The Pegasus Math: Opendoor's AI-Led Turnaround Claims Profitability Is Now Just Arithmetic
A re-rated but heavily sold-off stock bets that mortgage attach, AI leverage, and 5–7% contribution margins turn the breakeven math positive by year-end.
OPEN · Earnings Call · 2026-08-04
The "It's Just Math" Turnaround
Ten months into his turnaround, CEO Kaz Nejatian skipped the customary video clip and told a personal story: his wife told him not to come home until there was a plan to break even. The punchline, delivered with a wink, is that he is finally coming home. “It's just math, sweetheart, after this call, I am coming home.” — Kasra Nejatian, CEO · 2026-08-04 The math he presents is a four-variable framework — volume, contribution margin, OpEx, and financing costs — built around a acquisition GMV target of roughly 6,000 quarterly transactions at $375,000 per home, or ~$9B in revenue. His central claim: nothing needs to change for the model to cross into ANI profitability. Weekly contracts are already running above 500, contribution margin sits at 5.8% (the middle of the 5–7% guide), and the cost structure now runs around 2.4–2.9% of acquisition GMV versus the 3–4% he previously flagged as the breakeven band. The confidence rests on two company-unique data points. First, conversion without spread-buying: true seller conversion at the same spread levels as the pandemic-era peak has roughly doubled, directly rebutting the "are we buying growth through lower spreads?" skepticism. Second, AI leverage that broke the cost curve: home project managers now carry 10 renovations a month (up from 3, heading to 20), underwriters handle 50 valuations a day (up from 20, heading to 100), and under 20% of homes require an in-person visit (down from 80%). This is not aspirational boilerplate — CFO Christy Schwartz quantified it: “we purchased 4,378 homes and spent $14 million, 2.5x the volume on less variable cost.” — Christy Schwartz, Chief Financial Officer · 2026-08-04 The Variable operations expense per acquisition fell from $5,000 in Q1 to $3,000, and marketing collapsed from $19M to $5M quarter-over-quarter. The prior-call scaffolding is intact — Christy's seasonality framework, “Macro changes the level of the curve and seasonality is the shape of the curve,” — Christy Schwartz, CFO · 2026-05-08 underpins the claim that Q2→Q3 margin compression will be "way, way, way less" than the historical ~500 basis points, and that Q4 will beat Q3 for the first time in company history.Mortgage as the "Second Wing"
The bigger narrative shift is the mortgage. Kaz frames it through the "good Sears / bad Sears" parable — Discover grew born-inside-Sears financial infrastructure; Dean Witter failed as a bolted-on department-store booth. The current product is intended to be the former, and the early penetration numbers are the concrete evidence: over half of scheduled closes in Colorado, the first launch market, are financed through Opendoor Home Loans, and Texas is at "nearly 1 in 5" just six weeks after launch. The structural claim: “The legacy mortgage industry carries 65 to 85 basis points of yield in every single loan just to feed the pork barrel buffet of people taking margin.” — Kasra Nejatian, CEO · 2026-08-04 By stripping that out and passing it to the buyer as a lower rate, the mortgage becomes a velocity lever, not adjacent fee income — the "second wing" that turns the drag of a 7% mortgage rate into lift. This builds on, rather than replaces, prior framing. Last quarter Kaz said “We're not just discounting our way to a lower rate. We're actually building our way towards this.” — Kasra Nejatian, CEO · 2026-05-08 Today the mortgage product is live with real attach data, alongside graded progress on title/escrow (a self-awarded "B+") and insurance (an "incomplete" — postponed to focus on mortgage). The good Sears / bad Sears language is fresh and company-unique, a memorable hook for what is otherwise a standard attach-rate story.The Tape Doesn't Believe the Math Yet
Here is the tension. Management says the math is locked in, but the tape disagrees. The stock is down 36.7% from its late-April peak and has fallen 18.3% over the past 90 days. On a full-history basis, OPEN sits 90.2% below its February 2021 high. That is a market that has not yet paid up for the "pegasus" thesis — Kaz explicitly redefined the company from "unicorn" to a horse growing wings in public, acknowledging the awkwardness mid-transformation. The fundamentals confirm the turn but not yet the destination. Q1 2026 (the latest filed quarter, period ending April 30) showed revenue of $720M, down 38% YoY — the Q2 $883M print is a 23% QoQ bounce but still a fraction of the $1.5B quarters of mid-2025. The re-rating, though, is real: Price to Revenue of 1.1x, up ~690% YoY. Gross margin at 10.0% (up 1.4pp YoY) supports the contribution margin story, and Liabilities to Assets at 59.4% (down ~21pp YoY) shows the deleveraging that the "Cash Now More Later" (Step 2) product enables — earning Effective Net Cash of -$71M, versus -$1.4B a year ago. Whether the market is being perverse or prescient depends on the Q3 numbers: the guide is for revenue up at least 20% YoY, contribution profit dollars to more than double, and margin around 4–4.5%. If the "just math" holds, the wings have grown. If not, the horse was just running awkwardly.As things stand right now, Opendoor will become ANI profitable. It's just math, sweetheart, after this call, I am coming home.