Four Cohorts Make a Pattern: Upgrading Opendoor as the Margin Curve Goes Flat

Results vs. Consensus

Quality of Beat/Miss

• Revenue: High quality, and for the first time in three quarters the revenue line is not the story. The $720M print beat the ~$666M Street number by 8% and fell only 2.2% sequentially against a guide of down approximately 10%. Revenue remains depressed year over year, down 37.6% from $1,153M, because it is still being fed by the inventory drought of the second half of 2025. Homes sold of 1,921 against 2,474 purchased marks the first quarter in over a year in which the company bought more homes than it sold.
• Margins: This is the quarter's substance. Contribution margin of 4.4% against 1.0% is the largest sequential improvement in the recorded history of the metric, and it arrives with a mechanical explanation rather than a mix accident: the legacy book that produced the 1.0% print has been cleared, and aged inventory is now 10% of listings. Gross margin of 10.0% is the first double-digit print in the recent record, and adjusted gross margin of 9.3% against 6.1% confirms it is not an inventory-valuation artefact. The one caveat is that the 4.4% quarterly figure remains below the 5% to 7% target range, and Q2 is guided to the middle of it.
• EPS: The adjusted loss of $0.05 landed exactly on consensus. The GAAP loss of $0.18 is not comparable to that number and the reconciling item is now almost entirely stock-based compensation, which was $120M in Q1 against $14M in the year-ago quarter. General and administrative expense of $137M against $33M a year ago is essentially that one line. This gap is no longer an artefact of a financing event as it was in Q4; it is a recurring cost of the compensation structure and should be treated as such.

Key Operating Metrics

Opendoor reports a single operating segment. Read the table as a system and the quarter is coherent for the first time since we initiated: purchases exceed sales, inventory rebuilds, aged inventory collapses, unit margin recovers to within a point of the year-ago level on a third fewer homes sold, and the cost base holds flat. Contribution profit per home sold of $17K against $18K a year ago is the cleanest single comparison, because it strips out the volume decline entirely: the company is now earning nearly the same economics per transaction as it did before the reset, on a far healthier book.

Acquisitions and the contract funnel

Homes purchased of 2,474 rose 45% sequentially, the second consecutive quarter of meaningful growth, and signed contracts, the leading indicator, exceeded 5,000. March was the strongest single month for contracts since June 2022. Homes under contract to purchase at quarter-end reached 1,939, nearly triple the 710 at year-end and the first year-over-year increase in that line in the covered period.

"An acquisition contract will typically close about a month later. Q1's 2,474 purchases are mostly from late Q4, early Q1 contracts. Late Q1 contracts will close primarily in Q2. Also, we want to be clear, we don't close on every home we go into contract on." — Christy Schwartz, CFO

Assessment: The contract-to-close lag is the most useful piece of modelling guidance offered on the call, and it makes the Q2 revenue guide of approximately +25% sequential arithmetically well-supported rather than aspirational. The 1,939 homes under contract entering Q2 are a visible pipeline. The deliberate narrowing between contract and close is a discipline claim that cannot be verified from outside, but the cohort data is the indirect evidence for it.

Inventory health

Aged inventory fell to 10% of listings from 33% at year-end and 51% at the end of Q3, a 41-point improvement in two quarters, and the lowest level since Q2 2022. Management contrasted this against a broader market where the equivalent figure rose to 33%.

"Two quarters ago, more than half of our homes had been sitting on the market for over 120 days. At the end of Q1, that number was 10%. That is the lowest it's been since Q2 2022. To put it in perspective, the broader market was at 23% 2 quarters ago and rose to 33% at the end of Q1." — Christy Schwartz, CFO

Assessment: The relative comparison is the important half. Falling aged inventory while the market's own aged inventory rises cannot be explained by a friendlier housing backdrop, which is the standard bear rebuttal to this metric. Inventory dollars did rise 23% sequentially to $1,139M as the book rebuilds, so absolute balance-sheet exposure is growing again even as its quality improves.

Cash Now, More Later

The capital-light product, renamed from Cash Plus, accounted for more than a third of Q1 acquisition contracts against zero a year earlier, and management said the product tripled during the quarter. Under it the seller takes less cash upfront, retains more of the price risk, and pays lower fees.

"Cash Now, More Later, it's growing really fast. It's over 1/3 of our growing pie. And that's just really remarkable for a product that didn't exist a year ago. And that was totally reworked just 3 months ago. We're iterating how it works. We're fine-tuning it, trying to get the balance right between what the seller gets and what Opendoor keeps." — Kaz Nejatian, CEO

Assessment: A third of contract volume now carries materially less balance-sheet risk per transaction, which lowers the capital required per dollar of contribution profit. The candid admission that the economic split between seller and company is still being tuned is worth noting: as the mix shifts, reported contribution margin becomes a blend of two products with different economics, and the company still discloses no separate figures for either.

Key Topics & Management Commentary

Overall Management Tone: The most confident of the three calls we have covered, and the first where the argument was carried by a chart rather than by rhetoric. Management opened by replaying its own prior remarks for a second consecutive quarter, acknowledged the specific criticism that one month does not make a trend, and answered it with four. Where the posture stayed disciplined was in the explicit enumeration of what failure would look like, which is unusual and worth crediting. Where it drifted was into extended digressions on the history of money and tokenisation that added nothing to the investment case.

1. The cohort arrival curves

The analytical centrepiece of the quarter is a chart plotting cohort margin against percentage sold through. Legacy cohorts degrade steadily as they clear. The October through January cohorts run close to flat. The October cohort, now over 80% sold through, has given up only 90 basis points of core cash-product margin across that journey against more than 260 basis points for the comparable path a year earlier.

"We've now sold through over 80% of the October cohort and our trends have continued. Margins for our core cash products have come down only 90 basis points from where they were at 10% sold to over 80% sold. Last year, that same journey cost us over 260 basis points. So we've seen about a 3x improvement." — Kaz Nejatian, CEO

Assessment: This is the disclosure that changes our rating. In February the bear objection to the October cohort was sample size and the fact that difficult homes sell last. Both objections have now been tested: the cohort is over 80% cleared, which is deep into the difficult half, and three subsequent cohorts show the same shape. A flat arrival curve is the signature of pricing with genuine dispersion, and it is very hard to fake across four consecutive monthly vintages in a soft market.

2. Adjusted EBITDA profitability arrives

The company disclosed that as of April 1 it is adjusted EBITDA profitable on a forward twelve-month basis, delivering the milestone the CEO volunteered off-script in February. Q2 adjusted EBITDA is guided to breakeven plus or minus a few million dollars.

"As of April 1st, Opendoor is adjusted EBITDA profitable, on a 12-month go-forward basis. The October cohort was just the start. A full quarter later, we've gone from a claim to a track record." — Kaz Nejatian, CEO

Assessment: Delivered on time and one quarter after being promised without being formally guided. That is the second consecutive quarter in which this management team has met a commitment it made in the prior quarter, which is the pattern that justifies extending it some credit on the year-end adjusted net income target. The forward twelve-month framing does a lot of work here and should be read as a statement about the current run rate rather than about the trailing year, which remains firmly loss-making.

3. The strategic reframing: from price prediction to time

The clearest articulation yet of what actually changed at this company came in the prepared remarks, and it is a reframing of the question the pricing model is asked to answer.

"A year ago, the most important input into every decision was our home price appreciation forecast. Today, it's how fast we can sell the home we're looking to buy. Market makers do not win by being right about direction. They win by controlling their exposure to being wrong. They win by being right about time." — Kaz Nejatian, CEO

Management also described the failure loop of the prior regime with unusual specificity: wider spreads produced worse homes, worse homes produced worse margins, worse margins fed back into a more conservative model, which widened spreads further.

Assessment: The description of the doom loop is the most persuasive diagnostic this management has offered, because it explains the 51% aged-inventory figure as a systemic outcome rather than a run of bad luck. If the loop was the problem, breaking it is a durable fix. The claim that the model works across macro cycles remains untested against a genuine downturn.

4. Mortgage moves from beta to a real product

Opendoor Mortgage launched in Colorado during the quarter at a 4.99% rate. Management said attach rates are running above its most optimistic expectations, that rates are about 100 basis points below the market average, and that licensing is in flight in just over 20 states with a plan to roughly double that by the end of Q3.

"Big bank lenders take about 340 basis points in revenue per loan. Most of that is just a toil tax on the borrower ... We've built an AI-native mortgage platform from day 1. No legacy system, no commission-driven sales force ... Today, our rates are running about 100 basis points below the market average." — Kaz Nejatian, CEO

Assessment: The cost-structure argument for why an integrated lender can undercut incumbents by 100 basis points is sound in principle. Attach rates were described qualitatively as above expectations but never quantified, for the third consecutive quarter that a version of this question has been asked. We continue to model no ancillary contribution. The upside optionality is real and getting more concrete; the disclosure is not keeping pace with it.

5. Management named its own failure conditions

Unprompted, the CEO enumerated the three specific things that would indicate the strategy was not working: cohort curves reverting to the legacy shape, contracts plateauing at or below the low end of the published range for a sustained period, and aged inventory returning to Q4 levels.

"Cohort curves start looking like they did with the purple lines. They would start high, would have massive losses as we went through. Contracts would plateau at the low end of our range or below the low end of our range for a whole bunch of weeks and homes greater than 120 days in the market would go back to what we had in Q4. If those 3 things happen ... then we're not doing what we said we would do." — Kaz Nejatian, CEO

Assessment: Publishing falsification criteria is rare and analytically valuable, and we adopt all three as monitoring conditions. It also, deliberately or not, sets a high bar for being judged wrong, since the CEO framed it as all three occurring together. Any one of them deteriorating would be a meaningful warning; we will not wait for the conjunction.

6. Operating leverage without an AI spending problem

Fixed operating expenses fell to $33M from $35M, down $6M year over year, while trailing twelve-month operations expense held at 1.3% of revenue. The most striking comparison offered was historical: the last time quarterly acquisition contracts exceeded 5,000, fixed operating expenses were double the current level.

"There's a lot of noise right now about companies blowing their 2026 budgets on AI before the second quarter. That's not us. We're focused on results, not token leaderboards." — Christy Schwartz, CFO

The specific tools cited include an AI repair-negotiation tool that cut buyer fall-through by a double-digit percentage, AI scoping feedback reducing pre-list renovation spend by 10% to 20% per home in pilot markets, and ticket-triage automation redeploying three full-time employees. Management emphasised that most were built by operators rather than engineers.

Assessment: The half-the-fixed-cost-at-the-same-contract-volume comparison is the single best evidence for the operating-leverage pillar we have seen, because it holds volume constant and measures the cost base directly. The individual tool savings are small in aggregate; the point is the throughput of a non-engineering workforce shipping its own automation.

7. Capital: no equity raise to fund growth

Unrestricted cash rose to $999M, the highest balance in years, while inventory grew 23% sequentially. Asked directly whether accelerating acquisitions would require an equity raise, management gave the most explicit anti-dilution commitment of the coverage period.

"I despise dilution. If we issue a share, it has only one job to make every other share worth more for our existing shareholders. We will never issue shares to extend the runway. That's not what we're going to do. The goal is for Opendoor to never be in a position where it has to raise money to survive." — Kaz Nejatian, CEO

The CFO added that cash grew while inventory was acquired because younger homes with shorter days on market are structurally easier to finance, that warehouse capacity exceeds the acquisition plan, and that the outstanding warrants provide additional capital optionality.

Assessment: Growing the cash balance while growing inventory is the strongest possible refutation of the funding-risk version of the bear case, and it is what allows us to upgrade despite the share count. Note the careful construction of the commitment: no shares to extend runway or to survive. It says nothing about opportunistic issuance, and the CFO's own answer described future capital decisions as "opportunistic rather than necessary."

8. Seasonality gets a framework

For the first time, management laid out an explicit seasonal model: the selling season begins after the Super Bowl, peaks in early summer, tapers through autumn and bottoms in December, with acquisition cadence running inversely so that inventory is built through the fall for the spring selling season. The published contract roadmap has been reshaped accordingly.

"Macro changes the level of the curve and seasonality is the shape of the curve. ... Our acquisition cadence runs inversely to market resale activity. We acquire less in late spring when we'll be selling into weaker demand, and we build inventory throughout the fall in anticipation of the spring selling season." — Christy Schwartz, CFO

Assessment: This is a useful and previously missing piece of modelling infrastructure, and it pre-empts a misreading of softer summer acquisition numbers. It also functions as advance cover for a second-half acquisition path that was already rebalanced in February. Both things can be true, and the roadmap on the public dashboard is now the place to check which one is operating.

9. Product velocity and the Doma escrow acquisition

The quarter's product list included an expansion of Cash Now, More Later coverage, a rebuilt buyer app, the acquisition of Doma's escrow division, an AI underwriter pricing standard homes in Phoenix, a consolidated buyer messaging system replacing six, a rebuilt offer page, seller-led assessment scheduling now covering more than half of assessments at 6,000 in March alone, and a title intake process cut from up to five hours to fifteen minutes.

Assessment: The escrow acquisition is the most strategically significant item and received the least airtime. Owning closing infrastructure is a prerequisite for the integrated transaction the company keeps describing, and it is the first inorganic step in the story. No purchase price, revenue contribution or integration cost was disclosed anywhere in the release or the call.

10. What management says it has not figured out

A short section of the prepared remarks was given over to open questions, which is unusual enough to note. Mortgage performance across different market conditions and home-price tiers is untested. The economic split within Cash Now, More Later is still being tuned. And the housing market remains what it is.

"I've decided that the promise is not the same thing as a proof. You do not get credit for what should have happened. You only get credit what actually did. ... Proof will take more time, more reps, more shifts." — Kaz Nejatian, CEO

Assessment: Distinguishing evidence from proof, while sitting on the best quarter of the turnaround, is the correct posture and a meaningful contrast with the promotional register of the Q3 call. It does not change what we can underwrite, but it does change how much we discount what management says.

11. Tokenisation, again

A one-word shareholder question on tokenisation produced several minutes on the history of money, COBOL in banking infrastructure, and medieval English property records, concluding that title, mortgage and insurance will eventually move on chain, with the Doma escrow acquisition positioned as a step along that path.

Assessment: Our position is unchanged from November: zero investment content today. The difference this quarter is that there is now a concrete adjacent action, the escrow acquisition, that has a real business rationale independent of any blockchain thesis. Investors should value the escrow capability and ignore the framing.

Guidance & Outlook

Three consecutive quarters of guideposts have now been met or beaten. The Q2 framework is the most demanding yet, because for the first time it requires revenue growth and margin expansion simultaneously: approximately 25% sequential revenue growth to roughly $900M, contribution margin in the middle of the 5% to 7% range, and adjusted EBITDA at breakeven.

Implied Q2 shape: Revenue of approximately $900M at a 6% contribution margin implies roughly $54M of contribution profit, against $32M in Q1. With fixed operating expenses holding near $33M, that is the arithmetic that produces adjusted EBITDA breakeven. The 1,939 homes under contract at quarter-end, plus the roughly one-month contract-to-close lag, make the revenue half of that equation well-supported.

Street at: Consensus modelled roughly $666M for Q1 against the $720M print. Expect Q2 numbers to be marked toward the company's approximately +25% sequential framework, which puts the Street near $900M, and for the debate to shift from whether margin recovers to whether it holds at scale.

Guidance style: Conservative and now with three quarters of evidence behind that characterisation. Management has beaten its own revenue guidepost in each of the last two quarters by a wide margin and has hit every margin and EBITDA guidepost. The pattern is consistent enough to matter for modelling.

Analyst Q&A Highlights

The livestream format continues, with shareholder questions submitted through a brokerage platform alongside written sell-side questions read by investor relations. Still no live follow-ups.

Underwriting quality and funding as acquisitions accelerate

The most important question of the call combined the two live bear objections: whether faster buying means worse buying, and whether the growth requires an equity raise. Management answered both, and the funding half produced the most explicit anti-dilution language of the coverage period.

Q: "As you accelerate acquisition velocity, how are you ensuring that underwriting quality remains high and that you won't need to raise equity to fund this expansion?"
— Andrew L., shareholder question

A: "On the capital question, our cash position actually grew as we acquired more inventory, which reflects the underlying health of our inventory book. Younger homes with shorter days on market are structurally easier to finance, and we have sufficient warehouse capacity to more than keep up with our acquisition pace and plans."
— Christy Schwartz, CFO

Assessment: Cash up $37M sequentially while inventory rose $214M is the answer, and it is a good one. The underwriting half was answered by pointing at the cohorts rather than at process, which is the right hierarchy of evidence. The caveat is the CFO's own phrasing that future capital decisions would be "opportunistic rather than necessary," which is a narrower commitment than the CEO's rhetoric implies.

Confidence in the Q2 EBITDA inflection and the year-end target

A cluster of shareholder questions asked whether the profitability commitments still held with rates still elevated, and what the swing factors were. The answer reconfirmed both milestones and pointed to the three management objectives as the monitoring framework.

Q: "How confident are you today in the Q2 positive EBITDA and year-end profitability forecast? And what are the key swing factors from here, which might influence it?"
— Multiple shareholder questions incl. Arun Jacob V., Heejun C.

A: "The last time acquisition contracts exceeded 5,000 in a quarter, our fixed OpEx was double where it is right now, yes, double. ... We have made meaningful changes to what is required to run Opendoor 2.0, and we are beginning to demonstrate that those changes are durable as the volumes return."
— Christy Schwartz, CFO

Assessment: The volume-constant cost comparison is the most efficient answer available to the profitability question, because it converts an argument about the future into an observation about the present. Notably absent from the swing-factor discussion was any mention of stock-based compensation, which is the largest single reason adjusted and GAAP results diverge.

What is actually driving conversion and volume

A two-part sell-side question asked which specific change is moving seller conversion, and where the opportunity lies to scale homes per market without scaling operating expenses proportionately. The answer credited the shift from a single take-it-or-leave-it offer to a configurable one.

Q: "As you reflect on the moves you've made, which specific change do you believe is having the most measurable impact on seller conversion rates and acquisition volumes today? ... Looking ahead, where do you see the biggest opportunity to structurally drive more homes purchased per market without proportionately scaling OpEx?"
— Dae Lee, JPMorgan

A: "I actually insisted on launching every state in the Lower 48 because I want to test this hypothesis. It turned out that if you have a good underwriting model, a good product and a good partner network, you can buy homes anywhere, like we closed a home in South Dakota this week, and we have 0 employees in South Dakota."
— Kaz Nejatian, CEO

Assessment: The zero-employee South Dakota close is a small anecdote carrying a large structural claim: that geographic expansion is now marginal-cost-light. If that holds, the operating leverage available from here is considerably greater than the fixed-cost line alone suggests, because market count can grow without headcount. It is one home, and it is the right thing to watch.

Mortgage expansion and the rate advantage

A shareholder question about whether the Colorado mortgage rate was promotional produced a correction, a state-count disclosure and the underlying cost-structure argument.

Q: "I'm interested in your 4.99% mortgage promotion currently exclusive to Colorado. Are there plans to expand this offer to other regions or states soon?"
— Heejun C., shareholder question

A: "First of all, it wasn't a promotion. I want to be clear about that. That was the actual rate. We don't run rate connect here. We charge what the math allows us to charge ... we're in flight on licensing in about just over 20 states right now, and we expect to kind of roughly double that by the end of Q3."
— Kaz Nejatian, CEO

Assessment: The licensing figures are the first concrete timeline attached to the ancillary opportunity, and roughly 40 states by the end of Q3 would make mortgage a national product within two quarters. Attach rate remains undisclosed for the third consecutive quarter despite being described as above expectations, which is the gap between a story and a modelable line item.

Seasonality through the balance of the year

A sell-side question on seasonality produced the company's first explicit framework for it, and a reshaped public contract roadmap.

Q: "Andrew from Citizens is curious to help us understand a little bit more about seasonality kind of through the balance of the year."
— Andrew, Citizens (question read by Investor Relations)

A: "Seasonality is just like gravity. It's like a rule of nature. You don't blame gravity and if you try to fight it, you tend to lose. ... While we can't flatten the curve entirely, we can collapse the impact over time, and that's what we're working on."
— Kaz Nejatian, CEO

Assessment: Establishing the seasonal shape before it produces a softer quarter is good practice and materially improves the modelling picture. It also lowers the information content of the next two quarters' acquisition numbers, since any shortfall now has a pre-approved explanation. The published roadmap ranges are the discipline that keeps that honest.

AI and the long-run structure of the industry

A shareholder question on whether the company is positioned for what AI does to real estate produced a long structural answer, of which the operationally relevant part concerned complexity as a defence.

Q: "Do you believe when you look at the strategic direction of the company that we are well prepared for all the things that AI is likely to change about the way the real estate market operates in the coming years?"
— Mike Alfred, Alpine Fox LP

A: "Our goal isn't to use AI to cut 15% of our expenses by doing the same things we're doing just cheaper ... what we want to do, given everything AI can do is to rebuild our processes from scratch, from a blank piece of paper so that we can use AI to have a fundamentally different process."
— Kaz Nejatian, CEO

Assessment: The distinction between automating an existing process and redesigning it is the correct one and is corroborated by the fixed-cost-at-constant-volume comparison. The argument that transaction complexity protects an incumbent operator is plausible but self-serving, and the same complexity is what makes the business capital-intensive in the first place.

What They're NOT Saying

1. Stock-based compensation, still unmentioned: $120M in Q1 against $14M a year ago, driving general and administrative expense to $137M from $33M. It was not raised once on the call, in prepared remarks or in Q&A, and it is the entire difference between a $49M adjusted loss and a $173M GAAP loss. A profitability target defined on a measure that excludes it deserves to be discussed alongside it.
2. Mortgage attach rate: Described as "above even my most optimistic expectations" for the second consecutive quarter without a number. Until it is quantified, the ancillary pillar cannot move from optionality to model input.
3. Doma escrow acquisition terms: No purchase price, no revenue contribution, no integration cost, no mention in the release's financial tables discussion. The first acquisition of the new regime was disclosed in a product list.
4. Cash Now, More Later economics: Now more than a third of contracts, with management openly still tuning the split between seller and company, and no fee, take rate or per-transaction contribution disclosed. As mix shifts, blended contribution margin becomes progressively harder to interpret.
5. The spread: Still undisclosed in any period, four quarters into a strategy whose entire premise is spread discipline.
6. Contract-to-close conversion: Management acknowledged that not every contract closes and that the funnel narrows deliberately, but gave no conversion rate. With contracts of 5,000 and closes of 2,474, the implied relationship matters a great deal for translating the dashboard into forecasts, and the lag makes the naive ratio misleading.
7. Warrant dilution: Referenced approvingly as "capital optionality" without any statement of how many shares the $9, $13 and $17 warrants represent or where they sit against the current price.

Market Reaction

• Pre-print setup: The shares closed at $5.32 on May 7, down 8.7% year to date against the S&P 500's +7.2%, but up 16.9% over the preceding thirty days from $4.55 on April 7 and up 510.1% over twelve months. The 52-week closing range entering the print was $0.51 to $10.52.
• Next-day session: The shares opened at $5.35 on May 8, a 0.6% gap up, traded between $4.78 and $5.50, and closed at $5.01, down 5.8% or $0.31. Volume was 53.8 million shares against a thirty-day average of 33.3 million, or 1.6 times normal. The S&P 500 closed up 0.8%.

A 5.8% decline after the best operating quarter of the turnaround is the most interesting fact in this section. It is not a rejection of the print, which beat on revenue and delivered on every guidepost. The session opened marginally higher and sold off from there, which reads as position-trimming into strength following a 16.9% run over the preceding month, compounded by a headline GAAP loss per share of $0.18 against a consensus struck at $0.05 on the adjusted line. Any screen or automated summary comparing those two numbers records a large miss.

The volume pattern is less dramatic than the two prior prints, which traded at 1.5 and 2.3 times normal as the market re-rated the story in both directions. At 1.6 times normal with a modest opening gap and a steady fade, this looks like holders adjusting rather than the shareholder base changing hands.

The setup matters for the rating. Entering the print the stock had recovered 16.9% from its early-April level but remained down 8.7% year to date against a market up 7.2%, and it closed the session 52% below the 52-week closing high of $10.52. A quarter that delivered the margin recovery, the four-cohort pattern and the EBITDA inflection, met by a 5.8% decline, is the definition of an improving fundamental case meeting a de-rating multiple. That combination is what an upgrade is for.

Street Perspective

Debate: Does the margin recovery survive scale?

Bull view: The bull case on the Street is that the four-cohort arrival curves settle the pricing-quality question, that Q1's 4.4% was achieved while acquisitions grew 45%, and that Q2's guide to the middle of the 5% to 7% range with revenue up approximately 25% demonstrates margin and volume are no longer a trade-off.

Bear view: The bear camp contends that the easy margin recovery came from clearing a distressed book, that 4.4% is still below the target range, and that the true test comes when the company is buying at 6,000 homes a quarter into a seasonally weaker second half rather than at 2,474 in the spring.

Our take: The bulls have this quarter and the bears have the right test. What tips the balance is the arrival-curve shape rather than the level: a flat curve means the marginal home is being priced correctly, which is the property that should persist as volume grows. We would revisit if Q3 shows margin holding only because acquisitions decelerate.

Debate: Is adjusted net income a legitimate target given the compensation cost?

Bull view: Some desks argue that stock-based compensation reflects a one-time reconstitution of the entire executive team and a founder-mode incentive structure, that it is non-cash and therefore irrelevant to the going-concern question, and that adjusted EBITDA turning positive on a forward basis is the operational fact that matters.

Bear view: Skeptics observe that $120M of quarterly stock-based compensation against a company targeting adjusted net income breakeven means the target can be hit while per-share value is being transferred from existing holders to employees, and that management has now declined to discuss the line for three consecutive quarters.

Our take: We continue to side with the bears on the principle and have moderated on the magnitude. Stock-based compensation is a real cost and should be in any honest profitability measure. But the Q1 disclosure that cash grew while inventory grew removes the funding-driven dilution risk, which was the more damaging version of this concern. What remains is compensation dilution, which is expensive but bounded and, at the current share price, substantially priced in.

Debate: Is the second-half acquisition ramp achievable?

Bull view: A growing consensus view is that with contracts above 5,000 in Q1, 1,939 homes under contract entering Q2, and geographic coverage across the Lower 48 at near-zero marginal cost, the path from 2,474 quarterly purchases toward the 6,000 exit rate is now a throughput problem rather than a demand problem.

Bear view: The bear camp notes that the ramp has been rebalanced toward the second half once already, that the newly articulated seasonality framework conveniently predicts softer acquisitions through spring and summer, and that the exit-rate target has not moved despite two quarters of first-half moderation.

Our take: We model roughly 5,000 quarterly purchases exiting 2026 rather than management's 6,000, and we do not think the difference changes the investment case. At 5,000 homes and a 5.5% contribution margin the company clears its adjusted net income target with room. The exit rate is the headline; the margin is the thesis.

Model Update Needed

Valuation impact: At $5.01 on 959.3 million weighted-average shares the equity is capitalised at roughly $4.8B. Trailing multiples remain uninformative because the trailing year contains the trough: trailing twelve-month contribution profit is $128M, which puts the stock at roughly 38 times a number nobody should anchor on. The forward arithmetic is the relevant frame. Q2 guided revenue of approximately $900M at a 6% contribution margin annualises to roughly $216M of contribution profit, or about 22 times, against a fixed cost base of roughly $132M annually and an adjusted EBITDA line crossing into positive territory. On our exit-2026 assumptions of roughly 5,000 quarterly purchases at a 5.5% to 6.0% margin, the multiple compresses further without any multiple expansion. Our fair-value range moves to $6.00 to $8.00 from $4.50 to $6.50, which sits 20% to 60% above the current price.

Thesis Scorecard Post-Earnings

Overall: Thesis strengthened materially. The central margin pillar has moved from a single proof point to a four-cohort pattern tested deep into sell-through, the housing-risk bear point is now clearly challenged, and the funding-driven half of the dilution objection has been answered by a rising cash balance. What remains is compensation dilution, which is expensive and disclosed only in the filings.

Action: Upgrade to Outperform. The February upgrade trigger was a quarterly contribution margin at or above 4% with acquisitions above 2,300 homes; the quarter delivered 4.4% and 2,474. We would move back to Hold if any one of management's three self-declared failure conditions appears, if Q2 contribution margin lands below 5%, or if the company issues equity for anything other than an acquisition it can defend on returns.