Opendoor Technologies (OPEN) · Deep analysis

Report previewOPEN.USReport date Last updated
← Back to reports
Report contents
01

Business model

Business model: a thin spread on home price certainty

Opendoor buys houses directly from owners and resells them. The seller pays a service fee, which the company has generally set around 5% of the home price, and picks a closing date; Opendoor pays cash, does light renovation, and lists the property, typically selling within three to four months. The paying customer is a US homeowner with a deadline, usually a job relocation, divorce, estate settlement, or downsizing, who trades the last few thousand dollars of achievable price for speed and certainty. Buyers are ordinary owner-occupants and small investors. There is one reportable segment, essentially all revenue is US, and the metros served have shrunk and regrown with the company's appetite for risk.

This is a spread business dressed as a retail business. Revenue is gross home sales, so a single transaction books several hundred thousand dollars, while gross profit is what remains after the purchase price, renovation, holding costs, and the buyer-agent commission Opendoor pays on resale. That spread has run in the low single digits to low double digits as a percentage of revenue depending on market conditions. Every dollar of revenue needs roughly a dollar of inventory, financed through secured facilities, so growth consumes capital months before it produces profit. Opendoor reports residential real estate revenue separately from a small "other revenues" line covering title, escrow, and mortgage, but it does not break out those attachments by product. It also does not publish acquisition cost per home or cohort-level spreads, which are the numbers that would show whether added volume creates value or just consumes it.

Bargaining power is bounded on both sides. The fee is capped by the alternative, a traditional sale at roughly 5–6% in total commission, and by investors who will bid for the same house. On the resale, Opendoor still pays a buyer-agent commission, so a structural slice of the spread leaves before the company sees it. The offer price is where competition actually happens: Opendoor must guess what a house will clear for in three to four months, and it carries the error. The early advantage was a pricing model fed by its own transaction data and the willingness to commit capital at speed. Neither is durable on its own. Two better-funded rivals built the same capability and then quit, Zillow winding down its offers business in 2021 and Redfin exiting in 2022. That exodus reduced competitive intensity, but it is also evidence about the economics, not a validation of them.

The fragility is concentrated in the hold period. A portfolio bought near a peak and sold into falling prices produces losses on inventory already owned, and the secured facilities are marked against home values, so borrowing capacity can contract exactly when the company needs it. That sequence played out after 2022: Opendoor stopped buying, took inventory write-downs, and revenue settled far below the peak above $15 billion it reached that year. Current volume remains a fraction of that level, which means the fixed cost base of technology, marketing, and central operations is being spread over too few transactions to demonstrate operating leverage. The business type is a spread-taking market maker with retailer-like inventory and lender-like balance-sheet exposure; on the evidence, its stage is a turnaround whose unit economics are still unproven at anything like national scale.

Growth has three routes, and they are not equivalent. Deeper penetration in existing metros adds homes but adds inventory and market risk in the same proportion. Attaching title, escrow, and mortgage to each transaction adds fee income that uses no balance sheet, though the attach rates are not disclosed and the mortgage operation depends on the same buyer demand as the resale. The third route is the marketplace, where third-party investors and partner agents take the other side of a listing so Opendoor earns a fee without owning the home. Revenue per transaction there is much smaller, but so is the capital, and the question is whether the fee volume can eventually cover the overhead that the balance-sheet model was supposed to absorb.

What would overturn this assessment is specific: two or three consecutive years in which gross profit per home, after renovation and holding costs, stays positive through a flat-to-declining price market, with overhead covered by operating profit instead of inventory gains or one-off cost cuts. A sustained rise in homes sold without a matching rise in acquisition marketing and holding days would be the supporting evidence. A widening gap between homes bought and homes sold, or inventory days lengthening into a soft selling season, would say the opposite.

Questions worth following

  • Can Opendoor earn a positive spread on homes through a declining price market?

    The spread after renovation, holding costs, and buyer-agent commission is the entire economic engine; everything else is scale on top of it. If the spread stays positive when home prices are flat or falling, the model can be financed and grown. If it depends on rising prices, the business is a leveraged bet on housing, not an operator.

  • Can Opendoor hold its seller fee without losing transaction volume?

    The fee is the clearest read on pricing power against the roughly 5-6% cost of a traditional sale. Stable fees alongside stable volume would show sellers still pay for certainty; rising fees matched by falling contracts would show the tradeoff is real and that demand is elastic.

  • Can Opendoor's marketplace grow without owning the inventory?

    An asset-light channel could lift fee income per transaction without tying up capital or carrying price risk, which changes how much growth the balance sheet can support. Whether partners actually transact at scale determines if this is a second engine or a marketing feature.

02

Financial quality

Financial quality: cash moves with the housing book, not with earnings

Opendoor's financial quality is weak, and the weakness is structural. Operating income was negative in each of the five fiscal years from 2021 through 2025, and operating cash flow has tracked the direction of the home inventory almost exactly: positive when the housing book shrinks, negative when it grows. In the first nine months of FY2026 the company rebuilt the book, adding $1.64 billion of inventory, and funded most of it with short-term borrowings while gross margin fell to 3.72%. Cash and returns now hinge on how fast those houses resell and at what spread.

USD bn unless notedFY2021FY2022FY2023FY2024FY2025Q3 FY2026Assessment
Revenue8.0215.576.955.154.371.02Very weak
Gross margin (%)9.104.287.018.408.013.72Weak
Net margin (%)-8.25-8.69-3.96-7.61-29.74-21.33Very weak
ROE (%)-34.83-81.16-26.79-46.67-151.34-28.95Very weak
ROA (%)-11.33-16.79-5.41-11.71-46.99-6.51Very weak
Liabilities/assets (%)74.3481.1470.5173.7058.1184.21Weak
Current ratio2.124.3049.215.677.031.68Adequate
Quick ratio0.741.3323.861.584.200.49Weak
Inventory turns2.222.822.072.402.610.45Adequate
Asset turnover1.371.931.371.541.580.31Strong
Operating cash flow-5.790.732.34-0.591.05-0.83Weak
Free cash flow-5.830.692.31-0.621.04-0.84Weak

*Quarter ended September 30, 2026. Flow and turnover items are single-quarter and not comparable with annual columns. Leverage for FY2025 and Q3 FY2026 is computed from reported assets and liabilities.

Returns have never cleared the cost of capital

The best year in the table, FY2023, still produced a -3.96% net margin and a -26.79% ROE. FY2025 was the worst at -29.74% and -151.34%. Gross margin has ranged from 4.28% to 9.10% since FY2021, thin for a business that takes title to houses and carries their price risk between purchase and sale. Asset turnover of 1.4x to 1.9x is high for a balance sheet this heavy, so the model does move its capital; each turn simply earns less than a tenth of revenue in gross profit, which interest and overhead then consume. Total equity fell from $2.25 billion at the end of FY2021 to $0.59 billion at September 30, 2026. These are operating losses absorbing the shareholder base, not a small-denominator artifact.

Operating cash flow is the inventory line by another name

Cash generation here is a working-capital release, which makes it a poor quality signal. FY2023 produced $2.34 billion of operating cash flow in the same year inventory fell $2.69 billion, and FY2025 produced $1.05 billion as inventory fell $1.23 billion. FY2024, when inventory rose $0.39 billion, produced -$0.59 billion. The nine months to September 2026 added $1.64 billion of homes and burned $1.80 billion of operating cash. Capex runs around $10 million a year, so free cash flow barely differs from operating cash flow; the real investment is houses. Positive free cash flow in three of five years sounds sturdier than it is, because every positive year came from liquidation. Receivables are a rounding error ($10 million to $20 million) since buyers pay cash at closing, so the entire working-capital exposure sits in inventory.

USD bn unless notedQ1 2025Q2 2025Q3 2025Q4 2025Q1 2026Q2 2026Q3 2026AssessmentTrend
Revenue1.151.570.920.740.720.881.02WeakImproving
Gross margin (%)8.598.177.216.2510.009.743.72Very weakDeteriorating
Net income-0.08-0.03-0.09-1.10-0.17-0.16-0.22Very weakDeteriorating
Total assets3.282.912.702.412.352.963.74WeakDeteriorating
Total liabilities2.632.281.891.401.402.053.15WeakDeteriorating
Inventory2.361.531.050.931.141.842.57WeakDeteriorating
Operating cash flow-0.280.820.430.07-0.25-0.72-0.83WeakDeteriorating
Current ratio2.994.352.837.037.072.941.68AdequateDeteriorating
Quick ratio0.751.971.684.203.481.040.49WeakDeteriorating

Net PP&E is not disclosed for these periods; it was $0.10 billion at the end of FY2022, so fixed assets are immaterial to the analysis and homes are the operating asset.

The rebuild leans on short-term debt

Inventory rose from $0.93 billion at December 31, 2025 to $2.57 billion at September 30, 2026, liabilities from $1.40 billion to $3.15 billion, and short-term debt from zero to $2.02 billion against $0.91 billion of cash. Current ratios across the window swing mostly with debt classification, so the quick ratio is the cleaner liquidity read: 0.49, because 71% of current assets are unsold houses. Interest has become the binding constraint. In the September quarter, gross profit of roughly $38 million was smaller than $40 million of interest expense. Revenue also rose year over year for the first time in the window, from $0.92 billion to $1.02 billion, so volume is no longer the problem. Spread is.

Three entries need explaining before the loss is read as operating

In the December 2025 quarter, operating loss was $0.15 billion but EBIT was -$1.07 billion and the net loss $1.10 billion, leaving roughly $0.9 billion of non-operating charge that the income statement does not identify. Separately, total equity rose from $0.71 billion to $1.01 billion across FY2025 despite a $1.30 billion net loss and negative financing cash flow, implying a large non-cash equity movement. Both need the equity and debt roll-forwards in the 10-K before the loss is treated as purely operating. Current liabilities also fell from $0.91 billion to $0.33 billion in that quarter while long-term debt edged up, which points to reclassification as much as repayment, and flatters the FY2025 leverage figure.

The third flag is margin. Gross margin fell 6 percentage points in one quarter, from 9.74% to 3.72%, while inventory grew 40%. That is consistent with price cuts on older homes to keep them moving, though mix could also explain it; the evidence so far does not settle which. A change in the assessment needs two things together: gross margin back near the 8% to 10% range on the larger book, and operating cash flow that turns positive without the housing stock shrinking. If September's spread holds on a growing portfolio, the rebuild is destroying value at scale.

Generated by DeepVal AI. Data and models may contain errors or be outdated. For research only; not personalized investment advice.

Go beyond the preview

Continue reading the full report

  1. Key risksWhat could break the investment case
  2. Valuation and margin of safetyWhat the current price assumes, and how much room for error it leaves
  3. SummaryOur overall judgement and what to watch next

Free to sign up · 20 research points · unlocking this report uses 10

20 points cover 2 full reports. No card needed.

Research reflects its generation date. Preview the first two modules.

More on Opendoor Technologies Inc