakes.
The parts of the business that can be fast already are fast. Personal auto quotes are real-time. Small commercial coverage is instant for the simpler classes. Renters and homeowners admitted policies bind in minutes. The friction that remains in the system is overwhelmingly in places where speed has a tradeoff cost: underwriting discipline, regulatory compliance, capital adequacy, fraud detection, customer protection. Vendors who pitch speed without acknowledging that those tradeoffs exist are pitching to an audience that has spent careers managing exactly those tradeoffs, and they sound naive.
There is also a generational version of the slow trope that deserves explicit pushback. The framing that the insurance industry resists technology because it is dominated by older executives who do not understand modernization. This is wrong on its face. The carriers and MGAs investing most heavily in AI right now are the ones whose senior leadership has been in the industry for decades. They are not resisting technology. They are resisting bad technology, badly sold, by vendors who do not understand the business. That is a completely different problem from technological conservatism, and conflating the two is one of the surest ways to lose credibility with the people you are trying to sell to.
“Insurance Companies Don’t Want to Pay Claims”
This one needs to die, and the consumer-facing insurtech players have been the worst offenders. The marketing framing that traditional carriers are incentivized to deny claims and our AI-powered platform will make sure customers get paid is morally lazy, factually wrong, and creates real regulatory exposure for the vendors who repeat it. Carriers want to pay valid claims. Paying claims is what insurance is. Carriers also have a legal and fiduciary obligation to not pay invalid claims, because doing so transfers cost to every other policyholder in the pool. Adjudicating that line is the actual work of claims, and it requires judgment, evidence review, and policy interpretation that does not reduce cleanly to a sentiment about whether the carrier wants to pay.
The vendors who use this framing in their marketing are doing two things they probably do not realize. They are alienating every claims professional inside every carrier they want to sell to, because no claims VP wants to buy a platform from a vendor who has publicly accused the industry of acting in bad faith. They are also creating regulatory risk for themselves, because state insurance regulators do not appreciate vendors making generalized claims about industry-wide bad faith conduct that the regulator's own examination data does not support. The first time a carrier's general counsel pulls up the vendor's marketing site and shows it to the procurement committee, the deal is dead. The vendors who have built durable claims businesses know this and have scrubbed this framing out of their messaging. The ones who have not are pitching to an audience that has already filed them under "do not engage."
“Insurance Is Stuck in the 1980s”
A close cousin of the "slow" framing. Insurance is supposedly run on green-screen mainframes, paper applications, and fax machines, waiting to be modernized by Silicon Valley. This was a half-true generalization a decade ago. It is not true now. The major carriers have spent the last decade investing billions in modernization. Guidewire, Duck Creek, Sapiens, Insurity, and a dozen other modern PAS providers have replaced legacy systems across most of the industry. Cloud migration is largely done at the top end of the market and well underway in the middle. The carriers that have not modernized are smaller regionals or specialty players where the economics did not justify the investment, not because the entire industry is stuck in the past.
Vendors who walk into a major carrier in 2026 and start the pitch by referencing the carrier's legacy systems are pitching to a buyer who has just spent five years and tens of millions of dollars replacing those systems. The CIO in the room finds this framing insulting because it implies they have not been doing their job. They have been doing their job. They are likely further along in their modernization journey than the vendor realizes, and they want to talk about what the next layer of capability looks like on top of the modern stack they have already built, not be condescended to about a state that no longer exists. The vendors who do their homework on the carrier's actual tech stack before they walk in are the ones who get the meeting.
The pattern across all of these foundational tropes is the same. They are rhetorical shortcuts that vendors use to position themselves as the solution before they have established that they understand the problem. Each one signals to the buyer that the vendor has not spent serious time inside the industry and is working from a third-hand understanding that is roughly a decade out of date. The buyers have heard all of these. They have stopped politely waiting for them to pass. They mentally check out the moment any of them appears, and they remember the vendors who lean on them. The fastest way for an insurtech founder to dramatically improve their close rate would be to scrub all five of these phrases out of their pitch deck and never use them again. That single change, by itself, would do more for the category's credibility than another billion dollars of venture funding.
Myth One: AI Is the Differentiator
Walk through the marketing pages of any 20 insurtech companies. AI-native. AI-first. AI-powered. Agentic AI. Generative AI for underwriting. Generative AI for claims. AI copilots. AI underwriters. AI orchestration. The category has reached peak AI-as-adjective, and the result is that AI as a positioning claim now signals nothing. It is table stakes. It is not differentiation.
Here's the data. By 2024, 76 percent of US insurers had already integrated generative AI into their operations. Two-thirds of the $5.08 billion that flowed into insurtech in 2025 went to AI-focused companies. Every workbench vendor, every intake platform, every claims tool, every copilot now claims agentic capability, portfolio intelligence, and meaningful loss ratio improvement. When everyone has the same positioning, no one has positioning.
The real moat in insurance was never the algorithm. It was proprietary data, workflow lock-in, distribution relationships, regulatory posture, and capacity partnerships. These are the things that compound. These are the things that take years to build and are nearly impossible for a competitor to replicate. The model is the tool. The mine is the actual business, and most vendors are still selling shovels while pretending they own the mine.
I had a conversation with a chief underwriting officer at a top-25 carrier earlier this year. She had sat through eight vendor pitches in two weeks. Her observation: every single vendor opened with their AI capabilities, and every single vendor's AI capabilities sounded identical. The vendor that eventually won her business spent the first 20 minutes of their pitch talking about how their platform integrated with her treaty bordereau reporting. They didn't mention AI until 30 minutes in. That was the signal she was looking for. The vendor understood that AI was the easy part. The integration into her actual workflow was the hard part. They were the only vendor in the room who led with the hard part.
Myth Two: We're Going to Disrupt the Incumbents
This one was always going to age badly. The original insurtech thesis circa 2015 to 2020 was that legacy carriers were lumbering dinosaurs and a new generation of tech-native challengers would eat their lunch. A decade and tens of billions of dollars later, the scoreboard tells a different story. Lemonade was supposed to be the future of insurance. After ten years of operations, the company is still reporting losses, with a Q3 2025 net loss of $38 million, though it has finally hit positive free cash flow milestones. Root similarly slipped back into a $5.4 million loss after a profitable year. Hippo finally posted a full-year profit in 2025, but only after selling off business lines and absorbing a 113.1 combined ratio that would have ended a traditional carrier's career.
Meanwhile, the actual incumbents are doing fine. State Farm posted a $1.5 billion underwriting profit in 2025. Travelers acquired Corvus. Allianz X backed Openly. Munich Re acquired Next Insurance for $2.6 billion. The disruption never happened. What happened instead is that incumbents bought the best of the new technology, hired the best of the new talent, and integrated the innovations that worked into their existing operating models. The category got absorbed, not replaced.
This is the part that should make every insurtech founder pause. If your pitch deck still has a slide about disrupting the incumbents, you are pitching a story that the market has already disproven. The smartest insurtech operators figured this out three years ago and repositioned around being the technology layer that incumbents buy. The ones still selling disruption are selling a thesis that the funding environment has already abandoned.
Myth Three: We'll Replace the Underwriter
This trope shows up in almost every claims and underwriting AI pitch. Some version of "our platform handles the work so underwriters can focus on strategy" or "our AI replaces 80 percent of the manual underwriting work" or, in the most aggressive versions, "our autonomous underwriting agent removes the human from the loop." Every senior underwriting leader I work with has the same reaction to this framing. They tune out.
Here's why. Senior underwriters are not afraid of being replaced. They are afraid of being burdened with tools that promise to make them faster and instead make them slower. They are afraid of platforms that produce confident-sounding outputs that they then have to second-guess and rework. They are afraid of being held accountable for decisions made by a black box they don't understand and can't override cleanly. The replacement narrative is exactly backwards for this audience.
The platforms that have actually won enterprise deployments at scale figured out the right framing years ago. Sixfold positions its product as the AI Underwriter that handles operational work while strategy stays with the human. Kalepa calls its product Copilot. Federato emphasizes that RiskOps gives underwriters next-best-action recommendations, not autonomous decisions. The vendors who lead with augmentation win underwriting director confidence. The vendors who lead with replacement lose.
The data backs this up. According to industry research, while 88 percent of auto insurers and 70 percent of home insurers report using or planning to use AI, only 7 percent have successfully scaled AI systems into production. That gap is not a technology problem. It is a workflow and trust problem. The platforms that get stuck in pilot purgatory are overwhelmingly the ones that tried to replace the underwriter rather than augment them. The senior underwriter is the gatekeeper for whether a platform makes it from pilot to production, and selling against that gatekeeper is a losing strategy.
Myth Four: Quotes That Used to Take Weeks Now Take Minutes
This is the trope that gives away the vendor faster than almost any other. Some version of it appears on roughly half the insurtech pitch decks in circulation. Quotes that used to take days or weeks now take minutes. Submissions that languished for ten days are now bound on day one. The carrier of the past, the story goes, was a slow-moving institution that took forever to give brokers an answer, and the AI-powered future is going to fix that.
Quotes have not taken weeks for years now. If a carrier or MGA is still producing quotes on a multi-week cycle in 2026, they are a lagging operator with deeper structural problems than any insurtech platform will solve. The standard quote turnaround in most lines is hours to a few days, with same-day quoting common in admitted personal lines, small commercial, and increasingly in middle-market and E&S. Specialty and complex commercial still run longer cycles because the underlying risks require genuine analysis, not because the carrier is technologically backward. A vendor who walks into a sophisticated MGA and pitches them on going from weeks to minutes is telling that MGA, in so many words, that they have no idea what the current quote cycle actually looks like.
Speed matters. Let me be clear about that, because the contrarian framing in the other direction is also wrong. Faster is generally better, and the vendors who deliver real cycle time improvements are providing real value. The problem is not that speed is unimportant. The problem is twofold. First, speed cannot be the sole value proposition, because every vendor is now claiming speed improvements and the differentiation has collapsed. Second, the specific framing of "going from weeks to minutes" is anchored to a strawman that has not existed in most lines for years. Sophisticated buyers hear that framing and immediately discount the vendor as someone who has been reading three-year-old industry reports instead of talking to actual operators.
The speed conversation that actually resonates with buyers is more nuanced. It is about freeing up underwriter capacity to focus on the harder accounts. It is about responding to brokers faster on the submissions that are obviously in or out, so the underwriter can spend their judgment time on the close calls. It is about reducing the friction between submission and decision for the 60 to 70 percent of accounts that are essentially binary, so the team has bandwidth for the 30 to 40 percent that actually require analysis. That is a speed conversation that a senior underwriter will engage with, because it respects the reality that not all submissions are the same and not all of them should move at the same pace.
The other failure mode of speed-first selling is that it ignores everything that happens after the quote. Underwriting discipline. Capacity stability. Claims handling quality. Renewal continuity. Producer relationships. Treaty compliance. These are the things that actually determine whether an MGA hits its plan, whether a carrier expands its appetite, whether a program survives a renewal cycle. None of them are solved by faster quoting. The vendors who lead with speed-first messaging in the commercial market are pitching to a metric that the buyer values but doesn't optimize their business around. That mismatch is why so many of these pitches go nowhere even when the speed numbers are real.
There is also the Lemonade comparison that the speed crowd never quite engages with. Lemonade's marketing made fast claims and fast quoting central to its brand, and the underlying economics still required actuarial discipline that the speed-first culture struggled to build. Years of unacceptable combined ratios followed. Hippo's 113.1 combined ratio in 2025 was not a tech problem. It was an underwriting discipline problem dressed up in modern UX. Fast does not equal good. Fast plus disciplined equals good. Fast minus disciplined equals a slow-motion combined ratio crisis that takes years to unwind. The vendors selling speed without engaging with the discipline question are selling the same trade that has already burned the public insurtech carriers, and the buyers know it.
Myth Five: We Understand Your Industry
This is the most damaging trope of all, because it speaks to the credibility gap that has poisoned vendor-buyer relationships across the category. Every insurtech founder claims to understand insurance. The reality is that a meaningful portion of them have never bound a policy, chased a renewal, sat through a market cycle, or had to explain a denied claim to a customer. They have read industry reports. They have hired one or two former insurance executives. They have a vocabulary. They do not have lived experience.
The buyers can tell. I was in a conversation last quarter with the head of an MGA who had just finished evaluating three insurtech platforms. His comment, almost verbatim: "Two of them clearly didn't understand my business. The third one understood it. The third one didn't necessarily have the best technology, but they were the only ones I could imagine actually working with for five years." That's the buyer's actual decision criteria. Industry fluency beats feature parity every single time.
There is a specific tell that comes up in these conversations. When a vendor uses the phrase "loss ratio" but cannot explain why a particular book of business is running a higher loss ratio than expected, that is a vendor who has been coached on the vocabulary but does not actually understand the underlying dynamics. When a vendor talks about "capacity" but cannot articulate the difference between fronting capacity and quota share capacity, that is a vendor who learned the words without learning what they mean. When a vendor pitches an MGA without ever asking about their reinsurance treaty structure, that is a vendor who has no idea what the MGA actually optimizes for. These tells are obvious to anyone in the industry. The vendors who exhibit them lose deals they don't even realize they were close to winning.
The humility problem is real. Selling to insurance professionals while having spent very little time inside the industry is a hard sell, and the appropriate response is to be honest about it. The founders who have built credibility in this category typically open conversations with some version of: "I'm new to insurance. Here's what I've learned. Here's where I think we add value. Tell me what I'm missing." That kind of intellectual humility opens doors. The opposite, which is showing up with confident assertions about how the industry needs to be fixed by people who don't yet understand how it works, closes them.
Myth Six: Our Customer References Prove It
Every vendor has a deck full of customer logos and outcome statistics. 90 percent reduction in time to quote. 32 percent increase in premium per underwriter. 700 basis points of loss ratio improvement. These statistics are not lies, but they are also not what the buyer thinks they are. They are the best-case outcomes from a small number of deployments, often in narrowly scoped pilots, on specific lines of business, with specific carriers, under specific conditions that almost certainly do not match the buyer's actual operating context.
I have run enough pilot evaluations from the buyer side to know what these numbers actually look like under scrutiny. A 90 percent reduction in time to quote often turns out to be a 90 percent reduction in a narrowly defined slice of the quote workflow, measured under controlled conditions on a curated submission set. A 700 basis point loss ratio improvement claim, when traced back to the underlying analysis, often turns out to be a model-projected improvement based on a backtest, not a realized improvement in production. None of this is the vendor lying. It is the vendor stretching a real outcome into a marketing claim that doesn't survive contact with the buyer's actual data.
The discipline I now bring to every buyer evaluation is to demand three things. First, what was the baseline before the platform was deployed? Half the time the vendor cannot produce this number, which tells you they don't really know how much of the improvement was attributable to the platform versus other operational changes that happened in parallel. Second, what was the actual production outcome at 12 and 24 months, not the pilot result? The pilot is the easy part. Sustained production deployment is where most platforms underperform their pilot numbers by 30 to 50 percent. Third, who is the reference customer and can I talk to them directly without the vendor in the room? If the vendor controls the reference conversation, the reference is marketing. If the buyer controls it, the reference is data.
Myth Seven: Claims Is the Next Frontier and We’ve Solved It
Claims is the new submission triage. Three years ago, every insurtech pitch was an underwriting platform. Now the wave has moved to claims, and every vendor has a generative AI claims solution, an agentic FNOL platform, an automated subrogation engine, or a fraud detection model that promises 30 to 40 percent productivity gains. The marketing language has migrated almost word for word from the underwriting side, with the same tropes, the same overconfidence, and the same gap between pitch and production.
Claims is harder than underwriting, not easier, and most vendors have not figured this out yet. An underwriting decision is a forward-looking judgment based on incomplete information about a risk that has not yet materialized. A claims decision is a retrospective adjudication of facts, evidence, regulatory requirements, contract language, and customer relationships, often under emotional and legal stress. The variability in claims is staggering. A simple auto fender bender and a complex commercial property loss with multiple coverage parts and subrogation potential are not the same problem. Treating them as a single category to be solved by one platform is the kind of category error that produces failed deployments.
The other reality that claims-focused vendors consistently underweight is regulatory exposure. Unfair claims practices statutes, bad faith liability, and state-by-state market conduct regulation create a compliance environment where automating the wrong decision can produce six-figure or seven-figure penalties on a single file. The carriers that have been burned by aggressive claims automation in the past are now setting up governance frameworks that require explainable decisions, human review on any adverse determination, and audit trails that can stand up to regulatory scrutiny. A vendor that pitches their claims AI as a replacement for adjuster judgment is pitching to a buyer whose general counsel will veto the deal before it reaches the operations team. The vendors who have figured this out lead with augmentation and explainability. The ones who haven't are still selling autonomous claims agents to carriers whose chief claims officers stopped listening at the word autonomous.
There is also a customer experience angle that almost every claims-tech pitch misses. The claims moment is the moment of truth in the insurance relationship. It is the only time most customers ever interact with the carrier in a way that matters to them. A claims platform that processes faster but communicates worse, or that resolves correctly but feels cold, can destroy more retention value than the operational savings can ever recover. The carriers that have lived through this know it. The vendors who have not lived through it tend to optimize for throughput metrics that look great in a deck and produce churn metrics that look terrible eighteen months later.
Myth Eight: We’ll Modernize Distribution
The distribution tech wave has produced its own family of overused tropes. We are modernizing the broker. We are digitizing the producer workflow. We are eliminating the friction between carrier and agent. Every version of this pitch makes the same implicit assumption, which is that the existing distribution system is broken and needs to be replaced. Most of these vendors then discover, several million dollars into their roadmap, that the existing distribution system is not broken. It is the product of decades of relationships, regulatory structures, compensation economics, and trust networks that produce real economic value for carriers, brokers, and customers, even if the technology underneath it is dated.
The vendors who have actually moved the needle in distribution tech figured out that the right framing is not modernization. It is augmentation of an existing system that works imperfectly but works. Vertafore and Applied have spent decades building distribution infrastructure that the next generation of vendors is now plugging into rather than competing against. The acquisition of Cytora by Applied in September 2025 was a recognition that distribution and underwriting tech are converging at the broker management layer, and the standalone vendors who tried to bypass that layer are now finding it harder to scale than the ones who integrated into it.
The other distribution myth that needs to die is that producers and brokers are eager to adopt new technology if it makes their workflow easier. They are not, in the aggregate. A senior producer running a book of business worth several million in commission has very little incentive to risk that book on a platform change that promises efficiency gains they could not personally verify before the migration. The history of broker tech adoption is a long graveyard of platforms that were technically superior to the incumbent and lost anyway because the incumbents had relationship capital that the new entrant did not have. Vendors who underestimate the conservatism of the producer community end up with beautiful products that the actual sellers will not use. That is a real and recurring failure mode that the easy money cycle has obscured but never solved.
Myth Nine: Embedded Insurance Will Eat the Market
Embedded insurance has been the future of insurance distribution for six years running. The pitch is consistent: insurance will be sold at the point of need, embedded into the purchase of cars, homes, electronics, gig work, or travel, and the standalone insurance buying experience will become obsolete. The market opportunity numbers in these decks are always enormous, often citing trillion-dollar addressable markets. The actual revenue numbers, in practice, are smaller than the pitch suggests.
Embedded works extraordinarily well for specific product-market fits. Travel insurance bundled into ticket purchases. Renters insurance bundled into lease signing. Phone protection at the point of device purchase. These are real businesses with real economics. What embedded has not done, and shows few signs of doing in the near term, is replace the standalone purchase experience for complex coverage. Homeowners insurance with significant property values, commercial liability for a growing business, specialty coverage for a unique risk profile: these decisions still benefit from broker guidance, comparison shopping, and underwriting analysis that does not fit cleanly into a checkout flow.
The trillion-dollar embedded TAM slide deserves the same skepticism as any other pitch deck market sizing exercise. The realistic embedded opportunity is large but not unlimited, concentrated in specific product categories where the friction of a standalone purchase is high and the coverage decision is simple. Vendors who have built sustainable embedded businesses understand this and have positioned narrowly. Vendors who keep pitching embedded as the disruption of insurance writ large are recycling a pitch deck slide that the market has been politely declining to validate for half a decade.
Myth Ten: Data Is the New Oil and We’ve Got the Best Of It
Almost every insurtech vendor selling to carriers claims to have proprietary data that gives them a unique edge. The reality is that almost none of them do. Most insurtech data offerings are aggregations of publicly available sources, third-party data licensed from the same handful of providers that everyone else has access to, and a thin layer of platform usage data that does not meaningfully differentiate one vendor from another. The phrase "proprietary data" has been used so loosely in this category that it has lost almost all signaling value.
The actually defensible data positions in insurance are owned by the carriers themselves and by a small handful of data infrastructure companies. ISO and Verisk own decades of industry loss data that no vendor will replicate. LexisNexis owns identity and prior carrier history that is genuinely hard to compete with. CoreLogic owns property data. The carriers themselves own their own claims history, their own loss runs, their own renewal patterns, and their own producer performance data, which is more valuable to them than any external dataset a vendor might offer. When a vendor claims their proprietary data will improve a carrier's underwriting outcomes, the carrier's first question should be: what does your data tell you that my own data does not? In most cases, the honest answer is very little.
The vendors with actually defensible data positions are typically narrow and category-specific. CyberCube has built genuinely proprietary cyber risk data through years of dedicated focus. ICEYE has satellite imagery for catastrophe analysis. ZestyAI has property risk data that does meaningfully extend what the carrier already has. These are real data businesses. Most of the rest of the insurtech category is selling data positioning that does not survive a serious technical evaluation. The buyers who care about data are the ones who will catch this. The buyers who do not care about data will sign anyway, but the deal will underperform because the supposed data edge was never real.
What Easy Money Actually Bought
Here is where this gets uncomfortable for the people inside the bubble. Insurtech raised more than $50 billion globally between 2018 and 2025. A meaningful portion of that capital was deployed inefficiently. It funded marketing campaigns that overpromised. It funded sales teams that pitched without industry credibility. It funded product roadmaps designed to please venture board members rather than insurance buyers. It funded extended runway for companies that probably should have been allowed to die earlier so the talent could be reallocated to ventures with better unit economics.
The bill for that inefficiency is now coming due. Quarterly insurtech funding hit $1.0 billion in Q3 2025, down 17 percent from Q3 2024. Funding has stayed in the $1 billion to $1.4 billion range since early 2023, far below the $5.3 billion peak in Q4 2021. The number of investors making four or more insurtech deals fell to its lowest level since 2017. Acquisition activity is up: 21 insurtechs were acquired in Q3 2025 alone, the most since Q3 2022. The capital is concentrating and the consolidation is accelerating.
Three specific things easy money bought that the industry will have to live with for the next decade.
It Bought a Generation of Founders Who Confused Vocabulary for Understanding
The easy money cycle made it possible to raise serious capital for an insurance technology company without a single founding team member who had actually worked inside the industry. The result is a generation of insurtech companies whose product roadmaps reflect what venture investors think insurance should look like rather than what insurance operators know it actually is. Some of these companies have rebuilt their founding teams over time to address this gap. Many have not, and they are now facing the consequences in the form of stalled enterprise sales cycles and customer churn.
It Bought Marketing That Eroded Buyer Trust
When the broker community surveys show that two-thirds of buyers believe vendor claims are overstated, that did not happen by accident. It happened because the easy money cycle made it possible to pay for aggressive marketing that promised more than the products could deliver. Every insurtech vendor in the market today is paying the trust tax that the worst behavior of the boom years created. Buyers walk into evaluations assuming the claims are exaggerated until proven otherwise. That is a real cost, and it falls on the honest vendors as much as the dishonest ones.
It Bought Talent That Will Need to Be Redeployed
There are thousands of smart people working at insurtech companies that will not exist as independent entities in three years. Some of these companies will be absorbed by larger platforms. Some will be quietly wound down. Some will pivot into different categories. The talent in these companies is going to need to find new homes, and the carriers and MGAs that have been hiring insurtech operators into their internal innovation teams are going to be the beneficiaries. This is actually a positive outcome for the industry, but it is also a reminder that the easy money cycle did not create durable companies in many cases. It created a temporary holding pattern for talent that the incumbents will eventually absorb.
What Honest Selling Actually Looks Like
If I were advising an insurtech founder right now, here is the operating playbook I would walk them through. None of this is novel. All of it is what the vendors who are actually winning in this market are already doing.
Lead with the buyer's economics, not your product. The first ten minutes of every pitch should be about the buyer's loss ratio, retention, hit ratio, producer activation, or whatever the specific metric is that their boss cares about. If you cannot map your product to one of those metrics in the buyer's own language, you do not have a pitch. You have a feature list.
Quantify the cost of doing nothing, then quantify your delivered value against that. Most insurtech pitches focus on what the platform does. The buyer doesn't care what the platform does. The buyer cares what happens to their business if they don't adopt the platform versus if they do. Frame everything in that comparison. The platforms that do this well close faster and at better contract values.
Be honest about what you don't know. The senior insurance buyer has a finely tuned BS detector. They have heard every overpromise and they can spot a vendor stretching the truth from across the room. The fastest way to build trust is to be willing to say "I don't know" or "that's not what we do well" or "here's where another vendor would be a better fit for that specific use case." Vendors who can do this are rare and they win disproportionately because the buyer remembers the honesty.
Spend real time inside the workflow you claim to improve. Not a discovery call. Not a Gong recording. Actually sit next to a producer for a week. Watch an underwriter clear submissions on a Monday morning. Shadow a claims adjuster on a fender bender call. The reason so many insurtech products feel hollow is that they were designed by people who have never had to do the work. The products that succeed are the ones built by teams that internalized the workflow deeply enough that the product feels obvious to the users.
Stop chasing the headline customer logo. The biggest carrier in the room is rarely the best first customer for an early-stage insurtech. They take 18 to 24 months to sign, they require enterprise-grade compliance and security infrastructure that early-stage companies cannot afford to build, and they often expect heavy customization in exchange for the logo rights. The smart insurtech founders sign mid-market MGAs and regional carriers first, prove the product in production, build references, and then go after the headline accounts with case studies that close themselves. The vendors who skip this step often spend two years selling into Fortune 500 carriers with nothing to show for it.
The Reality in 2026
The insurtech category is not dying. Let me be clear about that, because the contrarian voices in this market sometimes overstate the case in the other direction. There are genuinely good companies in this category, building real products, serving real customers, and producing real outcomes. The point of this piece is not that insurtech is a fraud. It is that the gap between the vendors who deserve their seat at the table and the vendors who have been carried along by easy money is about to become very visible, and the industry needs to be honest about which is which.
Three things are going to happen over the next 24 months. The first is consolidation. The 21 vendors in the Datos Insights underwriting workbench navigator will not all exist as independent companies in 2028. Some will be acquired by larger platforms. Some will be acquired by carriers looking to own their underwriting infrastructure. Some will be wound down. The acquisition of Cytora by Applied in September 2025 is the template for what is coming.
The second is sorting. The vendors with genuine industry credibility, durable customer outcomes, and disciplined go-to-market motions will pull away from the rest. The funding environment is already enforcing this sort. Two-thirds of 2025 insurtech funding went to a narrow set of AI-focused leaders. The long tail of vendors is starting to find that capital is no longer available on the same terms, and many will not survive the transition.
The third is reckoning. The buyers who have spent the last decade burned by overpromises are now setting up procurement processes designed to catch them. Pilots are more structured. Reference checks are more rigorous. Contracts include more performance-based pricing. The vendors who built their go-to-market on confidence-and-charm sales motions are going to find themselves outmatched by the new buyer discipline, and they will lose deals to vendors who can actually walk into a CFO's office and defend their numbers.
None of this is a tragedy. It is the natural cycle of a category that raised too much money, made too many promises, and is now growing into the reality of what it actually can deliver. The companies that emerge from this cycle as genuine category leaders will be stronger for it. The talent that gets redeployed will be better for it. The buyers will get better products built by teams that respect the complexity of their business.
The Takeaway
Insurance is a hard business. It has been a hard business for 300 years. The people who run it are not stupid, they are not waiting to be saved by Silicon Valley, and they are not impressed by AI as a positioning claim. They are impressed by vendors who show up with humility, real industry understanding, and disciplined honesty about what their product actually does.
The myths and tropes that have dominated insurtech marketing for the last decade are running out of road. The data is in. Buyer trust has eroded. Capital is concentrating. The easy money cycle is over. What comes next is going to be slower, harder, and more honest, and the vendors who internalize that early will build the companies that actually matter in 2030.
The category does not need more shovels. It does not need more confident pitches from founders who have never worked inside the industry. It does not need more aggressive marketing of best-case outcomes as production reality. It needs operators. It needs humility. It needs honest selling. The vendors who provide those things will own the next decade. The ones who don't are going to find out, quickly and publicly, that the buyers have stopped believing them.
Fabio Faschi is an Insurance Product and Sales leader, National Producer, Board Member of the Young Risk Professionals New York City chapter and Committee Chair at RISE with over a decade of experience in the insurance industry. He has built and scaled over a dozen national brokerages and SaaS-driven insurance platforms, and is the founder of ScholarusAI.com and Hogglet.com for Enterprise AI transformation and risk management. Fabio's expertise has been featured in publications like Forbes, Consumer Affairs, Realtor.com, Apartment Therapy, SFGATE, Bankrate and Lifehacker.
