Opendoor Technologies Inc. (OPEN) 2026 Q1 法說會逐字稿

(video playing) Hey, everyone. Welcome to Opendoor's Q1 2026 financial open house earnings livestream. I'm Michael Judd, Opendoor's Head of Investor Relations.

(影片播放中) 各位好。歡迎收看 Opendoor 2026 年第一季財務開放日（Financial Open House）財報直播。我是 Michael Judd，Opendoor 投資人關係主管。

A few quick housekeeping items before we get started, like all things Opendoor, we're going to do this faster. Details of our results and additional management commentary are available in our earnings release, which can be found at investor.opendoor.com. The following discussion contains forward-looking statements within the meaning of the federal securities laws. All statements other than statements of historical fact are statements that could be deemed forward-looking, including, but not limited to, statements regarding Opendoor's financial condition, anticipated financial performance, business strategy and plans, market opportunity and expansion, and management objectives for future operations.

在開始之前先快速說明幾項事項；就像 Opendoor 的一切一樣，我們會更快完成。我們的業績細節與管理層補充評論已發布於財報新聞稿，可至 investor.opendoor.com 查閱。以下討論包含符合聯邦證券法定義的前瞻性陳述。除歷史事實陳述外，所有陳述均可能被視為前瞻性陳述，包括但不限於關於 Opendoor 的財務狀況、預期財務表現、業務策略與計畫、市場機會與擴張，以及管理層對未來營運的目標等陳述。

These statements are neither promises nor guarantees, undue reliance should not be placed on them. Such forward-looking statements involve risks and uncertainties that may cause actual results to differ materially from those discussed here. Additional information that could cause actual results to differ from forward-looking statements can be found in the Risk Factors section of Opendoor's most recent annual report on Form 10-K for the year ended December 31, 2025, as updated by our quarterly report on Form 10-Q for the quarter ended March 31, 2026, and other filings with the SEC.

這些陳述既非承諾亦非保證，不應過度依賴。此類前瞻性陳述涉及風險與不確定性，可能導致實際結果與此處討論內容存在重大差異。可能導致實際結果與前瞻性陳述不同的其他資訊，請參閱 Opendoor 最近一期截至 2025 年 12 月 31 日止年度的 Form 10-K 年報之「風險因素」章節，以及我們截至 2026 年 3 月 31 日止季度的 Form 10-Q 季報更新內容，與其他向 SEC 提交的文件。

Any forward-looking statements made on this webcast, including responses to your questions, are based on management's reasonable current expectations and assumptions as of today, and Opendoor assumes no obligation to update or revise them, whether as a result of new information, future events, or otherwise, except as required by law.

本次網路直播中所作的任何前瞻性陳述（包括對各位提問的回覆），均基於管理層截至今日合理的現行預期與假設；除法律要求外，Opendoor 不承擔因新資訊、未來事件或其他原因而更新或修訂該等陳述的任何義務。

The following discussion contains references to certain non-GAAP financial measures. The company believes these non-GAAP financial measures are useful to investors as supplemental operational measurements to evaluate the company's financial performance. For a reconciliation of each of these non-GAAP financial measures to the most directly comparable GAAP metric, please see our website at investor.opendoor.com.

以下討論將提及若干非 GAAP 財務衡量指標。公司認為，這些非 GAAP 財務衡量指標可作為補充性的營運衡量，用以協助投資人評估公司的財務表現。如需將各項非 GAAP 財務衡量指標與最直接可比的 GAAP 指標進行調節（reconciliation），請至 investor.opendoor.com 查閱。

With that, let's get into the open house with Kaz and Christy.

接下來，讓我們進入由 Kaz 與 Christy 主持的開放日內容。

Good afternoon, everyone. I opened the Q4 financial open house by showing you a clip from the Q3 financial open house. I did this because I think among the most important things you can do to build trust is to just do what you said you would do; don't promise the moon and deliver dust. Just do what you said you would do.

各位下午好。我在第四季財務開放日一開始，先給大家看了第三季財務開放日的一段影片。我這麼做是因為，我認為建立信任最重要的事情之一，就是說到做到；不要許下登月承諾，最後只交出塵土。就做你說你會做的事。

Our last open house might as well have been called the look at the October cohort open house. With that in mind, let's once again take you back to our last financial open house.

我們上一次的開放日，幾乎可以直接叫做「看看 10 月批次（October cohort）開放日」。基於這點，讓我們再次帶大家回到上一次的財務開放日。

(video starts)

(影片開始)

That the October cohort is going so well is not a plan, it's a proof point. The product launches I'm going to talk to you about aren't promises of things that might work. They're the explanation for why October happened and why it's repeatable.

10 月批次進展得如此順利，這不是計畫，而是驗證點（proof point）。我接下來要談的產品上線，不是對「可能會成功」的事情做出承諾。它們是在解釋為什麼 10 月會發生，以及為什麼它可以被重複。

Now, look, because we're committed to transparency, let me get ahead of a couple of things. October was not our largest cohort by volume, but it was about double the size of what we were doing just a few months ago. Given how the past few weeks have gone, I believe we're on track to significantly increase our acquisition size, as we said we would do.

現在，因為我們致力於透明化，我先把幾件事說在前面。10 月並不是我們以量計最大的批次，但規模大約是幾個月前我們在做的兩倍。依照過去幾週的進展，我相信我們正如我們所說，走在大幅提升收購規模的軌道上。

What October shows is that the structural changes we made under Opendoor 2.0 are working. Then we're compounding those learnings into every single cohort going forward.

10 月所顯示的是：我們在 Opendoor 2.0 之下所做的結構性變革正在奏效。而且我們會把這些學習持續疊加到未來每一個批次中。

(video ends)

(影片結束)

During that call, I told you that Opendoor 2.0 cohorts would perform fundamentally differently than Opendoor 1.0.

在那通電話會議中，我告訴各位，Opendoor 2.0 的批次表現，將會與 Opendoor 1.0 有本質上的不同。

Back then, some folks said October was a fluke, or that we'd fall apart when the markets got harder or the sample set got larger. One of my favorite investors said, look, Kaz, one month does not make a trend. That was fair. Fair enough.

當時，有些人說 10 月只是偶然，或是當市場變得更艱難、或樣本規模變大時，我們就會撐不住。我很喜歡的一位投資人說：Kaz，你看，一個月不構成趨勢。這很公平。確實如此。

So here are the facts. We now have a few more months of data, and we should compare the first full four months of Opendoor 2.0 against the last couple of years of Opendoor 1.0. Don't pay attention to me. Look at the chart. These are the cohort arrival curves. They show what happens when a group of homes margins on the Y-axis, as that group of homes sells on the X-axis. Every one of those purple lines is an old Opendoor cohort. They all do the same thing. They bleed margin as we sell through.

所以，事實如下。我們現在又多了幾個月的數據，應該把 Opendoor 2.0 的前四個完整月份，與 Opendoor 1.0 過去幾年的表現做比較。不要聽我說。看圖表。這些是批次到達曲線（cohort arrival curves）。它們顯示一組房屋在 X 軸（隨著該組房屋售出進度）推進時，Y 軸上的毛利（margin）會發生什麼變化。每一條紫色線都是舊的 Opendoor 批次。它們都呈現同樣的情況。我們在賣出過程中會不斷流失毛利。

Now, look at the blue curves. Margin doesn't drop the way it used to. This is a step function change in how this company operates. Four consecutive months tell us something October alone could not. This isn't an accident. This isn't small sample luck.

現在，看藍色曲線。毛利不再像過去那樣下滑。這是公司營運方式的階躍式（step function）變化。連續四個月告訴我們一些僅靠 10 月單月無法證明的事。這不是意外。這不是小樣本的運氣。

Mortgage rates are still far too high, the listings are at all-time highs. In a housing market that was supposed to break us, our cohorts are delivering. October wasn't a fluke; it was just the first month we could see it. We've now sold through over 80% of the October cohort, 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.

房貸利率仍然高得離譜，掛牌量也處於歷史新高。在一個原本應該把我們擊垮的房市裡，我們的批次正在交出成果。10 月不是偶然；它只是我們第一次能看見這件事的月份。我們現在已經賣出 10 月批次超過 80%，趨勢仍在延續。我們核心現金產品（core cash products）的毛利，從售出 10% 到售出超過 80% 的過程中，只下滑了 90 個基點。

Last year, that same journey cost us over 260 basis points. We've seen about a 3x improvement. November, December, January, they all showed the same pattern four months in a row.

去年，同樣的過程讓我們付出了超過 260 個基點。我們看到大約 3 倍的改善。11 月、12 月、1 月，連續四個月都呈現同樣的模式。

In fact, Q4 of 2025 and January 2026 cohorts have the best combination of margin stability, and resale velocity of any cohort in Opendoor history, obviously excluding the COVID era.

事實上，2025 年第四季與 2026 年 1 月的批次，在毛利穩定性與轉售速度（resale velocity）的組合上，是 Opendoor 歷史上表現最好的批次，當然不包含 COVID 時期。

Our cohort curves or the slope of our margins as homes sell through are basically flat, and we're doing this at great speed. Every single cohort from October through January is selling faster than any corresponding cohort since COVID, and we're meaningfully scaling growth.

我們的批次曲線——也就是房屋逐步售出時毛利的斜率——基本上是平的，而且我們是以很快的速度做到的。從 10 月到 1 月的每一個批次，賣得都比 COVID 之後任何對應批次更快，而且我們正在有意義地擴大成長規模。

In Q1, we entered into contract on over 5,000 homes. That's 2 times bigger than Q4 and 3 times bigger than Q3. In fact, when it comes to contracts, this was our single best quarter since 2022.

在第一季，我們簽約（entered into contract）超過 5,000 間房屋。這是第四季的 2 倍、第三季的 3 倍。事實上，就簽約量而言，這是我們自 2022 年以來單季表現最好的一季。

Cohorts are performing better, resale velocity is improving, and we're scaling growth. How can we do this? Well, let's talk about it for a second.

批次表現更好、轉售速度在提升，而我們也在擴大成長。我們怎麼做到的？好，我們花點時間談談。

Two quarters ago, I laid out the blueprint and told you exactly what we were going to do. Underneath this all, there was one simple goal: make Opendoor faster.

兩個季度前，我提出了藍圖，並告訴各位我們要做什麼，而且說得一清二楚。在這一切之下，有一個簡單目標：讓 Opendoor 更快。

Last quarter, we graded ourselves and we're green across the board, and I promise we will do this every single quarter. Let's do that.

上個季度，我們為自己打分，整體都是綠燈，我承諾我們每一季都會這麼做。我們開始吧。

Step 1, profitability. Break even by the end of 2026. We're on track. We'll be ANI positive on a forward 12-month basis by the end of the year. As of April 1, Opendoor is Adjusted EBITDA profitable on a forward 12-month basis.

第 1 步，獲利能力。在 2026 年底前達到損益兩平。我們正按計畫推進。到今年年底，我們將在未來 12 個月（forward 12-month）基礎上實現 ANI 轉正。截至 4 月 1 日，Opendoor 在未來 12 個月基礎上已達到調整後 EBITDA 獲利。

Step 2, unit economics that make the model work. Positive contribution margin while increasing velocity. We're on track.

第 2 步，讓商業模式可行的單位經濟。在提升周轉速度的同時，實現正向貢獻毛利（contribution margin）。我們正按計畫推進。

Our contribution margin has increased every single month since we bottomed out in September, and October, November, December, and January cohorts, they're all selling faster than any corresponding cohort since COVID. We're improving margins, speeding up clearance. And we're doing all of this in a worse market.

自從我們在 9 月觸底以來，我們的貢獻毛利每個月都在提升；而 10 月、11 月、12 月與 1 月的批次（cohorts），其銷售速度都比 COVID 以來任何對應批次更快。我們正在改善毛利、加快出清速度。而且我們是在更差的市場環境中做到這一切。

Acquisitions are growing. In Q1, we entered into contract in over 5,000 homes, two times what we did in Q4, three times Q3. Our Q1 D2C acquisition contracts are up more than 4x compared to Q3 2025. This was our single best contract quarter since 2022.

收購量正在成長。在第一季，我們簽約超過 5,000 戶住宅，是第四季的兩倍、第三季的三倍。我們第一季的 D2C 收購簽約量，相較於 2025 年第三季成長超過 4 倍。這是我們自 2022 年以來單季簽約表現最好的一季。

Step 4, we're making really good progress on our capital-light products for sellers and transacting directly with buyers. Opendoor Checkout has now helped us sell homes in a bunch of states. More than one-third of our acquisition contracts in Q1 were cash now, more later.

第 4 步，我們在面向賣方的輕資本產品，以及直接與買方成交方面，取得了非常好的進展。Opendoor Checkout 現在已協助我們在多個州售出房屋。第一季我們的收購合約中，超過三分之一是「現在拿現金、之後再拿更多」（cash now, more later）。

This time last year, that number was exactly zero. That's our scoreboard. We're green across the board. This quarter, the scaffolding came down, and what's underneath is a company that finally knows exactly what it is and how it wins.

去年這個時候，這個數字是零。這就是我們的記分板。我們各項指標全面轉綠。本季，鷹架拆掉了，而底下是一家終於清楚知道自己是什麼、以及如何取勝的公司。

For a long time, the core assumption of Opendoor was that we had to be better at predicting the future than the rest of the world. We operated like a prop desk. We looked at the macro and made directional bets based on where we thought the prices were going to be in three, six, nine months, and then we pushed billions of dollars onto the table. The issue was never the people and not the model. The problem was the wrong problem to solve. Even if our models had been perfect, they were still pointed in the wrong direction.

很長一段時間，Opendoor 的核心假設是：我們必須比世界上其他人更擅長預測未來。我們像自營交易台（prop desk）一樣運作。我們看宏觀環境，根據我們認為三、六、九個月後房價會往哪個方向走來下注，然後把數十億美元押上桌。問題從來不在於人，也不在於模型。問題在於我們解的是錯的題目。即使我們的模型完美無缺，它們仍然指向了錯的方向。

Everything flowed from a single question: Where are home prices going? That one guess drove everything. It set the spread, which set what we bought, and determined whether or not we made money.

一切都源自一個問題：房價會往哪裡走？這一個猜測驅動了所有事情。它決定了價差（spread），進而決定我們買什麼，並決定我們是否賺錢。

When we got the answer wrong, we blamed the market every single time. Macro became our excuse for everything.

當我們答錯時，我們每一次都怪市場。宏觀環境成了我們所有事情的藉口。

Look, when predicting the future is your North Star, the reflex in a down market is always the same. Widen spreads, slow down, pull back, pray for the market to recover. Every defensive move fed the thing that was actually killing us. We were playing prevent defense when we were down by a touchdown. Of course, we were losing. We widened the spreads to protect ourselves, but in doing so, we changed the funnel. We changed the thing that was making the company work.

你看，當「預測未來」是你的北極星時，在下行市場裡的反射動作永遠一樣。拉大價差、放慢速度、收縮規模、祈禱市場回升。每一個防禦性動作都在餵養真正正在殺死我們的東西。我們落後一個達陣，卻在打防守保守戰術（prevent defense）。當然會輸。我們拉大價差來保護自己，但這麼做的同時，我們改變了漏斗。我們改變了讓公司運轉起來的那個關鍵機制。

We got worse homes. Worse homes meant worse margins. Worse margins went back into the model. The system got more conservative, and spreads widened even more. We didn't just have risk that we could not calculate.

我們買到更差的房子。更差的房子意味著更差的毛利。更差的毛利又回灌到模型裡。系統變得更保守，價差又被拉得更大。我們不只是面臨無法計算的風險。

We actually built a machine that amplified it. Every move made everything worse. That was our fatal flaw. In a business where time is risk, the old model taught us to slow way down.

我們其實打造了一台會放大風險的機器。每一步都讓情況更糟。那就是我們的致命缺陷。在一個「時間就是風險」的生意裡，舊模型卻教我們大幅放慢。

And once that reflex exists, every department in the company, product, operations, finance, everyone, starts running the same defensive operating system. The default everywhere was slow down just to protect ourselves.

而一旦這種反射存在，公司裡每個部門——產品、營運、財務、所有人——就會開始跑同一套防禦型作業系統。到處的預設都是：放慢，只為了保護自己。

Look, I'm a nerd's nerd. I think models are really cool, but they're incredibly worthless when you have the wrong strategy. So what we did wasn't just improve the pricing model, we changed the question that it was meant to answer.

你看，我是個徹頭徹尾的宅。我覺得模型很酷，但當策略錯了時，它們幾乎一文不值。所以我們做的不只是改進定價模型，而是改變了它原本要回答的問題。

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.

一年前，每個決策最重要的輸入是我們對房價升值的預測。今天，最重要的是：我們能把正在考慮買的那間房賣得多快。做市商（market makers）不是靠判對方向來贏。他們靠的是控制「判錯」時的曝險。他們靠的是判對時間。

When a prop desk gets scared, it pulls right back. That's how the spiral starts.

當自營交易台害怕時，它會立刻全面撤退。螺旋式下滑就是這樣開始的。

When a market maker sees risks, it does the exact opposite. It speeds up and prices to clear. The faster you move, the less any single home can hurt you. And velocity is how we know our pricing is right. A home that sits doesn't give us any signal. It just increases risk.

當做市商看到風險時，它做的正好相反。它會加速，並以能出清的價格定價。你動得越快，任何單一房屋對你的傷害就越小。而周轉速度（velocity）就是我們判斷定價是否正確的方式。一間房子如果賣不動，給不了我們任何訊號。它只會增加風險。

Opendoor 1.0 was, it was Kobayashi Maru. It wasn't a game we should have played. Without fundamentally changing it, we would have just totally killed the company. You don't beat that game by getting better at simulation. You beat it by changing the program. So we're now running on a velocity OS. The difference is totally structural.

Opendoor 1.0 就是——它是「小林丸」（Kobayashi Maru）。那不是我們應該玩的遊戲。如果不從根本上改變它，我們只會把公司徹底搞垮。你不是靠把模擬做得更好來贏那個遊戲。你是靠改變程式來贏。所以我們現在運行在一個「速度作業系統」（velocity OS）上。差異是完全結構性的。

We have rebuilt our engine around a totally stacked team of high-frequency thinkers. Signal intelligence officers, hedge fund quants, and they're all maniacally focused on data loops. They have a mandate, ship a change every single week, optimized for both margin and velocity.

我們已經用一支高度堆疊、具高頻思維的團隊，重建了我們的引擎。訊號情報官、避險基金量化分析師（quants），他們全都以近乎偏執的程度專注於資料迴圈。他們的任務是：每週交付一次變更，同時針對毛利與周轉速度最佳化。

And as our models get better, and they're getting better every single week, the whole machine moves faster. The whole company runs faster. We now run on a weekly cadence across the company. Products ship every single week. We don't need to be perfect in order for this business to work. We just need to be faster. With hundreds of acquisitions a week, we see pricing signals, renovation costs, and clearance patterns faster than anyone else in this market. Every home teaches us something. Every single day we shave hold times, our capital turns go up, and our returns go up.

而隨著我們的模型變得更好——而且每週都在變得更好——整台機器就跑得更快。整家公司都跑得更快。我們現在全公司以每週節奏運作。產品每週都上線。我們不需要完美，這門生意也能運作。我們只需要更快。每週數百筆收購，讓我們比市場上任何人更快看到定價訊號、整修成本與出清模式。每一間房子都教會我們一些事。我們每天縮短持有時間（hold times），資本周轉次數就上升，報酬也上升。

Speed. Speed pays for everything. In a bad market or a great one, the variable thing that actually matters is time. When you ask what has made the change, what makes this whole thing work, it's one word, faster.

速度。速度為一切買單。不論市場很差或很棒，真正重要的可變因素是時間。當你問是什麼帶來改變、是什麼讓整件事能運作，答案只有一個字：更快。

I wear a T-shirt at every financial open house that says one thing, faster. You can see it. I'm wearing one right now. Most of you think, it's just a personality quirk, right? A founder nerd thing, a costume of a wartime CEO. It's really not.

我每次在財務開放日都穿一件 T 恤，上面只寫一個字：更快。你們看得到。我現在就穿著。你們大多數人以為，那只是個性怪癖，對吧？創辦人宅的習慣，一種戰時 CEO 的戲服。其實不是。

Look, we used to be in a business that lived or died on whether we got the future right. Now, we're in a business that lives or die on whether we move fast. Faster isn't just our competitive advantage. It's an absolute moral imperative.

你看，我們過去所處的業務，是成敗取決於我們是否把未來判斷正確。而現在，我們所處的業務，是成敗取決於我們是否行動夠快。更快不只是我們的競爭優勢。它是一項絕對的道德責任。

Let me just say this again so you don't think I'm being subtle about it. Faster is not just our competitive advantage. It's our whole reason for being here. It's our moral imperative.

我再說一次，免得你以為我在拐彎抹角。更快不只是我們的競爭優勢。它是我們存在於此的全部理由。它是我們的道德責任。

Every day, someone is stuck and cannot move is a day in their life that they cannot move on. They're on hold. If we're in the business of helping people move, then days matter. It's a job offers they haven't accepted, a planned retirement put on hold, a family not started.

每天，只要有人被卡住、無法前進，那就是他們人生中無法繼續往前的一天。他們被迫按下暫停鍵。如果我們的業務是幫助人們搬遷與前進，那麼每一天都很重要。那可能是一份還沒接受的工作邀約、一個被延後的退休計畫、一個尚未開始的家庭。

The traditional home sale process is more than just inconvenient. It holds these people back. 40 million homeowners in this country want to move in the next 12 months, but only one in five think they can actually do it. Not because moving is too expensive, but because everything about it is just too uncertain.

傳統的房屋出售流程不只是麻煩而已。它會拖住這些人。在這個國家，有4,000萬名屋主希望在未來12個月內搬家，但只有五分之一的人認為自己真的做得到。不是因為搬家太貴，而是因為整個過程的一切都太不確定。

Here's a simple test for any system. Would you design it this way if your family had to live in it? The legacy real estate system fails the test. It's our job to fix this.

任何系統都有一個簡單的測試。如果你的家人必須生活在其中，你會把它設計成這樣嗎？傳統的房地產體系通不過這個測試。修正它是我們的工作。

Every product decision Opendoor goes through goes through one single filter. We only care about one thing. Are we returning time to people?

Opendoor 的每一個產品決策，都會經過同一個唯一的篩選條件。我們只在乎一件事。我們是否把時間還給人們？

Last week, across all sellers, Opendoor gave back over 100 years of time. Over 100 years of time. Over 500 families said yes to an Opendoor offer and reached certainty about 90 days sooner than they would have in a traditional process. You do the math. In just one week, we got rid of a century of human waiting, time that got returned to families who got to move on. Faster is a moral imperative. It is a good in and of itself. That is what this company is for.

上週，綜合所有賣方，Opendoor 返還了超過100年的時間。超過100年的時間。有超過500個家庭接受了 Opendoor 的出價，比起傳統流程提早約90天獲得確定性。你自己算算。僅僅一週，我們就消除了整整一個世紀的人類等待，把時間還給那些得以繼續前進的家庭。更快是一項道德責任。它本身就是一種善。這就是這家公司存在的目的。

Every product launch ultimately serves one question: How do we move faster for sellers, for buyers, for Opendoor, for everyone? So let me run through some product launches.

每一次產品上線，最終都在回答同一個問題：我們如何讓賣方、買方、Opendoor，以及所有人都能更快？所以我來快速介紹幾個產品上線。

This quarter, we expanded cash now, more later coverage. Every week, hundreds of families who would have heard, sorry, we can't help you, are now getting real offers. We totally rebuilt the foundations of our buyer apps.

本季，我們擴大了「先拿現金、之後再拿更多」的適用範圍。每週都有數百個家庭，原本會聽到「抱歉，我們幫不了你」，現在都能拿到真正的出價。我們也徹底重建了買方 App 的基礎架構。

We acquired Doma's escrow division. Noah, our AI underwriter, prices normal homes in Phoenix now. We rebuilt every message a buyer gets from us. Six different systems became just one conversation. We rebuilt our offer page, giving customers the type of information they would have gotten from an expert who was at their kitchen table.

我們收購了 Doma 的代管（escrow）部門。我們的 AI 核保員 Noah 現在能為鳳凰城的一般住宅定價。我們重寫了買方從我們這裡收到的每一則訊息。六個不同的系統，變成一段單一的對話。我們重建了出價頁面，讓客戶獲得的資訊，就像有位專家坐在他們的廚房餐桌旁提供的一樣。

We also built a portable assessment scheduling. You can now get your home assessment done on your own terms. More than half of our assessments are now seller-led, 6,000 in March alone. We migrated our component library to an AI-native front end.

我們也打造了可攜式的估價排程功能。你現在可以依照自己的時間安排完成房屋估價。目前超過一半的估價由賣方主導，光是3月就有6,000件。我們把元件庫遷移到 AI 原生的前端。

While we were in there, we killed our legacy cake service. A transition that had failed three times in four years got finished in six weeks. The platform is what makes everything else faster.

在處理這些的同時，我們也淘汰了舊的 cake 服務。一個在四年內失敗三次的轉換，六週就完成了。平台正是讓其他一切變得更快的關鍵。

Talk to any Opendoor engineer, and they'll tell you this is a really big deal. We built an AI audit tool that automatically reconciles inspection scopes with actual repair decisions, giving our field teams real, actionable feedback to improve operator compliance and cost discipline.

你去問任何一位 Opendoor 工程師，他們都會告訴你這件事非常重要。我們打造了一個 AI 稽核工具，能自動把驗屋範圍與實際維修決策進行對帳，為我們的現場團隊提供真實、可執行的回饋，以提升作業遵循度與成本紀律。

At title intake, it used to take us up to five hours, and it now takes 15 minutes. We launched Opendoor Mortgage in Colorado. Now, one of our marketing managers replaced our half-a-million-dollar life cycle legacy email system with one cloud skill. A field manager in our Southeast division runs five states on cloud. A finance team turned 20 hours of SOX deliverables into one-minute query. None of these people, by the way, were engineers.

在產權資料導入（title intake）這一段，以前最多要花我們五個小時，現在只要15分鐘。我們在科羅拉多州推出了 Opendoor Mortgage。現在，我們的一位行銷經理用一個雲端技能，取代了我們一套50萬美元的生命週期舊式電子郵件系統。我們東南區的一位現場經理在雲端管理五個州。財務團隊把20小時的 SOX 交付物，變成一個一分鐘的查詢。順帶一提，這些人都不是工程師。

We also tripled our cash now, more later product. Our voice bots dropped seller contract time from 30 minutes to 5. We replaced 72 manual exports a month with one pipeline. We built a new listing operating consoles in eight days. We merged 8 different HR systems into one. In the process, we killed dozens of point solutions. Now, that's not the full list. It's just what I had time for before they played the walk me off stage music.

我們也把「先拿現金、之後再拿更多」產品的規模擴大到三倍。我們的語音機器人把賣方簽約時間從30分鐘降到5分鐘。我們用一條管線取代了每月72次的手動匯出。我們在八天內打造了新的刊登作業控制台。我們把8套不同的人資系統整併成一套。在這個過程中，我們淘汰了數十個點狀解決方案。當然，這不是完整清單。只是我在他們播放「請我下台」的音樂之前來得及講的內容。

As you can tell, we've changed a lot, but I also want to tell you what we haven't figured out. Look, I'm a Leafs fan. I know what it feels like to be promised lots of things and get absolutely none of them. I know what it feels like to watch the same group of people over and over again give you false hope and give you nothing.

如你所見，我們改變了很多，但我也想告訴你我們還沒搞定的事。你看，我是楓葉隊（Leafs）的球迷。我知道那種被承諾很多、最後卻什麼都沒有的感覺。我知道那種看著同一群人一再給你虛假的希望、卻什麼也沒帶來的感覺。

I've thought about this more than I probably should. I've decided that a promise is not the same thing as a proof. You do not get credit for what should have happened. You only get credit for what actually did. I know what it feels like to have momentum in March and tears in May, which is why this T-shirt says faster and not done.

我對這件事想得可能比我應該的還多。我得出的結論是：承諾不等於證明。你不會因為「本來應該發生」而得到功勞。你只會因為「實際發生」而得到功勞。我知道三月有氣勢、五月卻落淚是什麼感覺，所以這件 T 恤上寫的是 faster，而不是 done。

Faster is a setting. It's not a destination. We don't get to celebrate signals. Every quarter is just another shift for us. We're not done. We're not even close.

更快是一種設定。它不是一個目的地。我們不能為了訊號而慶祝。每一季對我們來說都只是另一個輪班。我們還沒完成。我們甚至還差得遠。

Mortgage is live. The early data is honestly going a lot better than I thought it would go. We're getting really good attach rates. Our customers love it.

房貸業務已經上線。老實說，早期數據比我原本預期的好很多。我們的搭售率（attach rates）非常不錯。客戶很喜歡。

Look, it's early. We don't fully know how the product is going to work across different market conditions in different home price tiers. We have a thesis. It's working really well. We haven't proven it at scale; cash now, more later. It's growing really fast. It's over a third of our growing pie. That's just really remarkable for a product that didn't exist a year ago, and that was totally reworked just three 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.

你看，現在還早。我們還不完全知道，在不同市場條件、不同房價區間下，這個產品會如何運作。我們有一個假設。目前運作得非常好。但我們還沒有在規模化層面證明它；「先拿現金、之後再拿更多」成長非常快。它成長得非常快。它已經占我們成長版圖的三分之一以上。對於一個一年前還不存在、而且三個月前才徹底重做的產品來說，這真的非常了不起。我們正在迭代它的運作方式。我們在微調，試著在賣方拿到的價值與 Opendoor 保留的價值之間取得正確平衡。

But let me be honest with you. Every product that Opendoor ships has to earn its place in our portfolio. Cash now, more later is earning it, but we're not done changing it. Like we said earlier, the housing market, look, it just remains what it is.

但我想對你坦白。Opendoor 推出的每一個產品，都必須在我們的產品組合中贏得它的位置。「先拿現金、之後再拿更多」正在贏得這個位置，但我們還沒停止改造它。就像我們先前說的，房市嘛，你看，它仍然就是它原本的樣子。

We believe the model we have built on faster works across macro cycles. We're no longer dependent on a macro. We control our own destiny. October, November, December, January cohorts, they were all bought during the most aggressive expansion in our history in a market that I don't think anyone would describe as favorable, and this is the best evidence we have. That's just what it is. It's evidence. It's not proof. Proof will take more time, more reps, more shifts, more aggression, more products shipped faster.

我們相信，我們建立在「更快」之上的模式，能夠穿越不同的總體經濟循環。我們不再依賴總體環境。我們掌握自己的命運。10 月、11 月、12 月、1 月的批次，都是在我們史上最激進的擴張期間買進的，當時的市場我想沒有人會形容為有利，而這就是我們手上最好的證據。事實就是如此。這是證據。但不是證明。要形成證明，還需要更多時間、更多次實作、更多輪轉、更多進攻，以及更快地推出更多產品。

And we've said this before, and you'll always hear us saying this. We're not asking you to take our word for it. We're asking you to watch and to hold us accountable. Christy is going to walk you through the numbers in a minute, but before she does, I want to close with this. I've been asked a lot what Opendoor is.

我們以前就說過，而且你們也會一直聽到我們這麼說。我們不是要你們只聽我們說。我們希望你們觀察，並要求我們負責。Christy 等一下會帶大家看數字，但在她開始之前，我想用這段話作結。很多人問我 Opendoor 是什麼。

We changed our LinkedIn profile from real estate to software, but software is too generic. Look, Opendoor is on a mission. Our job is to get people who are stuck moving. We're a machine that helps America move.

我們把 LinkedIn 的產業標籤從房地產改成軟體，但「軟體」又太籠統。聽著，Opendoor 有一個使命。我們的工作，是幫助那些被搬家卡住的人動起來。我們是一台幫助美國搬家的機器。

When I joined Opendoor, I did it because the homeownership matters. It is a thing. It is the single thing that leads to better families, better neighborhoods. When people buy a home they love, they're buying a share in this country.

我加入 Opendoor，是因為擁有住房這件事很重要。這是真實存在的。它是促成更好的家庭、更好的社區的關鍵因素。當人們買到自己喜歡的房子，他們買的是這個國家的一份參與。

We don't buy homes at Opendoor to hold them. We buy them. We buy them so we can get them into a next family faster, with less friction, at a better price.

在 Opendoor，我們買房不是為了持有。我們買房。我們買房，是為了能更快、摩擦更少、以更好的價格，把房子交到下一個家庭手上。

Every family we help move is a family that is putting down roots. It's a neighborhood that's getting better. It's children that get to grow up in a home that their parents love.

我們每幫助一個家庭搬家，就是一個家庭開始扎根。就是一個社區變得更好。就是孩子能在父母喜愛的家中長大。

Faster is what this company was built to do. The T-shirt, that's just a reminder.

「更快」是這家公司成立的初衷。那件 T 恤，只是一個提醒。

The Opendoor machine is now running, and every day it runs, every single day it runs, friction disappears and people move. We do not need a better market. We just need a better machine.

Opendoor 這台機器現在已經運轉起來，而且它每天運轉、每天運轉，每一天運轉，摩擦就會消失，人們就能搬家。我們不需要更好的市場。我們只需要更好的機器。

Last week, we gave back over 100 years. That's 100 years of human pain just gone. That's not corporate jargon; that's just families moving and building better lives.

上週，我們為人們省回了超過 100 年。那是 100 年的人類痛苦，就這樣消失了。這不是企業術語；這只是家庭在搬家、在打造更好的生活。

Please track it. Please hold us accountable. Christy?

請追蹤它。請要求我們負責。Christy？

Thank you, Kaz. I'm not wearing the T-shirt, but I promise I'll match the pace. Three things to know about Q1 before we get into the details.

謝謝你，Kaz。我沒有穿那件 T 恤，但我保證會跟上節奏。在進入細節之前，關於第一季有三件事要先知道。

One, we reduced aged inventory from 51% to 10% in two quarters. The book is the freshest it's been in nearly four years.

第一，我們在兩個季度內把高齡庫存占比從 51% 降到 10%。我們的庫存組合是近四年來最新鮮的一次。

Two, margins bottomed out in September and have improved every month for six months straight. Q1 closed at 4.4%, up 3.4 points quarter over quarter, and we expect the upward trend to continue into next quarter.

第二，毛利率在 9 月觸底，並且連續六個月每月改善。第一季收在 4.4%，較上一季提升 3.4 個百分點，我們預期這個上升趨勢會延續到下一季。

Three, acquisitions are up 45% from Q4, Q1 was our strongest quarter for signed contracts since Q2 2022. The headline behind those three, starting in Q2 2026, we expect to be Adjusted EBITDA profitable on a 12-month go-forward basis. The machine is working. Let's get into it.

第三，收購量較第四季成長 45%，第一季是自 2022 年第二季以來簽約量最強的一季。這三點背後的重點是：從 2026 年第二季開始，我們預期在未來 12 個月的前瞻基礎上，調整後 EBITDA 將可達到獲利。這台機器正在運作。我們開始吧。

As a reminder, we are executing against three management objectives on our path to profitability. The table in our earnings release shows our progress on each. Let me walk through the highlights.

提醒一下，我們正依循三項管理目標執行，以走向獲利。我們的財報新聞稿中的表格顯示了各項進度。我來帶大家看重點。

First, scale acquisitions. We purchased 2,474 homes in Q1, up 45% from Q4. This is the second-consecutive quarter of meaningful growth. And signed acquisition contracts, our leading indicator, tell an even stronger story. March was our highest single month for signed contracts since June 2022, and Q1 was our highest quarter since Q2 2022.

第一，擴大收購規模。第一季我們買入 2,474 間房屋，較第四季成長 45%。這是連續第二個季度出現有意義的成長。而作為領先指標的「已簽署收購合約」呈現更強勁的故事。3 月是自 2022 年 6 月以來單月簽約量最高的一個月，而第一季是自 2022 年第二季以來最高的一季。

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.

收購合約通常會在大約一個月後完成交割。第一季的 2,474 筆購買，多數來自第四季末、第一季初的合約。第一季後段簽的合約，主要會在第二季完成交割。

Also, we want to be clear, we don't close on every home we go into contract on. Under Opendoor 2.0, we're deliberate about which contracts we take all the way to purchase, so the funnel narrows between contract and close. More contracts mean more opportunities to be selective, and the trajectory matters.

另外，我們想說清楚：我們不會對每一間簽約的房子都完成交割。在 Opendoor 2.0 之下，我們會審慎決定哪些合約會一路走到實際購買，因此從簽約到交割之間，漏斗會收斂。更多合約代表有更多機會做選擇，而趨勢本身很重要。

In a short period of time, we've gone from our lowest contract volume since COVID to our highest since 2022. This is the tempo required to achieve the goals we've set for ourselves, and we're building the volume and the discipline at the same time. You can continue to track our weekly progress on accountable.opendoor.com.

在很短的時間內，我們從 COVID 以來最低的簽約量，提升到自 2022 年以來最高。這就是達成我們為自己設定目標所需要的節奏，而我們同時在建立量能與紀律。你們可以在 accountable.opendoor.com 持續追蹤我們每週的進度。

Volume only counts if the quality holds, and our second management objective is the scorecard for whether we're delivering the right kind of growth.

只有在品質維持的前提下，量才有意義；而我們的第二項管理目標，就是用來衡量我們是否交付了正確類型的成長的記分卡。

Second, improve unit economics and resale velocity. This is where the work really shows up, and there are three data points I want to highlight.

第二，改善單位經濟效益與轉售速度。這是成果真正顯現的地方，我想強調三個數據點。

One, resale contribution margin has improved every month since September 2025, closing Q1 at 4.4%, up 3.4 percentage points quarter over quarter.

第一，自 2025 年 9 月以來，轉售貢獻毛利率每個月都在改善，第一季收在 4.4%，較上一季提升 3.4 個百分點。

Two, our Q4 2025 and January 2026 cash acquisition cohorts have the best combination of margin stability, and resale velocity of any corresponding cohort in company history, excluding the COVID era.

第二，我們 2025 年第四季與 2026 年 1 月的現金收購批次，在毛利率穩定性與轉售速度的組合上，是公司歷史上所有對應批次中表現最好的（不含 COVID 時期）。

And three, the percentage of homes on the market for more than 120 days fell to 10%, down from 33% at year-end and 51% at the end of Q3, a 41-percentage-point improvement in just two quarters.

第三，在市場上掛牌超過 120 天的房屋占比降至 10%，低於年末的 33% 與第三季末的 51%，僅兩個季度就改善了 41 個百分點。

Let me stay at this point for a moment. 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.

我在這點上多停留一下。兩個季度前，我們超過一半的房屋在市場上已經放了超過 120 天。到第一季末，這個數字是 10%。這是自 2022 年第二季以來的最低水準。

To put it in perspective, the broader market was at 23% two quarters ago and rose to 33% at the end of Q1. We are now carrying a book that is materially fresher and healthier than the market.

作為對照，更廣泛的市場在兩個季度前是 23%，到第一季末上升到 33%。我們現在持有的庫存組合，明顯比市場更新鮮、也更健康。

Inventory health is both a leading indicator of forward margin and evidence that our approach is working. A faster moving book means lower holding costs, less market exposure, better resale outcomes, and more efficient use of capital, and that's exactly what's showing up in our margins. This didn't happen because the market got friendlier. It happened because of tailored underwriting, disciplined close to listing workflows, and resale systems designed to move homes quickly while protecting unit economics.

庫存健康度既是未來毛利率的領先指標，也是我們方法正在奏效的證據。更快周轉的庫存組合意味著更低的持有成本、更少的市場曝險、更好的轉售結果，以及更有效率的資本運用，而這正反映在我們的毛利率上。這不是因為市場變得更友善。而是因為量身打造的承保、從交割到上架的紀律化流程，以及為了在保護單位經濟效益的同時快速推動房屋成交而設計的轉售系統。

Third, build operating leverage. Fixed operating expenses were $33 million in Q1, down $2 million quarter-over-quarter and down $6 million year over year. Our trailing 12-month operations expense as a percentage of trailing 12-month revenue held steady quarter over quarter at 1.3%. We are holding the fixed cost base flat while simultaneously investing in the AI and infrastructure that powers our product, and it's worth pausing here for a minute.

第三，建立營運槓桿。第一季固定營運費用為3,300萬美元，較上一季減少200萬美元，較去年同期減少600萬美元。我們過去12個月的營運費用占過去12個月營收的比例，季對季維持在1.3%不變。我們在維持固定成本基礎持平的同時，也持續投資於驅動我們產品的AI與基礎設施，這裡值得先停下來說明一下。

We're going all in on AI, and we're doing it responsibly. 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. We have an internal Slack channel called Default to AI where teams celebrate measurable impact.

我們正全力投入AI，而且以負責任的方式來做。目前有很多雜音在談一些公司在第二季之前就把2026年的預算砸在AI上。我們不是那樣。我們專注於成果，而不是代幣排行榜。我們有一個內部Slack頻道叫做「Default to AI」，各團隊會在上面分享並慶祝可量化的影響。

Some highlights in addition to what Kaz shared earlier, an AI-powered repair negotiation tool cut our buyer fall-through rate by over double digits. Field managers are using AI scoping feedback, helping to reduce pre-list renovation spend by up to 10% to 20% per home in pilot markets. A ticket triage automation redeployed three full-time employees from classification to resolution.

除了Kaz先前分享的內容之外，再補充幾個亮點：一個AI驅動的修繕議價工具，讓我們的買方成交流失率下降了超過兩位數百分點。現場管理人員正在使用AI範疇界定回饋，在試點市場中，協助將每戶在上市前的整修支出降低約10%到20%。一套工單分流自動化，讓三名全職員工從分類工作轉而投入到解決問題。

What's notable is that most of these tools were built by operators, not engineers, using the AI infrastructure we've invested in. We're cutting waste and reallocating into capabilities that move the business.

值得注意的是，這些工具大多是由營運人員而非工程師所打造，使用的是我們已投資的AI基礎設施。我們正在減少浪費，並把資源重新配置到能推動業務的能力上。

Our flat fixed operating expense is the output of that discipline, not the absence of investment. Three objectives, three quarters of consistent progress. The plan is working.

我們固定營運費用持平，是這種紀律的結果，而不是沒有投資。三個目標，連續三個季度穩定進展。這個計畫正在奏效。

Turning to the balance sheet, we ended the quarter with $999 million in unrestricted cash, our highest cash balance in years. That's a product of two things: the strength of our parent level capital position following the work we did last fall and the health of our inventory book.

接著看資產負債表，我們在季度末擁有9.99億美元的非受限現金，為多年來最高的現金水位。這來自兩個因素：我們去年秋天完成相關工作後，母公司層級資本部位的強勁，以及我們庫存帳簿的健康狀況。

We held 3,420 homes in inventory at quarter end, representing $1.1 billion in net inventory. Our non-recourse asset-backed borrowing capacity remains robust at $7.1 billion, with $1.5 billion committed. Between liquidity, facility capacity, and the quality of what we're financing under those facilities, we have meaningful flexibility to execute against our plans.

我們在季度末持有3,420戶庫存住宅，對應淨庫存為11億美元。我們的無追索權資產擔保借款額度仍然強勁，為71億美元，其中15億美元已承諾。在流動性、融資工具額度，以及我們在這些融資工具下所融資資產的品質之間，我們具備相當的彈性來執行我們的計畫。

Now, let me give you the guidepost for Q2. Acquisitions. You can continue to track our acquisition contracts on accountable.opendoor.com. We've updated our contract road map for the remainder of the year. The ranges reflect our current outlook inclusive of typical seasonality, and we'll continue to update them each quarter as we learn more.

現在，讓我給大家第二季的指引重點。收購。你們可以持續在accountable.opendoor.com追蹤我們的收購合約。我們已更新今年剩餘期間的合約路線圖。這些區間反映了我們目前的展望，並已納入典型的季節性因素；隨著我們獲得更多資訊，我們也會每季持續更新。

Revenue. Our Q1 increase in home acquisitions will start to flow through to resales, leading to expected revenue growth of approximately 25% quarter over quarter.

營收。我們第一季房屋收購量的增加將開始反映在轉售上，預期營收將季對季成長約25%。

Contribution margin. Our contribution margin bottomed out in September and has been improving every single month since then. We expect the contribution Margin for Q2 2026 to fall in the middle of our 5% to 7% goal we shared in the first Opendoor 2.0 financial open house.

貢獻毛利率。我們的貢獻毛利率在9月觸底，之後每個月都在改善。我們預期2026年第二季的貢獻毛利率將落在我們在第一場Opendoor 2.0財務開放日所分享的5%到7%目標區間的中間位置。

Adjusted EBITDA. We expect Q2 adjusted EBITDA to be break even, plus or minus a few million dollars. We see Q2 as an inflection point. We expect to be adjusted EBITDA profitable on a 12-month go-forward basis starting in Q2.

調整後EBITDA。我們預期第二季調整後EBITDA將接近損益兩平，上下浮動幾百萬美元。我們將第二季視為一個轉折點。我們預期自第二季起，以未來12個月的前瞻基礎來看，將能達到調整後EBITDA獲利。

In closing, last quarter I said you can't build a great business in a spreadsheet. You build it by shipping product, operating with discipline, and learning from the market. Q1 is what that looks like when the machine starts to work.

最後總結，上個季度我說過，你無法只靠試算表打造一家偉大的企業。你要靠推出產品、以紀律營運，並從市場中學習來打造。第一季就是當這部機器開始運轉時的樣子。

Acquisitions, margin, resale velocity, inventory health, and cost all moved the right way at the same time. That's not a lucky coincidence. That's a system that's working.

收購、毛利率、轉售速度、庫存健康度與成本，同時都朝正確方向改善。這不是幸運的巧合。這是一個正在運作的系統。

Two quarters ago, we laid out our plan. Every quarter since, we graded ourselves against it and delivered. We have a lot left to prove. We intend to keep doing exactly that.

兩個季度前，我們提出了計畫。此後每一季，我們都依此自我評分並交出成果。我們還有很多需要證明。我們打算繼續這麼做。

With that, Michael, I'll turn it over to you for questions.

接下來，Michael，我把時間交給你進行提問。

Great. Thanks, Christy. Our first question comes to us via video submission from Mike Alfred.

好的。謝謝你，Christy。我們的第一個問題是透過影片投稿，由Mike Alfred提出。

Hey, guys, it's Mike Alfred, Founder and Managing Partner of Alpine Fox LP, as well as a Board Director at IREN and Bakkt. Great job on the execution side. I really like the way the business is integrating AI into everything you're doing.

嗨，各位，我是Mike Alfred，Alpine Fox LP的創辦人兼管理合夥人，同時也是IREN與Bakkt的董事。在執行面做得很棒。我非常喜歡你們把AI整合進你們所做的一切的方式。

My question is about the longer term implications of AI. 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? Thanks, guys.

我的問題是關於AI的長期影響。當你們看公司的策略方向時，你們是否相信我們已為AI在未來幾年可能改變房地產市場運作方式的各種變化做好充分準備？謝謝各位。

Oh, wow, that's a great question. Look, I think the answer of this is like in a bunch of layers. I kind of think about the layers, so let me just go through them.

哇，這是個很棒的問題。我想答案可以分成好幾個層次。我會用分層的方式來思考，所以我就逐一說明。

Layer one is the, like, the earning call answer. AI is important. We're leaning right in. We're spreading across the entire business. You kind of hear that from every corporate CEO. I mean, it's true, but it tells you like nothing actually useful.

第一層有點像財報電話會議的標準回答。AI很重要。我們正積極投入。我們把它擴展到整個業務。你幾乎會從每位企業CEO那裡聽到這些。我的意思是，這是真的，但其實沒有告訴你任何真正有用的資訊。

Layer two is actually important. That's like the software leverage story. Like the original SaaS era, the insight was that you could take a CRUD database, wrap business logic around it and some workflow around it, and then you'd find that people could do a lot more, right? Software would get cheaper.

第二層其實很重要。那是軟體槓桿的故事。就像最初的SaaS時代，洞見在於：你可以拿一個CRUD資料庫，外面包上一層商業邏輯與工作流程，然後你會發現人們能做更多事情，對吧？軟體會變得更便宜。

People could do a lot more because you could encode the rules and the processes and the decision-making into software, and the leverage was just insane.

人們能做更多，因為你可以把規則、流程與決策制定編碼進軟體裡，而那種槓桿效應簡直驚人。

AI extends that by quite a bit because you're now encoding judgment on top of rules, the leverage becomes really high. That's real, it's important, we're capturing a lot of this, but that's just a story of software broadly. It doesn't say anything specific about Opendoor. We just happen to be honestly just really good at this.

AI在此基礎上又延伸了不少，因為你現在是在規則之上再編碼「判斷」，槓桿就會變得非常高。這是真實的、很重要，我們也正在捕捉其中很多價值，但那終究是廣義軟體的故事。它並沒有說明任何Opendoor特有的東西。我們只是碰巧——老實說——在這方面做得非常好。

Layer three is actually fundamentally more interesting. It's like the automation versus the collaboration split. AI as collaboration software is very misunderstood. Let me talk about that for a second.

第三層在本質上更有意思。這有點像自動化與協作之間的分野。把AI當作協作軟體，其實被嚴重誤解。我花點時間談談這個。

Look, our goal isn't to use AI to cut 15% of our expenses by doing the same things we were doing, just cheaper, right? Like, that's automation applied to cost. And the goal isn't that, like, a black box replaces a human process end to end. That's just not what we're doing. Like, 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.

你看，我們的目標不是用AI把我們的費用砍掉15%，只是把原本在做的事情用更低成本做一遍，對吧？那是把自動化用在成本上。目標也不是讓一個黑盒子端到端取代人類流程。那不是我們在做的事。我們真正想做的是，鑑於AI能做的一切，從零開始、從一張白紙重建我們的流程，讓我們能用AI打造一個本質上不同的流程。

Layer four is actually our complexity as a structural advantage. This one is important to understand, and this is why we're not afraid of AI as the way like some software incumbents are. Real estate is atoms and risk and not just bits. It's also some bits. The underlying transaction involves a level of complexity and condition and local dynamics and human emotion, and all of it, like, makes the system very complex, and that's actually our advantage, right?

第四層其實是我們把複雜性當作結構性優勢。這點很重要要理解，這也是為什麼我們不像某些既有軟體業者那樣害怕 AI。房地產是由實體資產與風險構成，而不只是位元。當然也有一些位元成分。底層交易牽涉到一定程度的複雜性、條件、在地動態與人類情緒，所有這些讓整個系統非常複雜，而這其實就是我們的優勢，對吧？

AI doesn't eliminate this complexity. It just makes navigating it a lot, a lot easier. What we don't need to do here is just get to some hypothetical end state. We just need to be meaningfully better than the alternative and the legacy process at every step.

AI 不會消除這種複雜性。它只是讓你在其中穿梭變得容易很多、很多。我們在這裡不需要做的是抵達某個假想的終局狀態。我們只需要在每一步都顯著優於替代方案與傳統流程。

Like, this is a Red Queen's race dynamic, and it works in our favor here. Look, we've been running in this very complex environment for years, and we have a crap ton of operational knowledge, and that is deeply, deeply useful.

這有點像「紅皇后競賽」的動態，而在這裡它對我們有利。你看，我們多年來一直在這個高度複雜的環境中奔跑，我們累積了大量的營運知識，而那是非常、非常有用的。

Okay, last, I promise this is the last layer. I like five-layer cakes. The fifth layer is about what AI does to the other side of transaction. Like, there are two parts to this, right? What the customer feels and sees and what it does to the category.

好，最後，我保證這是最後一層。我喜歡五層蛋糕。第五層是關於 AI 對交易另一端會造成什麼影響。大概有兩個部分，對吧？客戶的感受與所見，以及它對整個品類的影響。

On the customer side, look, the traditional real estate process is defined or by information asymmetry, right? That's just not an accident. That's the foundation of the whole process, right? Experts who know the market make profit from transaction friction because the parties themselves can't navigate it.

在客戶端，你看，傳統房地產流程的特徵是資訊不對稱，對吧？這不是偶然。這是整個流程的基礎，對吧？熟悉市場的專家，因為交易摩擦而獲利，因為交易雙方自己無法順利穿越這些摩擦。

AI totally dissolves this asymmetry. What that means is the customers are being, like, upgraded. We can build AI concierge things that feel to the customer like the expert is sitting at their kitchen table, right? That's an incredibly important thing, and it's what we're doing.

AI 會徹底瓦解這種不對稱。這代表客戶正在被「升級」。我們可以打造 AI 禮賓式服務，讓客戶感覺就像專家坐在他們的廚房餐桌旁一樣，對吧？這非常重要，而這正是我們正在做的事。

On the category side, this is the actual meta bet. Every major internet transition, every industry has had winners that didn't just jam the Sears catalog into a browser, they actually helped with the transaction. Travel, retail, fintech. That's been true across every internet.

在品類端，這其實是更高層次的押注。每一次重大的網際網路轉型、每個產業，都出現過贏家；他們不是把 Sears 型錄硬塞進瀏覽器，而是實際幫助完成交易。旅遊、零售、金融科技。這在每一波網際網路浪潮中都成立。

It just hasn't happened in real estate. Real estate's, like, honestly the last, major holdout. Not because the category is fundamentally immune from this, but because the underlying complexity made it a little too messy to transact at scale. AI just totally removes this constraint.

只是房地產一直沒有發生。老實說，房地產是最後一個主要的堡壘。不是因為這個品類天生對此免疫，而是因為底層的複雜性讓大規模交易變得有點太混亂。AI 會徹底移除這個限制。

I think I should have actually started the answer by saying yes. Yes, we believe we're well-positioned. It's honestly on the inside, it feels as though our business was built waiting for this manna to fall from heaven, and it now has.

我想我其實應該用「是」來開頭回答。是的，我們相信我們處於有利位置。老實說，在公司內部的感覺就像我們的業務是為了等待這份天降甘霖而打造的，而現在它真的降臨了。

Great. We had a few questions submitted via Say Technologies that all kind of clustered around profitability, so I wanted to pull out two.

很好。我們透過 Say Technologies 收到幾個提問，主題大多集中在獲利能力上，所以我想挑出兩個。

The first comes from Heejun C., who's asking, you said in the last earnings call that turning profitable by the end of the year was achievable. Now that the first quarter has passed and interest rates remain high, is that still a realistic goal?

第一個來自 Heejun C.，他問：你在上一季財報電話會議中說年底前轉虧為盈是可達成的。現在第一季已經過去、利率仍然偏高，這仍然是現實的目標嗎？

Also, Arun Jacob V. asks, how confident are you today in the Q2 positive EBITDA and year-end profitability forecast? What are the key swing factors from here which might influence it?

另外，Arun Jacob V. 問：你們今天對於 2026 年第二季 EBITDA 轉正以及年底獲利預測有多大信心？從現在起有哪些關鍵的變動因素可能影響結果？

Those are great questions. Thank you. We reconfirmed our goal and expectations earlier on the call, and I'll say it again here. We expect Opendoor to be break-even or profitable, adjusted net income profitable by the end of this year on a 12-month go-forward basis.

這些問題很好。謝謝。我們在稍早的電話會議中已再次確認我們的目標與預期，我也在這裡再說一次。我們預期 Opendoor 在今年年底前，按未來 12 個月滾動基礎計算，將達到損益兩平或獲利（以調整後淨利計）。

Arun, to answer your question, we also shared in the call earlier that we're going to reach an important milestone on that path to profitability, in that starting in Q2 2026, we expect to be adjusted EBITDA profitable on a 12-month go-forward basis.

Arun，回答你的問題，我們也在稍早的電話會議中分享過：我們將在通往獲利的路徑上達成一個重要里程碑，也就是從 2026 年第二季開始，我們預期按未來 12 個月滾動基礎計算，將達到調整後 EBITDA 獲利。

Our management objectives that we report every single quarter are the three legs to the stool that help ensure we're on the right path. We're building momentum. Acquisition closes are up. Acquisition contracts, the leading indicator to closes, are also up. In fact, Q1 2026 had over 5,000 contracts. That's the highest quarter of contracts since Q2 2022.

我們每一季都會對外報告的管理目標，是確保我們走在正確道路上的「三隻腳凳」的三條腿。我們正在建立動能。收購成交量上升。收購合約（成交的領先指標）也在上升。事實上，2026 年第一季合約數超過 5,000 份。這是自 2022 年第二季以來合約數最高的一季。

Resale contribution margin has improved every single month since September, and we guided Q2 to the middle of our 5% to 7% targeted CM range. Long-held inventory went from 51% to 10% in two quarters. And we did all of this while holding fixed OpEx down.

轉售貢獻毛利率自 9 月以來每個月都在改善，而我們對 2026 年第二季的指引落在 5% 到 7% 目標 CM 區間的中位數。長期持有庫存占比在兩個季度內從 51% 降到 10%。而我們在做到這些的同時，仍把固定營運費用（OpEx）維持在低位。

The last time acquisition contracts exceeded 5,000 in a quarter, our fixed OpEx was double where it is right now. Yes, double. That's the AI investments and operator empowerment that we talk about every single quarter. That's what's happening here in fixed OpEx. We have made meaningful changes to what is required to run Opendoor 2.0. We are beginning to demonstrate that those changes are durable as the volumes return.

上一次單季收購合約超過 5,000 份時，我們的固定 OpEx 是現在的兩倍。是的，兩倍。這就是我們每一季都在談的 AI 投資與賦能營運人員。這就是固定 OpEx 這裡正在發生的事。我們已對營運 Opendoor 2.0 所需的條件做出實質改變。我們也開始證明，當量能回升時，這些改變具有持久性。

We're clear on our profitability goals, and we will continue to check back in every quarter with updates.

我們對獲利目標很清楚，並將持續在每一季回來更新進展。

Can I add something here? I think Warren Buffett famously said you find out who's swimming without shorts when the tide goes out. I have four kids, they actually sometimes go swimming, and I have to worry about them wearing shorts. I feel for Warren.

我可以在這裡補充一下嗎？我想華倫・巴菲特有句名言：潮水退了，你才知道誰在裸泳。我有四個孩子，他們有時候真的會去游泳，我得擔心他們有沒有穿短褲。我能理解華倫的感受。

But right now, like, the tide is out in housing, right? In the real estate market, the tide is out. Most CEOs will tell you that they wished conditions were friendlier. I'm telling you the opposite.

但現在，房市就像潮水退了，對吧？在房地產市場，潮水正在退去。多數 CEO 會告訴你，他們希望環境更友善。我告訴你的是相反的。

When I took this job, I knew the tide was out. That was the entire point. I didn't take this job because I was hoping macro would turn, and it would bail us out. Like, I wasn't looking for a company of sunshine patriots.

當我接下這份工作時，我就知道潮水是退的。這就是重點所在。我接這份工作不是因為我希望總體環境好轉、來拯救我們。我不是在找一家只在陽光普照時才愛國的公司。

I think Kelly Clarkson famously retweeted Nietzsche, and said, what doesn't kill you actually makes you stronger. I chose hard mode. We choose hard mode because that's what's going to make us stronger.

我想凱莉・克萊森曾經很有名地轉推尼采，並說：殺不死你的，會讓你更強大。我選了困難模式。我們選擇困難模式，因為那會讓我們更強。

Look, we do not need permission from the Fed to put on our shorts to go swimming. Everything we've accomplished so far, everything has been done in the face of an unforgiving macro.

你看，我們不需要聯準會的許可，才能穿上短褲去游泳。到目前為止我們所完成的一切，都是在嚴苛的總體環境下做到的。

I think we've told you how it looks like when we're winning. And some of you are watching this. I think I should tell you what it would look like if we were losing, if we could not do the things Christy said we will do. This is the thing most company CEOs don't do because they're afraid they're going to end up losing, and they want to be able to hide it, but I want you to hold us accountable.

我想我們已經告訴各位，當我們在贏的時候會是什麼樣子。而你們之中有些人正在看著這一切。我想我也應該告訴你，如果我們在輸、如果我們做不到 Christy 說我們會做到的那些事，會是什麼樣子。多數公司的 CEO 不會做這件事，因為他們害怕最後會輸，想要能把它藏起來，但我希望你們能對我們究責。

Here's how you would know. Cohort curves start looking like they did with the purple lines. They would start high. We'd 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. Homes greater than 120 days on the market would go back to what we had in Q4.

你會這樣知道。同批次曲線會開始看起來像紫色線條那樣。它們會從高點開始。我們在過程中會出現巨額虧損。合約會在我們區間的低端，或甚至低於我們區間低端，連續好幾週呈現平台期。在市場上超過120天的房屋會回到我們在第四季時的水準。

If those three things happen, if all those three things happen, then we're not doing what we said we would do, right. It's all about slope, acquisition, inventory health. That's the business. Those three things.

如果這三件事發生，如果這三件事都發生，那我們就沒有在做我們說過要做的事，對吧。一切都關乎斜率、收購、庫存健康度。這就是這門生意。就這三件事。

Look, I don't think any of those three things are going to happen, and I don't think all three of them are going to happen together because we believe we've built a model that works better.

你看，我不認為這三件事中的任何一件會發生，而且我也不認為它們三件會同時發生，因為我們相信我們打造了一個運作得更好的模型。

Faster is the key. We can't ignore the macro. We're not stupid. It will never be our excuse. Good excuses don't make great companies, right? We control our own destiny. We don't need the market to recover. We don't need rates to fall. We don't need perfect conditions. We just need to keep moving more families faster and faster through a machine that's already working.

更快是關鍵。我們不能忽視總體環境。我們不蠢。這永遠不會成為我們的藉口。好的藉口不會造就偉大的公司，對吧？我們掌控自己的命運。我們不需要市場復甦。我們不需要利率下跌。我們不需要完美的條件。我們只需要持續讓更多家庭，以越來越快的速度，通過一台已經在運作的機器。

As I've said before, look, we're not asking you to take our word for it. We're just asking you to watch those three things that Christy talked about.

就像我之前說過的，你看，我們不是要你們只憑我們的一面之詞。我們只是請你們觀察 Christy 談到的那三件事。

Great. Our next question, Andrew L. asks, as you accelerate acquisition velocity, how are you ensuring the underwriting quality remains high and that you won't need to raise equity to fund this expansion?

很好。下一個問題，Andrew L. 問：當你們加速收購速度時，你們如何確保承保品質仍然很高，並且不需要透過增發股權來為這次擴張提供資金？

Thank you for the question, Andrew. It's important to know that we're not accelerating acquisitions by driving, like, blunt spread compression. It is driven by a combination of tailored underwriting that allows us to give really compelling offers to high-quality homes, product expansion through our cash now, more later product, geographic expansion, and just conversion improvements realized from such things as making improvements to the offer page.

謝謝你的問題，Andrew。重要的是要知道，我們加速收購並不是靠那種很粗暴的利差壓縮來推動。這是由多項因素共同驅動：量身打造的承保讓我們能對高品質房屋給出非常有競爭力的報價；透過我們「先拿現金、之後再拿更多」產品進行產品擴張；地理版圖擴張；以及像是改善報價頁面等帶來的轉換率提升。

While we've removed the requirement for a first in-person visit from pre-contract to post-contract, we still perform an in-person inspection before we purchase the home. This sequencing change helped remove friction from the contracting process, and it saved the cost of an in-person inspection for higher intent sellers without compromising our understanding of the home we're about to acquire.

雖然我們已將「首次到場實地看屋」的要求從簽約前移到簽約後，但在我們購買房屋之前，仍然會進行一次到場檢查。這個流程順序的調整，有助於移除簽約流程中的摩擦，並且把到場檢查的成本留給意願更高的賣方，同時不會犧牲我們對即將收購房屋的理解。

What I just described isn't proof that our underwriting quality remains intact and high. The proof is in the cohorts themselves. Our October, November, December, and now January cohorts are each coming in with higher contribution margin, improved margin stability, increased resale velocity compared to their prior year cohorts.

我剛才描述的並不能證明我們的承保品質仍然完整且維持高水準。證據在各批次本身。我們10月、11月、12月，以及現在1月的批次，分別都呈現更高的貢獻毛利、更好的毛利穩定性，且相較於前一年同批次有更快的轉售速度。

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. We also have warrant structures that provide additional capital optionality.

關於資本的問題，我們在收購更多庫存的同時，現金部位其實還增加了，這反映出我們庫存帳簿的基本健康狀況。屋齡較新的房屋、在市天數較短，結構上更容易融資；而且我們有足夠的倉儲融資額度，足以因應甚至超過我們的收購速度與計畫。我們也有權證結構，可提供額外的資本選擇權。

To the extent any future capital decision is made, we expect to be opportunistic rather than necessary, and we will continue to evaluate the capital stack with an eye toward minimizing dilution.

若未來做出任何資本決策，我們預期會是出於把握機會，而非出於必要；我們也會持續檢視資本結構，並以將稀釋降到最低為目標。

Sorry, I'm just going to say a couple of things here. Like, first, there's a persistent myth that to move fast, you have to be sloppy. I just fundamentally reject this.

抱歉，我這裡再補充幾點。首先，有一個長期存在的迷思：要跑得快，就必須很草率。我從根本上不接受這種說法。

Look, there was a rumor when I joined Opendoor that Opendoor was a best buyer of homes with foundation issues. Like, whether or not that was true, it's definitely not true anymore. Today, we use AI to remove this toil we had accrued. We no longer have 11 people touching every single home so the one person that does touch it can actually do their job well, right? That's actually all I want to say about underwriting because I don't want to give away all of our secrets.

你看，我加入 Opendoor 的時候，有個傳聞說 Opendoor 最擅長買的是地基有問題的房子。不管那當時是不是真的，現在肯定不是了。今天，我們用 AI 來移除我們累積下來的這些繁瑣勞務。我們不再需要 11 個人去碰每一間房子，讓那個真正需要接手的人能把工作做好，對吧？關於承保我就說到這裡，因為我不想把我們所有的祕密都透露出去。

On the equity piece, let me add to what Christy said. I said it in my very first earning calls, but I want to repeat it. I despise dilution.

關於股權那一塊，我再補充 Christy 說的。我在我第一次的財報電話會議就說過，但我想再重申一次。我痛恨稀釋。

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.

如果我們發行一股，它只有一個任務：讓我們現有股東手上的每一股都更值錢。我們永遠不會為了延長資金續航而發行股票。那不是我們要做的事。目標是讓 Opendoor 永遠不會處在必須為了生存而募資的境地。

In the history of this company, it has raised way too much money. We're going to stop doing that. The discipline we need going forward is that we're going to fund this business from the cash flow we generate. I'm not interested in, like, building a company that needs a life raft every time. I'm interested in building a ship that actually floats, right? Like what Christy talked about isn't the best case scenario. It's the only way we were going to run this company.

在這家公司歷史上，它募了太多錢。我們要停止這樣做。我們未來需要的紀律是：我們要用我們產生的現金流來為這門生意提供資金。我對那種每次都需要救生圈的公司沒興趣。我有興趣打造一艘真的能浮起來的船，對吧？像 Christy 談到的那樣，並不是最佳情境。那是我們經營這家公司的唯一方式。

Great. Our next question comes from Heejun C., who asks, 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? If so, please provide an estimated timeline or a list of upcoming locations. Thank you.

很好。下一個問題來自 Heejun C.，他問：我對你們目前僅在科羅拉多州提供的 4.99% 房貸方案很感興趣。是否計畫很快把這個方案擴展到其他地區或州？如果是，請提供預估時程或即將推出的地點清單。謝謝。

Well, I mean, look, 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 gimmicks here. We charge what the math allows us to charge, right?

嗯，我的意思是，你看，首先，那不是促銷。我想把這點講清楚。那就是實際利率。我們不在這裡玩利率噱頭。我們收取的是數學允許我們收取的價格，對吧？

Look, mortgage is early. Right now, we're live in Colorado. Loans are doing well without any optimization. Attach rates are above even my, like, most optimistic expectations. I'm not going to give you a launch calendar for every market. 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, and we're rolling this out as fast as we can.

你看，房貸業務還在早期。目前我們在科羅拉多州已上線。在沒有做任何優化的情況下，貸款表現就很好。搭售率甚至高於我最樂觀的預期。我不會給你每個市場的上線行事曆。我們目前正在約 20 多個州進行牌照申請流程。我們預期到第三季末大致會把這個數量翻倍，而且我們正以最快速度推進上線。

But we've gotten some early feedback that I think is helpful. One of the customers told us that our rates blew the other lenders out of water. And I want to, like, talk about our math and why our rates blow other lenders out of water, right. The math here is simple. Big bank lenders take about 340 basis points in revenue per loan.

不過我們收到一些早期回饋，我覺得很有幫助。有位客戶告訴我們，我們的利率把其他放款機構遠遠甩在後面。我想談談我們的數學，以及為什麼我們的利率能把其他放款機構甩開，對吧。這裡的數學很簡單。大型銀行型放款機構每筆貸款大約會拿走 340 個基點的收入。

Most of that is just a toil tax on a borrower, right. It pays for branch offices, loan officers, manual underwriting, paper shuffling, terrible ads, and, like, expensive lunches. We built an AI native mortgage platform from day one. No legacy system, no commission driven sales force, as few humans as possible to get the job done.

其中大部分其實就是向借款人收取的「繁瑣勞務稅」，對吧。那用來支付分行據點、貸款專員、人工承保、紙本文件搬運、糟糕的廣告，以及像是昂貴的午餐。我們從第一天起就打造了一個 AI 原生的房貸平台。沒有舊有系統包袱、沒有以佣金驅動的銷售隊伍、用盡可能少的人力把事情做好。

So we're not just discounting our way to a lower rate, we're actually building our way towards this. That structural advantage means that the regular mortgage on our homes will always be the lowest rate the customers can get, right? Today, our rates are running about 100 basis points below the market average, that translates to about 10% to 15% lower mortgage rates per month. That's the gap, right? Our job is to just chip away at this to make sure that we actually make housing affordable in this country.

所以我們不只是靠打折把利率壓低，我們其實是在「建構」一條通往這個目標的路。這個結構性的優勢意味著，我們房屋的一般房貸利率，永遠會是客戶能拿到的最低利率，對吧？今天，我們的利率大約比市場平均低 100 個基點，換算下來每月房貸利率大約低 10% 到 15%。這就是差距，對吧？我們的工作就是一點一點把這個差距縮小，確保我們真的能讓這個國家的住房變得可負擔。

Great. James M. on Say asks, tokenization of real estate?

很好。Say 上的 James M. 問：不動產的代幣化？

That was the question? That was the whole thing?

問題就這樣？就全部只有這個？

That's the question.

就是這個問題。

Okay. Well, I think this is a question that gets asked frequently, and I have a rule of not announcing product launches before they're ready. I think the worst thing tech companies do is, they mistake software for PowerPoint presentations, and that just sucks. Like, it what makes people hate software companies.

好。我覺得這個問題常常被問到，而我有個原則：產品還沒準備好之前，不會先宣布要推出。我覺得科技公司做得最糟的一件事，就是把軟體誤當成 PowerPoint 簡報，這真的很糟。這就是讓人討厭軟體公司的原因。

Opendoor exists to tilt the world in favor of homeowners, right? Simpler, faster, fairer. You do that by reducing the friction tax. Like, the embedded friction tax in the system today on a given transaction is a double-digit percentage of the home's value, and tokenization is an incredibly important way of reducing this.

Opendoor 的存在，是為了讓世界更偏向屋主，對吧？更簡單、更快速、更公平。你要做到這點，就得降低「摩擦稅」。像是，今天系統裡一筆交易內建的摩擦稅，往往是房屋價值的兩位數百分比，而代幣化是降低這種成本非常重要的方法。

Here's, like, how I think about it, and it's important to be mindful of this. The patterns that we treat today as the natural order of things are usually just the last hack that someone installed on our machines, right? This is when Judd starts rolling his eyes. It matters, so I'm going to talk about my favorite topic, history of money.

我大概是這樣看待這件事的，而且要特別留意這點。我們今天視為理所當然的運作模式，通常只是某個人最後一次在我們的機器上裝上的權宜之計，對吧？這時候 Judd 開始翻白眼了。但這很重要，所以我要聊我最喜歡的主題：貨幣史。

Look, we went from barter to coinage to bills of exchange to checks to ACH to SWIFT, right? It really does feel like we're living, in the future, but the entire system of money that we rely on runs on banks running COBOL software. This is a programming language from 1959. The infrastructure powering our banking system that moves trillions of dollars is older than the Moon landing. We feel like we're in a stable place, but the people who were bartering also felt like they were in a stable place, right?

你看，我們從以物易物到鑄幣、匯票、支票、ACH 到 SWIFT，對吧？我們確實感覺像活在未來，但我們所依賴的整個貨幣系統，其實是建立在銀行運行 COBOL 軟體之上。那是一種 1959 年的程式語言。推動銀行體系、搬動數兆美元的基礎設施，比登月還要老。我們覺得自己處在一個穩定的狀態，但當年以物易物的人也覺得自己很穩定，對吧？

These are not permanent solutions. None of them are permanent because of the following: they all require intermediaries between people to get anything done. That cannot be the end state. On-chain settlement is the first time in the history of money where you don't need permission from other people to move value between two parties. This isn't an incremental improvement. It's the inevitable category end. And within our lifetime, we're going to see what it does, and everything we do today will seem antiquated.

這些都不是永久的解法。沒有一個是永久的，原因如下：它們都需要在人與人之間插入中介，事情才辦得成。那不可能是最終狀態。鏈上結算是貨幣史上第一次，你不需要取得他人的許可，就能在兩方之間移轉價值。這不是漸進式的改良。這是這個類別不可避免的終局。而在我們有生之年，我們會看到它帶來的影響，今天我們做的一切都會顯得過時。

And title is the same story. It's just about 100 years behind. Animals marked their territory physically, and humans mostly have done the same thing for most of history. The real innovation here was in medieval England. We formalized this with a clod of dirt and some witnesses, and now, we have some paperwork.

產權（title）也是同樣的故事。只是大概落後了 100 年。動物會用實體方式標記領地，而人類在大部分歷史中也大致如此。真正的創新發生在中世紀的英格蘭。我們用一塊土和一些見證人把它制度化，然後現在，我們有了一堆文件。

All that has happened between then and now is that some of these are searchable on the internet. That's the entire innovation, that these paper records that live in courthouses are now searchable.

從那時到現在，真正發生的事就是：其中一些資料可以在網路上搜尋了。這就是全部的創新——原本放在法院裡的紙本紀錄，現在變得可搜尋。

Look, the fact that there is a lobbying group defending the current way of doing things is the most reliable evidence that we're due for the next thing. It's like the petition of the candle makers against the sun. When I look at a housing transaction, I find it really hard to imagine that title to the most expensive asset in our lifetime does not live on-chain. It's hard to imagine that we have three transactions doing the same thing, title, insurance, mortgage, and they all have data trapped in silos. These will all move on-chain.

你看，居然有遊說團體在捍衛現行做法，這反而是最可靠的證據：我們該迎來下一個階段了。這就像蠟燭製造商聯名請願反對太陽一樣。當我看一筆房屋交易時，我很難想像：我們一生中最昂貴資產的產權，不會存在鏈上。也很難想像，我們有三個交易在做同一件事——產權、保險、房貸——而且資料都被困在各自的孤島裡。這些都會搬到鏈上。

Look, I'm not announcing any of this today, but we are doing work that's on the green path to this end. Our acquisition of Doma's escrow business is one example, right? We're taking the closing infrastructure of America, building checkout for real estate. And this is not tokenization, but it's clearly the step in the right direction. In that world, title and mortgage and insurance, all of it can move on-chain and this all gets better.

你看，我今天不會宣布任何這些事情，但我們確實正在做一些工作，走在通往這個終局的正確道路上。我們收購 Doma 的代管（escrow）業務就是一個例子，對吧？我們正在接手美國的交割基礎設施，為不動產打造「結帳（checkout）」系統。這不是代幣化，但顯然是朝正確方向邁出的一步。在那個世界裡，產權、房貸、保險——全部都可以上鏈，整體也會變得更好。

Great. Thanks, Kaz. Can't wait for the TED Talk.

很好。謝謝你，Kaz。等不及你的 TED 演講了。

Our next question comes to us from Dae Lee from JPMorgan. Kaz, you've now been leading Opendoor for over half a year and have had time to implement meaningful changes across the product and operations. 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, and what does the data tell you about that's compounding across your markets? Looking ahead, where do you see the biggest opportunity to structurally drive more homes purchased per market without proportionally scaling OpEx?

下一個問題來自摩根大通（JPMorgan）的 Dae Lee。Kaz，你領導 Opendoor 已經超過半年，也有時間在產品與營運上推動有意義的改變。回顧你做過的調整，你認為哪一項具體改變，對目前的賣方轉換率與收購量帶來最可衡量的影響？數據告訴你，這項影響如何在各個市場中持續累積放大？展望未來，你認為在不按比例擴張營運費用（OpEx）的前提下，結構性地提升每個市場購屋量的最大機會在哪裡？

Dae, I want you to know that I know that that was two questions. Let me answer them one at a time.

Dae，我想讓你知道，我知道那其實是兩個問題。我會一個一個回答。

On what's actually moving the numbers, look, I don't think any single thing we shipped is moving anything by itself, but the real structure change in our system is, right? Like, think about the classic sell me this pen story. The old Opendoor was the guy who would say, this pen is amazing. It's so smooth. It's lovely. The guy who'd aggressively show up and give you one choice, right? That was the old cash offer world. We'd show up at your door, give you one choice, say yes or no. That's not how people transact, right? The new Opendoor starts by asking the customer what they want. What do they actually need? What are they worried about?

關於到底是什麼在推動數字變化，你看，我不覺得我們推出的任何單一功能能單獨推動一切，但系統真正的結構性改變在於——對吧？就像那個經典的「把這支筆賣給我」的故事。以前的 Opendoor 就像那個人會說：這支筆太棒了。寫起來很滑順。很漂亮。然後很強勢地出現，只給你一個選項，對吧？那就是以前的現金報價世界。我們會到你家門口，給你一個選擇，讓你回答是或不是。但人們不是這樣交易的，對吧？新的 Opendoor 會先問客戶想要什麼。他們真正需要的是什麼？他們在擔心什麼？

How much cash they want upfront, what do they want later, what timeline they want. Cash now, more later isn't a single offer. It actually allows the customer to change Opendoor's business logic so that it works for them, right?

他們想先拿多少現金、之後想要什麼、想要什麼時間表。「現在拿現金、之後拿更多」不是單一報價。它其實讓客戶可以調整 Opendoor 的商業邏輯，讓它為他們服務，對吧？

The new offer page also does the same thing. It is the digital equivalent of sitting down with someone and explaining to them the realities of their neighborhood, their home, instead of just flashing a headline number, right? We want the customer to have the full picture and make the life choice that is best for them, and we want to be helpful in that process.

新的報價頁面也在做同樣的事。它相當於數位版的「坐下來跟你解釋」：你所在社區、你的房子實際面臨的情況，而不是只閃過一個醒目的數字，對吧？我們希望客戶看到完整全貌，做出對他們最好的生活選擇，而我們希望在這個過程中提供協助。

And most people think that in order to do that, you need a human at a kitchen table. I think that's just wrong. Most people just want the information themselves, so they can decide for themselves what's best for themselves. That's the shift. That's driving the conversion improvement.

而大多數人以為要做到這點，就需要一個人在廚房餐桌旁面對面解釋。我覺得那是錯的。大多數人只是想自己拿到資訊，這樣他們就能自己決定什麼對自己最好。這就是轉變。這正在推動轉換率的提升。

We also used to believe we'd need boots on the ground everywhere we had homes. I actually insisted on launching every state in the lower 48 because I wanted to test this hypothesis. 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 zero employees in South Dakota. That actually helps a lot.

我們過去也一直以為，在我們有房源的每個地方都需要有前線人員駐點。事實上，我曾堅持要在美國本土下方48州全部上線，因為我想驗證這個假設。結果證明，只要你有良好的承保模型、好的產品，以及良好的合作夥伴網路，你就能在任何地方買房。例如，我們這週在南達科他州完成了一筆房屋成交，而我們在南達科他州一名員工都沒有。這其實幫助很大。

To your second question on OpEx, I mean, I think we've already answered a lot of this. But the same machine does both of these things. More offer types mean more sellers. More sellers per market means we can have more transactions without adding headcount city by city.

關於你第二個有關營運費用（OpEx）的問題，我的意思是，我想我們其實已經回答了很多。但同一套機器同時能做這兩件事。更多的出價類型意味著更多的賣方。每個市場有更多賣方，意味著我們可以在不需要逐城增加人頭的情況下，完成更多交易。

Look, the big prize obviously is the tens of millions of people who want to move who can't, right? Between supply and demand, there's friction, right? If you reduce friction, you move both supply and demand lines. I actually saw this every day at Shopify. We made entrepreneurship easier, therefore we created more entrepreneurs. The same dynamic is true in housing.

你看，最大的獎賞顯然是那數千萬想搬家卻搬不了的人，對吧？在供給與需求之間存在摩擦，對吧？如果你降低摩擦，你就會同時推動供給與需求兩條曲線。我在 Shopify 每天都看到這點。我們讓創業變得更容易，因此創造了更多創業者。同樣的動態也適用於住房市場。

As we make things easier in buying a house, selling a house, mortgage, title, and eventually insurance, all of this will increase the demand and increase the supply, and none of this requires, like, significant incremental headcount.

當我們讓買房、賣房、房貸、產權（title），以及最終的保險等流程變得更容易時，這一切都會提升需求並增加供給，而這些都不需要，例如，大幅新增的人力。

Now, like look, none of this works if the underlying engine isn't good, but I think we've shown you that we're no longer peanut butter spread across cohorts, and we've shown you we have now four cohorts of data, and Q1 was our largest contract quarter in years.

現在，你看，如果底層引擎不好，這些都行不通；但我認為我們已經向你們展示，我們不再像把花生醬薄薄抹在各個 cohort 上那樣分散資源；我們也向你們展示，我們現在有四個 cohort 的數據，而且第一季是我們多年來合約量最大的季度。

The last time we had this many homes in contract, our fixed OpEx was twice as high. I think that answers your second question.

上一次我們有這麼多房屋處於簽約中（in contract）時，我們的固定營運費用（OpEx）是現在的兩倍。我想這回答了你的第二個問題。

Great. Andrew from Citizens is curious to help us understand a little bit more about seasonality kind of through the balance of the year.

很好。Citizens 的 Andrew 想請我們協助更進一步理解一下，今年剩餘時間的季節性大致會是怎樣。

Sure. I'm happy to provide some color on seasonality, and I'm sure Kaz will be happy to add something as well.

當然。我很樂意就季節性提供一些補充說明，我也相信 Kaz 也會很樂意再補充一些。

Each quarter we provide a series of macro charts, and those charts show a consistent pattern. In every macro, strong macro, neutral macro, challenged macro, one thing remains the same. It's the seasonal pattern. They present themselves year after year.

每個季度我們都會提供一系列宏觀圖表，而這些圖表呈現出一致的模式。在各種宏觀環境下——強勁的宏觀、中性的宏觀、承壓的宏觀——有一件事始終不變。那就是季節性模式。它年復一年地出現。

Macro changes the level of the curve, and seasonality is the shape of the curve. The selling season kicks off shortly after the Super Bowl, peaks in early summer, tapers through fall and bottoms out in December. This affects our resale velocity, which is considered in our spreads and therefore impacts our acquisition cadence.

宏觀因素改變的是曲線的高度，而季節性決定的是曲線的形狀。銷售旺季通常在超級盃之後不久開始，於初夏達到高峰，秋季逐步趨緩，並在12月觸底。這會影響我們的轉售速度（resale velocity），而轉售速度會被納入我們的價差（spreads）考量，進而影響我們的收購節奏（acquisition cadence）。

Days on market lengthens in the back half, margins compress in Q4. 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.

在下半年，市場在售天數（days on market）會拉長，第四季的利潤率會受到擠壓。我們的收購節奏與市場轉售活動呈反向關係。在春末我們收購較少，因為那代表我們會在需求較弱時賣出；而我們會在整個秋季累積庫存，以迎接春季銷售旺季。

You'll now see seasonality more reflected in our estimates on accountable.opendoor.com. We've updated our projected acquisition range with the shape easing through spring and summer and building through the fall.

你們現在會在 accountable.opendoor.com 上看到，我們的預估更能反映季節性。我們已更新預計的收購區間：在春夏期間形狀趨緩，並在秋季逐步建立。

Thanks. I will add something. Look, seasonality is just like gravity. It's like a rule of nature. You don't blame gravity. If you try to fight it, you tend to lose. We know how to fly planes. We don't do it by fighting gravity. We just build math to fly them, right? And while we can't flatten the curve entirely, we can collapse the impact over time. That's what we're working on.

謝謝。我再補充一點。你看，季節性就像重力一樣。它就像自然法則。你不會怪重力。如果你試圖對抗它，通常會輸。我們知道怎麼讓飛機飛。我們不是靠對抗重力來做到的。我們只是建立數學模型讓它飛起來，對吧？雖然我們無法把曲線完全拉平，但我們可以隨時間推移縮小其影響。這就是我們正在努力的方向。

Opendoor is like a retail like Walmart, like Home Depot, like Amazon, like Shopify. These retailers have known seasonality, like obviously Q4 is a better quarter for them because of Christmas. And Q1 numbers are always lower than Q4. No one would argue that Walmart's strategy has failed because January sales were lower than December sales. That would just be, like, insane.

Opendoor 就像零售商一樣，像 Walmart、像 Home Depot、像 Amazon、像 Shopify。這些零售商都知道季節性，例如很明顯第四季因為聖誕節而是更好的季度。而第一季的數字總是低於第四季。沒有人會因為一月銷售低於十二月，就主張 Walmart 的策略失敗了。那就太荒謬了。

The shape is just the shape. The same general seasonal shape that shows up in housing in 2021 when the market was on fire, in 2022 when the rates spiked, and in 2025 when de-listings hit record highs, that's the shape. They're different macroenvironments, the same calendar, since Pope Gregory invented it, I guess.

形狀就是形狀。同樣的整體季節性形狀，在2021年市場火熱時會出現、在2022年利率飆升時會出現、在2025年撤牌（de-listings）創新高時也會出現——那就是那個形狀。宏觀環境不同，但日曆相同；自從教宗格里高利發明它以來，我想就是如此。

Opendoor knows more about the shape of the curve than almost anyone else in the world, and we shape our underwriting engine around it. We underwrite homes based on when we plan to sell them. That's what our underwriting engine does, and I think it is working better and better every day.

Opendoor 對曲線形狀的理解，幾乎比世界上任何人都更深入，而我們也圍繞它來調整我們的承保引擎。我們在承保房屋時，會以我們計畫何時出售為基礎。這就是我們承保引擎在做的事，而我認為它每天都運作得越來越好。

Look, I want to end this answer with what I said earlier. We committed to being ANI profitable on a go-forward 12-month basis at the end of this year. Hard macro or not, we will do that. Our floor model assumes this hard macro will continue. If there's an interest rate cut or the macro improves, our floor will be higher.

你看，我想用我先前說過的話來結束這個回答。我們承諾在今年年底，於未來12個月的前瞻基礎上達成 ANI 獲利。不管宏觀環境艱難與否，我們都會做到。我們的底線模型（floor model）假設這種艱難的宏觀環境會持續。如果降息或宏觀改善，我們的底線就會更高。

I think we're running out of time, so let me just close what this is, okay.

我想我們時間快不夠了，所以讓我簡單做個收尾，好嗎。

Look, we're not asking you to believe in vibe here. We're asking you to watch the scoreboard. The cohort slope, acquisition contracts, inventory health. That's it. Those are the tells, right.

你看，我們不是要你相信某種氛圍（vibe）。我們要你看記分板。cohort 斜率、收購合約、庫存健康度。就這些。這些就是關鍵訊號，對吧。

If we keep moving the way we moved this quarter, then the machine is doing exactly what we said it would do. The market didn't bail us out here. Rates didn't save us. The team just did the work. They rebuilt the company's operating system. They shipped products. They cleaned up the book. They grew contracts, and they did it way more efficiently than anyone thought we could do it.

如果我們持續以本季度的方式前進，那這台機器就正在精準地做我們說它會做的事。市場並沒有在這裡救我們。利率也沒有拯救我們。團隊只是把工作做完了。他們重建了公司的作業系統。他們推出了產品。他們整理了帳本。他們成長了合約量，而且他們做得比任何人想像我們能做到的都更有效率。

That doesn't just give me optimism. It gives me confidence. We will have a lot left to prove, and we always will. When we reach profitability, the next part is how much? It just won't stop, right? We're going to keep shipping. We're going to keep showing you the data, and we're going to keep moving faster because families matter.

這不只是讓我感到樂觀。這讓我有信心。我們還有很多需要證明的，而我們永遠都會如此。當我們達到獲利後，下一個問題就是：能有多少？它不會停止，對吧？我們會持續交付。我們會持續向你們展示數據，而且我們會持續加快腳步，因為家庭很重要。

Okay. That's it. Thank you, thank you, and see you all next quarter.

好。就到這裡。謝謝、謝謝，我們下個季度見。