Yesterday Lemonade reported a quarter with 32.4% premium growth and one central claim: an LAE ratio of 5% against about 9% for the industry is measurable proof of technological superiority. The claim is correct, and the number is real.
This morning two more Israeli companies in the same world reported - and both crossed into profit.
Pagaya: Beat Its Own Guidance, on Two Metrics
This is the point easy to miss and the most important in the report: Pagaya did not beat consensus - it beat the top of the guidance range it issued itself.
| Metric | Q2 2026 | Company guidance | Year-over-year |
|---|---|---|---|
| GAAP net income | $45 million | $25-45 million | +$29 million |
| Adjusted EBITDA | $124 million | $100-115 million | +43% |
| Total revenue and other income | $387 million | - | +19% |
| Network volume | $3.5 billion | - | +33% |
Both of those overshoots are rare. Beating an analyst consensus is one thing - that is an outside party's estimate. Beating the top of your own guidance is something else: it says management, which sees the numbers from the inside, was too conservative about its own quarter.
And the company raised its full-year net income guidance.
What Pagaya actually does - and why the model is interesting
Pagaya is not a bank and not a classic lender. It sells an AI credit-underwriting model to other financial institutions - banks, fintechs, auto dealers.
The idea: when a customer applies for a loan and the bank's own system declines them, Pagaya evaluates the application through its model. If the model says the borrower is sound - it enables the loan and takes on part of the risk.
Which makes "network volume" its central metric - the total loans that passed through the system. $3.5 billion in a quarter, growing 33%, means more partners routing more volume through it.
CEO and co-founder Gal Krubiner: "Our record quarter reflects a flywheel that is clearly working: partners are sending more volume, adopting more products, and our network effects compound with every relationship we add."
Hippo: Reached Underwriting Profitability - the Metric That Decides in Insurance
| Metric | Q2 2026 | Year ago |
|---|---|---|
| Gross written premium | $482 million | +61.5% |
| Net income | $10 million | $1 million |
| Adjusted net income | $21 million ($0.79 per share) | +23.5% |
| Revenue | $145 million | +23.4% |
| Combined ratio | 95.8% | improved 4 points |
| Net loss ratio | 50.4% | 47.0% |
| Loss ratio excluding catastrophes | 45.8% | improved from 46.4% |
| Book value per share | about $17 | - |
And full-year guidance was raised: over $1.65 billion of gross written premium, and up to $70 million of adjusted net income.
Why the combined ratio is the only metric that truly matters in insurance
The combined ratio is total claims and expenses as a percentage of premium.
Above 100% - the company loses money on the underwriting itself. It collects $100 of premium and spends more, and if it is nonetheless profitable, that comes only from investing the float.
Below 100% - the company makes money from insuring. That is underwriting profitability, and it is the real test of an insurer: does it know how to price risk.
Hippo reached 95.8%. Meaning for every $100 of premium it spends $95.80 - and earns $4.20 from underwriting, before investment income.
And a 4-point improvement in a single year is very fast in insurance. It comes from better pricing, a better risk mix, and better customer selection.
And one point that needs clarifying: the net loss ratio rose from 47.0% to 50.4%, which looks like deterioration. But excluding catastrophes it actually improved - from 46.4% to 45.8%. The gap between the two is weather, not underwriting. In home insurance that distinction is what determines whether management is good or the year was kind.
And the Comparison That Demands Thought: Hippo Against Lemonade
| Hippo | Lemonade | |
|---|---|---|
| Premium growth | +61.5% | +32.4% |
| GAAP net income | +$10 million | -$43 million |
| Underwriting profitability | Yes (95.8% combined ratio) | No |
| Adjusted EBITDA profitability | Yes | Guides to Q4 |
| Market cap | about $0.8 billion | about $3.6 billion |
Two digital home insurers, both with Israeli roots, both selling the same story of technology making insurance more efficient.
One grows nearly twice as fast, is GAAP profitable, and is underwriting profitable. The other is worth four and a half times as much in the market.
How to explain the gap - three possibilities
First: the market prices the story, not the quarter. Lemonade has built a recognized consumer brand, a presence across five insurance categories, and a technology thesis that is easy to explain - a 5% LAE ratio against 9% for the industry. A clear story is worth a multiple, even when the profit is not there yet.
Second: scale and trajectory. Lemonade has reached $1.43 billion of in-force premium, while Hippo is running at $1.65 billion gross annually - a similar order of magnitude, but Lemonade got there through consistent growth across five categories, which reads as a repeatable path.
And third, the uncomfortable possibility: the market is simply wrong about something. Either it is pricing Lemonade too richly, or it is pricing Hippo too cheaply. This morning's reports do not settle it - but they put the question on the table in a way that is hard to ignore.
And fairness requires stating the other side: Lemonade's LAE ratio has real meaning. 5% against 9% for the industry is a measurement, not a slogan - and it is a metric regulators require everyone to publish under the same definition. Hippo did not publish a comparable LAE ratio in its release, so a direct comparison on the operating-efficiency axis is not available right now. Anyone who wants it will need both companies' full financial statements.
And the Thread Connecting Both to Yesterday's Piece
Pagaya and SoFi: exactly the same question
Yesterday SoFi posted records across every metric and still trades far from its high, and the question we unpacked was: why does a company clearing the Rule of 40 nearly twofold not receive a technology multiple.
The answer we reached: the Rule of 40 measures growth and profitability. It does not measure credit risk. A software company finishes the transaction when the customer pays; a lender begins it.
And Pagaya is exactly the same question in a purer form. It sells a model, not a loan - its revenue should be free of credit risk. But it also takes on part of the risk, which places it between the two.
And the stock is down 24.7% year-to-date, despite a record quarter and beating its own guidance. That is precisely the same phenomenon.
What to Track in Both
At Pagaya: (1) actual credit losses against the model - that is the entire business. An AI underwriting model is worth exactly as much as its accuracy; (2) network volume - $3.5 billion, and whether 33% growth continues; (3) partner count and product adoption - Krubiner talks about a "flywheel", and this is the metric that validates it; (4) whether GAAP net income holds, or was one-time.
At Hippo: (1) the combined ratio - whether 95.8% holds, or a catastrophe quarter pushes it back above 100%; (2) the loss ratio excluding catastrophes - 45.8%, the clean measure of underwriting quality; (3) premium pace - 61.5% is very fast, and the question is what it cost in underwriting; (4) the LAE ratio, if and when it publishes one - that is what would allow a direct comparison to Lemonade.
The debate in one line
The bulls see two Israeli companies that crossed into GAAP profit, beat their own guidance, raised outlooks, and are growing fast - at market caps that look low relative to that. The bears see two small companies in cyclical markets - consumer credit and home insurance - where one good quarter proves nothing, and whose profitability is too new to know whether it is structural. Both sides are reading the same reports.
My Angle
A personal opinion of Ilan Abramov - not advice, not a recommendation
What catches me in both of these reports is one thing they did that almost no company does: they beat their own guidance.
Beating an analyst consensus is nice, but it mostly means the analysts were wrong. Beating the top of the guidance you issued yourself three months earlier is something else - it is a statement that management, which sees the pipeline from the inside, was too conservative. Pagaya did it on two metrics at once.
And what interests me more than any single number is the comparison between Hippo and Lemonade.
I do not think the market is "wrong" - but I do think this gap deserves a question. A company with a 95.8% combined ratio, premium growing 61.5% and GAAP net income is worth a fifth of what a loss-making company in the same industry is worth. The plausible explanation is brand and story - and in markets, brand and story are real value, not illusion. But they are also the thing that eventually converges to numbers.
And the caution I hold on myself: new profitability is not structural profitability. Hippo went from $1 million to $10 million of net income - and one catastrophe quarter in home insurance can reverse that. Its loss ratio rose from 47% to 50.4%, and the explanation - weather - is correct and also a reminder that this is a business dependent on the sky.
And what I will watch: at Pagaya, actual credit losses against the model. That is the whole business. A company selling an underwriting model is worth exactly as much as its accuracy, and a record profit quarter does not prove the model is right - it proves it was right about loans written two years ago. The real test arrives when the credit cycle tightens.
Summary
Pagaya reported a record quarter: GAAP net income of $45 million - above the top of the guidance the company issued itself - adjusted EBITDA of $124 million (+43%), above its guide, revenue of $387 million (+19%) and network volume of $3.5 billion (+33%). Full-year net income guidance was raised.
Hippo reported gross written premium of $482 million (+61.5%), net income of $10 million against $1 million a year ago, and a combined ratio of 95.8% - underwriting profitability, improved by 4 points. The loss ratio excluding catastrophes fell to 45.8%. Guidance was raised to over $1.65 billion of premium and up to $70 million of adjusted net income.
And the comparison that stays open: Lemonade, which reported yesterday, grows at half Hippo's pace, loses $43 million a quarter - and is worth four and a half times as much. The question for the investor is not which of the two is better managed, but what the market is pricing when it prices "insurtech" - the numbers, or the story.
Sources: Pagaya Technologies' official results release for the second quarter of 2026 (July 30, 2026), as filed with the U.S. Securities and Exchange Commission on Form 8-K, including GAAP net income against the guidance range, adjusted EBITDA, revenue, network volume, the raised outlook and CEO Gal Krubiner's remarks; Hippo Holdings' official results release for the same quarter and date, including gross written premium, net and adjusted income, revenue, the combined ratio, loss ratios, book value and the raised guidance; Lemonade figures from its results release of July 29, 2026. Market capitalizations and share returns are accurate as of the time of writing. Hippo did not publish an LAE ratio in its release, so a direct comparison to Lemonade on that axis is not available. Data accurate as of the time of writing. Charts are shown in real time via TradingView. Nothing herein constitutes a forecast, recommendation or advice - see the full disclaimer at the bottom of the page.