How do InsurTech companies actually make money?
Whether you're building or investing in the InsurTech market, our report covers everything you need to know: the players, the money, and what's coming next.
SUMMARY
InsurTech companies make money by taking a profitable slice of the insurance value chain, and the strongest business models today are usually software, distribution and MGAs because they can earn recurring revenue without putting their own balance sheet behind every claim.
The biggest mistake is to compare InsurTechs on premium volume alone. A company can help place billions of dollars of insurance and keep only a small fraction as revenue, while a carrier can book much more of the economics but also inherit the claims.
Full-stack insurers have the biggest upside per customer, but only when underwriting works. Root’s move from a 133.2% combined ratio in 2023 to 98.2% in 2025 shows how quickly the economics change once pricing, claims and operating costs stop consuming more than the premium base.
MGAs sit in an unusually attractive middle ground. They can control product design, underwriting and distribution while outside carriers or reinsurers supply much of the capital, which makes growth easier to finance but leaves the MGA dependent on continued insurance capacity.
Digital brokers benefit from something that looks almost like subscription economics without actually being SaaS: renewals. Once a large retained book is in place, old customers can generate far more commission revenue than newly acquired customers because the acquisition cost has already been paid.
Marketplaces such as EverQuote replace claims risk with traffic risk. Their core question is not whether policyholders generate losses later, but whether consumer acquisition remains cheaper than what insurers are willing to pay for leads.
Embedded insurance is powerful when the distribution moment is already built into another transaction. The economics come from owning the connection, APIs, insurer relationships and policy infrastructure rather than from carrying the insurance risk itself.
Insurance software has the cleanest operating model in the group. Guidewire can grow recurring software revenue without creating a matching claims pool, which is why its economics are structurally easier to scale than those of a carrier even if enterprise sales remain slow and expensive.
Reinsurance can change the apparent economics of an InsurTech without much change in customer demand. Keeping more premium can lift reported revenue and underwriting upside, but it also means keeping more losses and tying up more capital.
AI is economically useful only when it lowers a real insurance cost: claims handling, loss ratios, pricing error, acquisition expense or back-office work. Funding concentration around AI says a lot about investor appetite, but much less about whether insurers are actually earning better margins.
The clearest way to judge any InsurTech is to identify the cost that scales with growth. For carriers it is claims, for marketplaces it is traffic, for brokers it is producer compensation and retention, and for software companies it is cloud infrastructure, product development and enterprise sales.
The old InsurTech idea that technology itself was the business model has not held up. The winners are using technology to improve an old insurance profit pool, take a larger share of it, or reach it with much less capital.
Big announcements don't always turn into real results. Our InsurTech market report shows where things actually get stuck.
What actually counts as an InsurTech company today?
InsurTech currently covers several completely different businesses, from insurers taking claims risk themselves to software companies that never touch an insurance policy.
That distinction has become more important as the market has matured. Lemonade, Root and Oscar Health behave economically like insurers: customers pay premiums, claims consume part of those premiums, and the company makes money when pricing and operating costs work in its favor. EverQuote sits somewhere else entirely. It brings insurance shoppers to carriers and agents and gets paid for those leads. Guidewire sells software to insurers. Cover Genius and bolttech help other companies add insurance to their own customer journeys.
MGAs create another category. They can design products, select risks and distribute policies while outside insurers or reinsurers provide much of the capital behind those policies.
The funding market is also pushing the definition wider. Gallagher Re's latest quarterly InsurTech report put global funding at $2.44 billion, the highest quarterly figure since 2022. More strikingly, 99.1% went to AI-focused companies. Some of today's best-funded InsurTech businesses therefore look much more like insurance technology suppliers than the digital insurers that dominated the first InsurTech boom.
So when we ask how InsurTech companies make money, the useful question is where they sit in the insurance chain and which piece of the economics they actually keep.
| InsurTech model | What it actually sells | Main revenue source | Who usually carries the insurance risk? |
|---|---|---|---|
| Full-stack insurer | Insurance coverage | Earned premium | InsurTech |
| MGA | Product creation and underwriting | Commissions, fees, profit share | Carrier or reinsurer, sometimes shared |
| Broker or agency | Distribution and advice | Commissions and fees | Carrier |
| Marketplace | Insurance shoppers | Referral or lead payments | Carrier |
| Embedded insurance platform | Distribution inside another product | Commission, revenue share, platform fees | Carrier |
| Insurance SaaS | Software | Subscription, license and usage fees | Customer insurer |
| Claims or risk software | Claims automation or risk tools | SaaS and service fees | Customer insurer |
| Hybrid InsurTech | Several of these services | Premiums plus commissions or fees | Depends on the business |
Is insurance premium the same thing as InsurTech revenue?
No. Premium volume can make an InsurTech look much larger than the amount of money it actually keeps.
A company can help place $1 billion of insurance and record only a fraction of that amount as revenue. That happens constantly with brokers, MGAs and embedded insurance companies because most of the premium ultimately belongs to the insurer responsible for paying claims.
Kin gives us a useful example. The reciprocal insurance exchanges it manages generated $218.9 million of gross written premium in one recent quarter, while Kin itself recorded $68 million of revenue. Kin earns management and insurance-related fees around the business written by those exchanges rather than booking every premium dollar as ordinary company revenue.
Hippo shows how the distinction also affects insurers. In a recent quarter, Hippo produced $482.2 million of gross written premium but reported $144.7 million of revenue. A large reason for the gap is reinsurance: Hippo passed part of the premium and part of the underlying risk to other insurers.
This is why gross written premium needs to be read carefully. Premium tells us how much insurance is flowing through the business. Revenue tells us more about how much of that flow the company can actually monetize.
Price is often what holds a market back. See how fast that's changing in our InsurTech market report.
How do full-stack InsurTech insurers actually make money?
Full-stack InsurTech insurers make money when premiums comfortably cover claims, claims handling, operating expenses and the cost of running the insurance balance sheet.
Root currently gives us one of the clearest examples of that model working. In its latest reported quarter, Root earned $727.2 million of net earned premium and recorded a 92.1% net combined ratio. Roughly speaking, claims and underwriting expenses consumed about 92 cents for each dollar of net premium, leaving about eight cents of underwriting room before other corporate expenses and investment income. Root still finished the quarter with $25.4 million of net income.
The interesting part is how different those economics look from only a few years ago. Root's net combined ratio was 133.2% in 2023. At that level, every dollar of premium brought in far more than a dollar of claims and underwriting expenses. By 2025, the ratio had fallen to 98.2%, and Root moved from a $147.4 million annual net loss in 2023 to $40.3 million of net income in 2025.
Oscar Health gives us the same basic model in health insurance on a much larger revenue base. Its latest quarterly revenue reached $4.88 billion. The medical loss ratio was 79.2%, while selling, general and administrative expenses represented another 14.2% of revenue. Oscar still produced $388.6 million of operating income.
Lemonade is currently further back on that path. Its gross loss ratio has improved sharply, reaching 60% in its latest reported quarter versus 67% a year earlier and 79% two years earlier. The improvement is real, although Lemonade still recorded a $43.4 million quarterly net loss.
A digital insurer eventually has to win the same mathematical game as every other insurer. Better apps and automated claims help, but the business starts making serious money when the company consistently charges enough for the risks it accepts.
Why would an InsurTech become an MGA instead of an insurance company?
An MGA lets an InsurTech control underwriting and distribution without putting the full insurance balance sheet behind every policy.
That can be a very attractive compromise. An MGA may design a specialist insurance product, decide which customers qualify, set pricing within agreed rules and manage distribution. A carrier or reinsurer then supplies much of the regulated insurance capital.
The MGA gets paid through commissions, management fees and sometimes performance-related compensation. It gives up part of the potential underwriting profit, but it also avoids funding the entire claim exposure.
The scale of this model has become hard to dismiss. Conning's latest study estimates that about $128 billion of U.S. premium flowed through MGAs in 2025. The statutory data it analyzed captured $102.6 billion, up roughly 12% in a year when the wider U.S. property and casualty industry grew much more slowly.
TWFG gives us a concrete view of the economics. Its MGA programs can carry commission rates above 20%, compared with roughly 12% for its core Agency-in-a-Box network. In its latest quarter, the company generated $87.5 million of total revenue on $569.9 million of written premium and reached a 19.7% net income margin.
wefox also provides an interesting strategic example. After spending years building a carrier, it sold that insurance company and shifted its focus toward an asset-light model built around MGAs and distribution. That move says quite a lot about the trade-off. Owning the carrier gives a company access to more of the insurance profit pool, while the MGA route usually makes growth easier to finance.
There is still a catch. MGAs depend on outside insurance capacity. If carriers decide that a product is underpriced or stop wanting the exposure, an MGA cannot keep writing unlimited business on its own.
If you want more recent data on this point, please see our latest InsurTech market report.
Some businesses here depend on just a few buyers. Find out who in our InsurTech market report.
How do digital insurance brokers make money?
Digital insurance brokers mainly make money every time they place or renew a policy, without having to pay the claims on that policy.
Goosehead makes the renewal economics especially easy to see. In one recent quarter, renewal commissions and renewal royalty fees together generated about $73.5 million, compared with roughly $22.1 million from new-business revenue.
So the old customers were producing more than three times as much directly comparable revenue as newly written customers during the quarter.
The economics can become attractive once the customer base is large enough. Goosehead reported an 86% client-retention rate, which gives it another chance to earn commission when a policy renews without repeating the full cost of finding the customer from scratch.
TWFG works in a similar way. Commission income accounted for $80.6 million of its $87.5 million of quarterly revenue, while the company also earned smaller amounts through policy fees, branch fees, licensing and contingent income.
The broker never captures the whole insurance premium, but it also avoids the largest expense in insurance: paying claims. For a well-retained customer base, that can be a very good trade.
How does an insurance marketplace like EverQuote make money?
EverQuote makes money by finding people who want insurance and selling access to those potential customers to carriers and agents.
The company generated $195.1 million of revenue in its latest quarter, including $172.1 million from auto insurance and $23 million from home and renters insurance. It also produced $19.2 million of net income and $30.1 million of adjusted EBITDA.
Those figures show that insurance lead generation can clearly be profitable. The more revealing number, however, is EverQuote's $56.9 million of variable marketing dollars.
EverQuote spends heavily on advertising to attract consumers in the first place. Variable marketing dollars measure what remains after those advertising costs. Roughly 29% of quarterly revenue survived that first layer of spending.
The marketplace model replaces insurance risk with acquisition risk. EverQuote does not care whether a referred customer later crashes a car and generates a $30,000 claim. It cares a great deal whether obtaining that customer costs $80 when an insurer will only pay $70 for the lead.
That difference explains why digital marketplaces can scale quickly when insurers are aggressively chasing new policies and slow down when carriers pull back their marketing budgets.
If you want more recent data on this point, please see our latest InsurTech market report.
Rules can make or break this market, and companies know it. Our InsurTech market report covers who's shaping them.
Why is embedded insurance becoming such a big InsurTech business?
Embedded insurance can produce attractive distribution economics because the customer is already buying something else when the insurance offer appears.
Travel insurance during a hotel booking is the obvious example. The same logic works with smartphones, cars, financial accounts, concert tickets, electronics and dozens of other products.
The InsurTech does not have to convince someone to visit an insurance website from scratch. The customer is already at the exact moment when a new risk appears.
Cover Genius has built its model around that idea. Its platform works with more than 200 partners and over 50 insurers, including digital businesses such as Booking.com, Agoda, Revolut and Uber. The company says it has passed $3 billion in cumulative gross written sales and recently reported 50% annual revenue growth.
bolttech operates on an even broader network. It connects more than 700 distribution partners to thousands of insurance products across dozens of markets and says its platform processes more than $90 billion of quoted premium annually.
Neither figure should be confused with revenue. What those numbers show is the distribution leverage available when an InsurTech sits between insurers and companies that already have millions of customers.
The strongest embedded insurance businesses make money by owning the connection rather than the underlying risk. They provide the APIs, policy administration, licensing infrastructure and insurer relationships needed to make insurance appear inside someone else's product.
Is selling insurance software actually a better business than selling insurance?
For many InsurTech companies, yes. Insurance software currently has cleaner economics than underwriting insurance directly.
Guidewire shows why. Its latest fiscal-year revenue reached $1.48 billion, up 23%, while subscription and support revenue increased 33% to $970.9 million. Annual recurring revenue reached roughly $1.24 billion, and the company generated about $390 million of operating cash flow.
Guidewire can grow those recurring software payments without creating a matching pool of insurance claims.
The contrast with a carrier is substantial. An insurer running at a 95% combined ratio has only about five underwriting cents left from each net premium dollar before other corporate costs and investment income. Guidewire's subscription economics leave a much larger share of each revenue dollar available after the direct cost of delivering the software.
Guidewire still has real costs. It spends heavily on research, cloud infrastructure, sales and implementation, and large insurers can take years to change core systems.
Even with those drawbacks, software removes the nastiest form of insurance volatility. A hurricane can create billions of dollars of unexpected claims for property insurers in a few hours. It cannot suddenly create billions of dollars of claims against an insurance software vendor.
| Company/model | Latest useful revenue indicator | Recent profitability indicator | Main thing that can go wrong |
|---|---|---|---|
| Guidewire / software | $1.48B fiscal-year revenue | $390M operating cash flow | Slow enterprise sales and competition |
| TWFG / distribution + MGA | $87.5M quarterly revenue | 19.7% net margin | Losing carrier relationships |
| EverQuote / marketplace | $195.1M quarterly revenue | $30.1M adjusted EBITDA | Traffic becoming too expensive |
| Root / carrier | $727.2M quarterly net earned premium | 92.1% combined ratio | Claims and bad pricing |
| Hippo / hybrid insurer | $144.7M quarterly revenue | 95.8% combined ratio | Property losses and catastrophe exposure |
| Lemonade / carrier | $294.4M quarterly revenue | Still loss-making | Claims plus operating expenses |
| Oscar / health insurer | $4.88B quarterly revenue | $388.6M operating income | Medical costs and regulation |
If you want more recent data on this point, please see our latest InsurTech market report.
Many players in this market still lose money on every sale. See who's closest to profit in our InsurTech market report.
Why are renewals so valuable to InsurTech companies?
Renewals can turn an expensive first insurance sale into several years of much better economics.
The broker model shows this particularly clearly. As we saw above, Goosehead recently generated more than three times as much directly comparable revenue from renewals as from new business. That is what a mature insurance book can eventually look like.
The principle also applies to carriers. Lemonade's annual dollar retention recently stood at 85%, while premium per customer increased from $402 to $433 over twelve months. Existing customers therefore continued contributing a large part of the company's premium base while spending more on average.
Kin also breaks out fee revenue associated with new and renewal premium because the distinction is economically important. New policies require distribution effort. Renewals can generate another round of fees from customers the company already knows.
This changes the way we should look at growth. Two InsurTechs can add the same amount of premium during a year while creating very different long-term value. The company that keeps most of those customers can gradually build a large recurring revenue base. The company replacing churned customers every year stays trapped in acquisition mode.
Why does reinsurance change an InsurTech's economics so much?
Reinsurance decides how much insurance risk an InsurTech keeps, so it can change revenue growth, capital needs and profitability even when customer demand barely moves.
Lemonade provides a recent example. Its latest quarterly revenue grew much faster than gross earned premium partly because the company kept more of its insurance premium instead of ceding it to reinsurers. Its renewed quota-share structure reduced the effective share ceded from roughly 20% to around 18%.
That sounds like a small change, but it means Lemonade keeps more premium and also keeps more of the losses attached to that premium.
Root has moved in the same broad direction over time by reducing its reliance on external quota-share reinsurance as its balance sheet and underwriting results improved. Keeping more risk gives Root access to more underwriting profit when its pricing works. It also requires more capital and creates more exposure when losses are worse than expected.
Property insurance shows how dramatic the reinsurance bill can become. Kin said the reciprocal exchanges it manages recently reduced catastrophe reinsurance costs by 25% per dollar of protection. For a homeowners business exposed to hurricanes, that kind of reduction can improve economics without adding a single new policy.
Risk-bearing InsurTechs also get one extra source of income from keeping insurance assets on their balance sheets: investment returns. Hippo, for example, generated $25.4 million of net investment income in 2025. That helps, especially as an insurer becomes larger, although a decent investment yield cannot rescue badly priced insurance for long.
Failures teach as much as successes. We break down the biggest falls in our InsurTech market report.
Is AI actually making InsurTech companies more profitable?
AI can improve InsurTech margins today, but it only deserves credit when an actual insurance cost comes down.
The current investment enthusiasm is extreme. Gallagher Re found that AI-focused companies attracted 99.1% of global InsurTech funding in its latest quarterly report, up from 95.2% in the previous quarter. Every funding round above $5 million went to an AI-focused company.
That tells us where investors are putting money. It does not tell us how much insurers are saving.
The better evidence comes from operating metrics. Lemonade's loss-adjustment-expense component recently fell to about 5% of gross earned premium from roughly 7% a year earlier. Automated claims handling is one factor the company points to when explaining how it can process a growing policy base without building a traditional claims operation at the same rate.
Guidewire gives us another version of the AI opportunity. Insurers are buying AI functions through the same cloud platform they already use for policy administration, billing, pricing and claims. Guidewire's subscription and support revenue recently increased 33%, and management specifically pointed to growing customer demand for AI-focused products.
The bar should stay high. An AI chatbot does very little for an insurer that consistently misprices risk. A two-point improvement in loss ratio across billions of dollars of premium is far more valuable than a flashy interface.
If you want more recent data on this point, please see our latest InsurTech market report.
Why can an InsurTech grow quickly and still lose money?
An InsurTech can grow very quickly while destroying value if every new policy brings too many claims, advertising costs or operating expenses with it.
Root's history makes the point better than almost any theory. Gross written premium grew substantially between 2023 and 2025, yet the decisive improvement was its combined ratio falling from 133.2% to 98.2%. Root moved into annual profitability when the economics of each premium dollar improved.
Lemonade currently shows the earlier stage of the same process. Its latest quarterly revenue reached $294.4 million and grew sharply, while its gross loss ratio continued to improve. The company still lost $43.4 million on a GAAP basis. Revenue growth alone has not finished the job.
The problem looks different at EverQuote because it does not pay insurance claims. Its constraint is marketing. The company recently generated $195.1 million of quarterly revenue but only $56.9 million of variable marketing dollars after advertising costs.
A useful way to judge any InsurTech is to identify the expensive input that scales with growth. For a carrier, it is usually claims. For a marketplace, it is traffic. For a broker, commissions paid to producers can matter. For a software company, cloud costs, R&D and sales are much more important.
Once we know which cost grows alongside revenue, headline growth becomes much easier to interpret.
Some companies do a lot more with fewer people. Our InsurTech market report shows who runs the leanest business.
Which InsurTech business models have the best economics right now?
Insurance software, mature distribution and well-run MGAs currently offer the cleanest economics, while carriers can make more money per customer if they are genuinely good at underwriting.
Guidewire is a strong example on the software side. The company has roughly $1.24 billion of annual recurring revenue and generated about $390 million of operating cash flow in its latest fiscal year.
TWFG shows what good distribution economics can look like. Its latest quarterly revenue rose 45% to $87.5 million, net income reached $17.3 million and adjusted EBITDA margin reached 30.4%. Its higher-commission MGA programs are also becoming more important to the revenue mix.
EverQuote demonstrates that marketplaces can be solid businesses once customer acquisition is under control. Its latest quarterly adjusted EBITDA reached $30.1 million, up 37% from a year earlier.
Carriers have a harder road, but the upside is larger. Root's 92.1% combined ratio shows that a digital carrier can eventually generate real underwriting profit. Oscar's latest quarter produced almost $389 million of operating income.
The trade-off is pretty clear. Software and distribution businesses surrender much of the premium pool in exchange for avoiding most claims risk. Carriers keep far more of the economics when underwriting goes well and absorb the damage when it does not.
| Model | How attractive are the economics today? | Capital needed | Where the upside comes from | Biggest weakness |
|---|---|---|---|---|
| Insurance SaaS | Very attractive at scale | Low | Recurring high-margin software revenue | Long enterprise sales cycles |
| Broker or agency | Attractive with strong retention | Low | Renewal commissions | Dependence on carriers |
| MGA | Very attractive when underwriting works | Low to medium | Underwriting economics without funding every claim | Capacity can disappear |
| Embedded insurance | Attractive with strong partners | Low | Cheap access to existing customers | Partner concentration |
| Marketplace | Good when acquisition costs behave | Low | Scalable lead generation | Advertising prices |
| Full-stack carrier | Potentially excellent | High | Full underwriting and investment economics | Claims volatility |
| Hybrid InsurTech | Depends heavily on mix | Medium to high | Several revenue streams | More complicated economics |
If you want more recent data on this point, please see our latest InsurTech market report.
So how do InsurTech companies actually make money?
InsurTech companies currently make money by taking a profitable slice of the insurance value chain, and the best businesses are becoming much more selective about which slice they want.
Full-stack insurers such as Root, Lemonade and Oscar collect premiums and try to keep what remains after claims and expenses. MGAs get paid for designing, pricing and distributing insurance while outside carriers supply much of the capital. Brokers earn commissions from sales and renewals. EverQuote sells access to insurance shoppers. Embedded insurance companies earn money by putting policies inside other companies' products. Guidewire gets recurring software revenue from the insurers themselves.
The evidence today points to a fairly clear hierarchy.
Software, distribution and MGA models usually offer the cleanest risk-adjusted economics because they can earn recurring revenue without putting their own balance sheet behind every claim. The recent performance of Guidewire and TWFG shows how profitable that setup can become at scale.
A good carrier can ultimately capture more value. Root has now shown that digital insurance can produce real underwriting profit once pricing and claims economics work. Oscar's recent profitability makes the same point at much greater revenue scale.
That upside comes with a price. Carriers need more capital, buy reinsurance, hold reserves and live with the possibility that claims suddenly move against them.
The original InsurTech boom often treated technology itself as the business model. That idea has aged badly. Technology becomes economically valuable when it lowers claims costs, finds customers more cheaply, prices risks better, automates expensive work or gives an InsurTech control of a profitable distribution channel.
So the best answer today is straightforward: InsurTech companies make money in many of the same places where the insurance industry has always made money, but the strongest ones are using technology to take a larger margin from those activities or to reach that margin with much less capital.
OUR METHODOLOGY
The economics of InsurTech are easy to oversimplify because companies grouped under the same label can make money in fundamentally different ways. Instead of relying on broad industry narratives or judging the sector from headline premium or revenue figures, we broke the question into the dimensions that actually determine the business model: where each company sits in the insurance value chain, how much of the underlying premium it keeps economically, which costs scale with growth, how recurring the revenue is, how much insurance and balance-sheet risk the company carries, and whether technology is producing a measurable economic advantage.
For each dimension, we used the freshest operating evidence available and focused on the metric that best fits the model. For risk-bearing insurers, that means earned premium, loss ratios, combined ratios, reinsurance and capital intensity. For brokers and MGAs, it means commissions, renewal economics, retention and written premium. For marketplaces, it means acquisition economics. For software companies, recurring revenue, subscription growth and cash generation matter much more than premium volume.
Premium, revenue and profit are not treated as interchangeable. Gross written premium is useful for understanding how much insurance activity passes through a business, but brokers, MGAs, embedded-insurance platforms and reinsured carriers may keep only part of that amount economically. Where reinsurance materially changes premium retention or risk retention, we look through the headline number and focus on what the company actually keeps and what losses it still has to absorb.
The final assessment is therefore based on several pieces of evidence pointing in the same direction rather than on one profitability ranking. That is especially important in InsurTech because a software vendor, an MGA, a lead marketplace and a full-stack insurer can all report strong growth while having completely different unit economics, capital needs and failure modes.
We prioritized regulatory filings, earnings releases, company operating disclosures and specialist insurance-industry research. Key sources include Gallagher Re's Global InsurTech Report Q2 2026, Conning's 2026 MGA Study, Root's Q2 2026 shareholder materials, Root's 2025 Form 10-K, Oscar Health's Q2 2026 results, Lemonade's Q2 2026 SEC filing, Kin's Q2 2026 operating results, and Kin's 2026-27 reinsurance update.
We also used TWFG's Q2 2026 results, Goosehead Insurance's Q2 2026 Form 10-Q, EverQuote's Q2 2026 results, Hippo's Q2 2026 Form 10-Q, Guidewire's FY2026 results, Cover Genius's 2026 company update, bolttech's current operating footprint, and wefox's strategic restructuring update.
Some valuations look stretched, others look cheap. We dig into the numbers in our InsurTech market report.
