KE Holdings Inc. (BEKE) Earnings Call Transcript
November 16, 2026
Earnings Call Speaker Segments
Good day, everyone, and welcome to the KE Holdings Third Quarter 2026 Earnings Conference Call. This is Siting Li, IR Director at KE Holdings. I want to let you know that everything on today's call, both management's prepared remarks and the Q&A session, will be conducted in Chinese. An English simultaneous interpretation line will be available throughout the call. If you'd like to follow along in Chinese, please dial in to the Chinese language line. <<operator_script>> This conference call is being recorded. The company's financial and operating results came out in the press release earlier today and have been posted to the company's IR website. Joining us today are Mr. Stanley Peng, our Co-Founder, Chairman and CEO, and Mr. Tao Xu, our Executive Director and CFO. Mr. Xu will walk you through our business updates and financial performance. After that, Mr. Peng will discuss the company's consumer-centric transformation practices and cultural renewal. Before we go further, please refer to the safe harbor statement in our earnings press release, which applies to this call, since we will be making forward-looking statements. Also note that Baker's earnings press release and this conference call include discussions of unaudited GAAP financial information as well as unaudited non-GAAP financial measures. Please see the company's press release, which contains a reconciliation of the unaudited non-GAAP measures to comparable GAAP measures. Finally, unless otherwise stated, all figures mentioned on this conference call are in RMB. Certain statistical and other information relating to the industry in which the company is engaged that will be mentioned on this call has been obtained from various publicly available official or official sources. Neither the company nor any of its representatives has independently verified such data, which may involve a number of assumptions and limitations, and you are cautioned not to give undue weight to such information and estimates. For today's call, management will use Chinese as the main language. Please note that the English translation is for convenience purposes only; in case of any discrepancy, management's original statements in Chinese will prevail. With that, I will turn the call over to our CFO, Mr. Xu Tao.
Please continue. Thanks, Siting. Hi, everyone, and thanks for being with us on the Q3 2026 earnings call. I'll begin with a quick look at how we performed in Q3. GTV kept rising compared with a year earlier. Revenue was down modestly year-over-year, yet we still posted robust profit growth that far exceeded both GTV and revenue gains. That shows once more that our better cost structure and operating efficiency keep feeding through into stronger profitability. GTV came in at RMB 769.67 billion in Q3, a rise of 4.5% year-over-year, while net revenue was RMB 20.99 billion, a fall of 8.9% year-over-year. The revenue drop came chiefly from revenue recognition packs, since our home rental business keeps moving toward native-based products, along with earlier adjustments in home renovation. Non-GAAP net income was RMB 2.605 billion, an increase of 102.6% year-over-year, and the non-GAAP net margin was 6.8%, up 12.4% year-over-year and still at a historically high level. These profit gains were mainly due to above-market operating performance from cost structure improvements and strict expense explained. Contribution margins rose year-over-year in every major business line, pushing gross margin to 27.7%, up 6.3% a year, while GAAP operating expenses dropped 22.4% year-over-year, and both the higher gross margin and the lower operating expenses drove profitability. Next, I'll walk you through the business lines. First, existing home transaction services: both GTV and revenue increased year-over-year in Q3, and the quarter-over-quarter performance reflected seasonality. GTV was RMB 520.95 billion, up 3% year-over-year and down 17.3% quarter-over-quarter. Net revenues were RMB 6.1 billion, up 1.8% year-over-year and down 13.2% quarter-over-quarter. Non-linear GTV platform revenue reached RMB 1.52 billion, up over-year and down 12.5% quarter-over-quarter, beating the segment's overall revenue. With network size holding steady, we keep refining operations to raise output per store and create efficiency. The contribution margin for existing homes was 42.5%, up 3.6 percentage points year-over-year, mainly on lower personal costs. On a quarter-over-quarter basis, contribution margin fell 3.6 percentage points because revenue was lower, while the cost structure stayed fairly stable. Now for our new home transaction services. Our new home business kept outperforming the market. Q3 GTV was RMB 206.76 billion, up 5.3% year-over-year and down 20% quarter-over-quarter. Net revenues were RMB 6.83 billion, up 29% year-over-year and down 23.6% quarter-over-quarter. GTV beat revenue both year-over-year and quarter-over-quarter, largely because we proactively optimized the project mix, raising the share of customer-centric new home offerings. New home contribution margin was 27.9%, up 3.8 percentage points year-over-year, driven by optimizing the cost structure through refund operations. Quarter-over-quarter, contribution margin dropped 0.9 percentage points, mainly because we proactively adjusted channel incentives and moderately raised commission sharing in Tusan cities to support partner operations, strengthen network vitality, and ensure long-term health for the platform and stores. Third, our home renovation and furnishing business. Natural revenues in Q3 were RMB 3.31 billion, down 22.9% year-over-year and up 3.9% quarter-over-quarter. Contract value actually returned to year-over-year growth this quarter. The year-over-year revenue decline reflects earlier contract value drops and the lag between signing and revenue recognition. Contribution margin came in at 40.2%, an increase of 8.2 percentage points from a year earlier and 0.6 percentage points versus the prior quarter. The year-over-year improvement was driven by lower material cost ratios achieved through centralized procurement, and the sequential gain came from reduced personnel costs thanks to higher operating efficiency. Fourth, home rental services. Net revenues for Q3 reached RMB 4.27 billion, a decline of 25.4% year-over-year and 11.6% quarter-over-quarter, largely reflecting the continuing transition to net basis revenue recognition for carrier free rent. Although this lowers accounting reported revenue, units under management expanded quickly. As of the end of Q3, units under management surpassed 330,000 units, roughly 25% higher year-over-year. Net basis products now represent more than 65%, versus over 24% a year ago. Contribution margin stood at 16.6%, up 7.9 percentage points year-over-year and 1.3 percentage points quarter-over-quarter. The year-over-year gain reflected a shift in product mix toward anatase recognition as well as greater personnel productivity. The sequential improvement was due to a larger net basis product share. Fifth, our emerging and other services. Q3 < > revenues totaled RMB 470 million, rising 19.8% year-over-year while falling 13.2% quarter-over-quarter. Let's now turn to costs, operating expenses and profitability. Store-related costs were RMB 500 million in Q3, 24% lower year-over-year and 10.7% lower quarter-over-quarter, helped by optimized LNG rental costs and network restructuring. Total GAAP operating expenses amounted to RMB 3.36 billion in Q3, down 22.4% year-over-year and 15.7% quarter-over-quarter. The year-over-year improvement reflects better organizational efficiency and optimized marketing spend. The sequential declines reflect continued expense discipline and the absence of last quarter's one-off new home provision. To be specific, G&A expenses were RMB 1.6 billion, down 14.2% year-over-year on lower personnel costs and strict expense discipline, including cuts to travel and meeting expenses, and down 21.4% quarter-over-quarter because of the high base from last quarter's credit loss provision for new homes. Sales and marketing expenses were RMB 1.25 billion, down 27.5% year-over-year thanks to optimized advertising, promotion and sales costs, and down 10.7% quarter-over-quarter owing to last quarter's one-off expenses and lower share-based compensation. R&D expenses were RMB 510 million, down 22% year-over-year from optimized R&D personnel costs, and down 7.8% quarter-over-quarter from lower personnel and technical service expenses. Turning to profitability. Q3 GAAP income from operations was RMB 2.463 billion, up 305.1% year-over-year and down 18.6% quarter-over-quarter, with an operating margin of 11.7%, up 9.1 percentage points year-over-year and down 0.6 percentage points quarter-over-quarter. Adjusted income from operations was RMB 2.883 billion, up 114.9% year-over-year and down 19.7% quarter-over-quarter, with an adjusted operating margin of about 13.7%, up 8.6 percentage points year-over-year and down 0.9 percentage points quarter-over-quarter. The year-over-year improvement stemmed from higher gross margins and lower operating expense ratios, while the sequential declines reflected a seasonal revenue decline. GAAP net revenue was RMB 2.195 billion, up 193.9% year-over-year and down 16.4% quarter-over-quarter. Non-GAAP net revenue was RMB 2.6 billion, up 102.6% year-over-year and down 18.2% quarter-over-quarter. Delivering more than double year-over-year non-GAAP net income growth highlights our strong property elasticity, driven by above-market performance and a healthy cost expense at baseline. Finally, let's cover cash flow, the balance sheet, and returns to shareholders. Operating cash inflow for Q3 came to RMB 3.28 billion. Our new home accounts receivable term per day averaged roughly 49 days < > which shows risk management working effectively. At the close of Q3, the broader cash balance was approximately RMB 71.8 billion, and that figure leaves out customer escrow refund. This ample liquidity bolsters our resilience to risk while also backing business development and shareholder returns. During Q3, we bought back roughly USD 160 million worth of shares. So on a year-to-date basis through the end of Q3, total buybacks have come to approximately USD 610 million, which is about 3.2% of total shares as of December 31, 2025. From the time we launched the buyback program in September 2022 through the end of Q3, cumulative repurchases have amounted to roughly USD 6.16 billion, equal to about 15.6% of the total shares before the program is launched. To sum up, across the past 3 quarters we have consistently confirmed that our refinery operations work effectively, along with the structure of optimization and expense management, and this has translated into tangible profit improvement. Looking ahead, we stay consumer-centric, and while preserving cost efficiency, we will steer our resources toward areas that enhance customer experience, service efficiency and long-term value; for new investments, we will take a paced approach with rapid validation against actual operating results, dynamically adjusting pacing and resource allocation to drive sustainable growth and shareholder value. Thank you. Now I'll hand it over to Stanley, our Chairman and CEO.
Thank you, Mr. Xu. In Q3, our operating results carried forward the good momentum we've seen throughout the year. That gain comes from market activity picking back up along with substantial improvements in how efficiently we operate, even as new home adjustments carry on and store and agent counts stay flat to slightly lower. Given that backdrop, transactions combining existing and new homes rose more than 10% year-over-year; existing units were up 14%; new home GTV climbed 5.3%; volume per agent increased by more than 30%; and average connected historic volume was up almost 17%; in home renovation, conversion rates stayed level versus a year ago, and contracted value went back to year-over-year growth. Managed rental units were up 25% year-over-year, and rental inventory per manager increased more than 20%; so what do all these figures tell us? Where is this growth coming from? It comes, in fact, from productivity per agent and efficiency per store. It does not come from simply adding headcount or expanding stores; our consumer-centric transformation has now been running for more than 200 days. To tackle slow turnover of older homes, we are pursuing multidimensional initiatives spanning layout redesign and renovation folded into the traditional transaction flow, refreshing rother, our efforts such as refreshing out-home in batches so homes sell faster. We've got a smooth, carefree ramp here, and that makes these homes more pleasant to live in and simpler to trade. At the same time, we also provided subsidies for aging inventory units — those are a safety investment rather than a growth play, aimed at preventing hazards and reducing risk costs. Behind these efficiency gains, unlocking existing home inventory and renewing old urban housing are turning into a new growth area in China; in fact, this is a structural opportunity where we're only at the beginning. And looking at the years ahead, there's still a great deal of room for us to improve. So today, I'd like to share two points. To start, how does transformation come about through time practice, and second, how does culture give lasting direction? So on the first point, transformation is entering certain client practices. It isn't dreamed up in meeting rooms. It emerges from real front-line problems, so let me walk you through three examples. Example one, the home adviser. In existing home stores, we assist consumers in answering the question of how they will live there during < >. In the past, property tours concentrated purely on community price and floor plan layout. Yet once billings were upgraded, consumer feedback changed. They typically ask, "How can this go? Is it possible for two bedrooms to become three?" And what about using the kids at various ages? Those questions really highlight the first thing to address — not merely the house itself, but the amount that purchasers are unsure about regarding how they'll live there or what could be modified afterward. So in my view, the know-how about property and the know-how about living space function separately, and the pivotal moment arrived at Acumen, the Chief Director, Matteo at Manager Galpin Child. They selected common floor plans, letting agent Caseta, matching buyers' requirements and pain points. Take, for instance, a child's age, what matters most, identifying the constraint. And designers, along with reworking options and approaches to change the space and how to execute the design, so all of that can meet household needs more effectively. On site, we'll explain these things in everyday terms, tying them back to the consumers, to the home renovation, and to the life ahead. So the in-depth, on-site programs — those three points all connect. The team got behind these narratives, and now they have their first group of agents who can actually talk about renovation and houses. Of 29 consumers, 13 added a redesign, and that helped them make the decision to buy. By September, predesign in renovation intent had risen from 18% to 65%. And with existing home renovation attachment rates coming in almost 3 percentage points above the bidding average. So what really changed here isn't an additional step — it's an added layer of expertise that's there at the moment of decision. Looking ahead, templates for frequently used layouts are being standardized so agents can handle receivables while giving tours, and the incentive structure is likewise being adjusted to reward designers who work together. This transformation therefore involves three levels. The first is that Beike shifts away from listings toward solutions for living; the second is that different disciplines work jointly around a single consumer; the third is that companies generate shared value around what consumers need, producing value and partnering with one another. That gives buyers a promising future: they ought to prioritize utilities, making renovation scenarios clear before they purchase. This amounts to a different approach to service, and it is simultaneously establishing a fresh purchasing habit among consumers. The second example comes from Wuhan. It concerns roughly new homes going from < > to a full-service solution. Historically, our model was that of a channel. Candid sales mandates were managed by developers, while we sourced specific, shorthold and collective shared commissions. In an < > new home had Juan and operation had Shanghai, the shift went from channel distribution toward full service integration, and projects in the designed and economic development zone are somewhat puzzling since Wuhan performs strongly on other measures. So what does the innovation consist of? And while they were on site, they financed the auctions close by and raised the road, and it comes to 300 units sold and < >, questioning whether selling these units might mislead consumers. But after going to stores and talking with agents, < > shifted quite a bit. So what mattered wasn't figuring out how to look at days—it was who to put forward. That means step one isn't breaking down metrics; it's doing consumer research. Through more than 100 interviews, the team pinpointed factory-worker buyers and first-time homebuyers, and charted exactly where they are. Following that consumer research comes their margin plan: they work out each category for the consumers—for instance, the down payment, mortgage, rent, renovation—and they also weigh every payment and evaluate every auction, whether yours, an existing one, or an event. So how do the various auction formats differ, and what does a sound final decision look like when evaluating the layout along with how the surroundings reverberate? Should there be no < > coming from the overpass, and since we have the partners in Chengdu who dispatched installers to put some loss into the show flat and to find — to gather feedback from the commerce side and directed us that the noise has to be dealt with. Each person ought to notice how conditions shift between various forces, whether windows are shut or not; so afterward, upon watching an agent close — bringing the glass store to a halt in order to conceal noise, we instructed the team < > and in fact, while we are on location the default remains closing whatever the expense, and that is not the shift, since this initiative is merely the beginning of the exploration. And with additional assistance from purchasers, observe whether the home suits their family over the next five to ten years. And after that assist them in deciding. So before we actually take a family and look at a nearby factory—they purchased a house but used up their savings, and they were planning to put off the renovation for a year, yet school couldn't wait, rent kept accruing, and the new home sat empty. < > undue in a renovation plan period with loans kept monthly payments affordable and week a year into start now. And they can live in the house now. So being able to afford the purchase doesn't just solve the cost—being able to afford the renovation and moving in solves the family's problems. The result is that < > to 179 subscriptions and 1 deal per 4 visits.
The norm in the business is one out of fifteen; about a third choose group renovation, meaning renovation for the new homes. What began as a single commission has expanded into agency fees, marketing incentives, and group renovation synergy, all tracked through unit economics. Each revenue stream, then, comes from giving consumers certainty; in my view that metric shows meaningful gains; and this method is being copied, as you'd expect, in other projects around the country, because it offers a repeatable way to find solutions. We presented two cases on strengthening frontline service and on becoming more professional; for case three, I'd like to walk you through — let's discuss the plans. Let me have our store design lead walk through the twelve renovated stores in the notion district. The manager is <>, the director is <>, and store design in the past was when companies rolled out nationwide and produced business every 2 to 3 days, which was efficient but rested on the fourth premise that all stores resi the exact same consumer. So actually, this is a fourth premise. Google stores serve families with school enrollment needs. He noted existing maps are drawn by supermarkets, but parents care more about the schools, and the school zone follows streets, not the supermarket. He requested a dedicated school map, even sketching it himself. Wait — I need to re-check my output: I reused several long runs verbatim ("store design in the past was when companies rolled out nationwide and produced business every 2 to 3 days, which was efficient but rested on the fourth premise that all stores resi the exact same consumer. So actually, this is a fourth premise. Google stores serve families with school enrollment needs." and "existing maps are drawn by supermarkets, but parents care more about the schools, and the school zone follows streets, not the supermarket. He requested a dedicated school map, even sketching it himself."). That violates the constraint. Let me redo it properly. Rewritten excerpt: First, our store design lead will take you through the twelve renovated locations in the notion district. The manager there is <>, the director is <>, and in the past, store design worked like this: companies expanded across the country and generated business every 2 to 3 days — efficient, yes, but it depended on that fourth premise holding that every store served an identical consumer. So in fact, that constitutes a fourth premise. Stores under Google serve households needing school enrollment. He observed that today's maps come from supermarkets, yet what parents value more is schools, and school zones trace streets rather than supermarket locations. He asked for a school-specific map and even drew a sketch of one himself. Thus, he brought the schools into the picture, along with the primary and middle schools within his assigned area. To took it a step further. He went to these stores in the role of a parent and remained for an entire 90-minute briefing. While that briefing was underway, he noticed two things. First, the agents came across as much more professional than he had anticipated, handling almost every question from parents without a hitch—covering schooling as well as streets and houses too. Second, the agents kept flipping back and forth among multiple pages. As a result, the agents did not even perceive this structure. That said, it’s quite obvious when you look at it. So when it comes to switching content, being able to correct those pitches on a secondary display would make things flow better and be simpler to track. Their map design, how information is shown, and especially the concrete empirical support behind it — the designs were tailored to actual requirements. At first it was intended for 2 to 3 families, and afterward a photo from 2 revealed 14 people, then 20, and 21 days after that. As a result, the initial plan was abandoned. That area was well known as an adaptable open space used for meetings or group briefings, and it can be opened up for receptions; it can also host briefings for more than 10 families, and we provide plenty of convenient amenities for the community. Take, for instance, what we did with the printer: we moved it over to a desk with seating, right next to community services and the building 4 agencies, and we raised the chairs a bit since a wet market is close by, which makes it simpler for seniors to use whether they sit or stand; say, early on at the bazaar, the elderly regularly walk past, and they ask whether anyone is sitting in those chairs; so in fact quite a few older folks took those seats and chatted with one another. That scenario, then, is meeting reality. The school map was folded into the digital world display, and we now have the synthetic map — a static map that carries live data while agents converse about particular zones. And we — the coming version aims at one screen that connects all business lines: rental existing, new home renovation and community service. So every one of them can show up on a single display; and delivery isn't the finish line. Two or 3 months afterward, we return to check visibility and to polish things further. These cases, then, also give us 2 lessons. First, specialist functions are moving. Experts must move onto the front line, understand workflows and deliver solutions alongside partners and keep an eye on results once launched. Next, the role of our stores is changing as well. They ought to serve not just as uniform work areas, but as venues that display expertise, centers where customers engage with us, and community-service outposts linking the capabilities of our Body 3 Wings, all of which can be tied together inside such a space. So underlying these shifts, the logic of growth is turning from adding a single extra customer and signing one additional deal toward gaining a deep grasp of one customer and taking care of a whole family. That prompted us to revisit four principles. Redefining quality service. Judging the work requires seeing it through the lens of customers' real circumstances. Out of the 29 customers we followed up with, 13 told us the redesign plans helped them make the decision to purchase. So the metric doesn't really mean higher conversion; what it means is that these consumers, the buyers, have more clarity when they're deciding. And when Jon asked if anyone was sitting in the chairs at the entrance and heard yes — that single word was more real, more moving, than any survey result. So good service isn't about ticking boxes. It's really about giving consumers genuine clarity. Second, it's about redefining what the professional services we provide actually are. Sometimes, you know, customers' problems represent only what they're able to put into words. Our professionalism—our expertise—is aimed at helping them grasp the full picture. Take one case: Li, a designer, and the agent toured the homes side by side. The agents thus understand where the pain points lie, while the designers understand the distinctive opportunities. By joining these two skill sets, buyers come to see how the property might connect with how they live. With the Wuhan projects, buyers were fearful of—or disliked—the highway noise. Instead of dodging the issue, we put it plainly before them: how the acoustics change across floors, and with sealed windows as opposed to without—and then the final decision rests with them. Expertise, then, isn't about showing how much we know; it's about offering something consumers can grasp, apply, and depend on. Third, it's about redefining the support our organization provides. When a top-tier employee or partner succeeds, we should ask whether an ordinary partner could get the same results under identical conditions. Take the success at Dah — it can be replicated and regulated, since it turned into a layout library and a collaboration rulebook. The solution at the Wuhan project can likewise be replicated, because reviews and scripts became frontline tools that anyone can use. The success at KUO store, too, produces a repeatable methodology. AI isn't merely a matter of deploying some new tool—it's really aimed at that very problem: turning top-tier expertise into scalable instruments. When a large number of people can tap into that expertise, it becomes an institutional capability. The question is no longer whether an organization has elite talent on staff, but whether it lets a broader set of people—employees or partners—perform at a high level. From there, we need to rethink how we view the execution of our changes; so far, we have three lessons. Lesson one: take things away before you add. Across the three cases we mentioned, things worked because the organization eliminated legacy KPIs. Take Min and Suni District in Beijing—there, for instance, we dropped the market share metric, that KPI. And for the Wuhan one as well, we simply set no KPI. That lets our partners serve consumers' real needs. Subtraction, therefore, is undervalued.
But at times, the market zeroes in on which new offerings we roll out, when genuine transformation actually begins with pulling up old roots, which gives us something like a floor where new possibilities can grow. Takeaway number two is iterating through practice rather than aiming for upfront perfection; the Google store, for instance, was redesigned after observing 20 people in the photo and adjusted a month later. Then it takes yet another round of adjustment once it has been in use for 2 to 3 months. That also applies to the projects in Wuhan. We've been making adjustments all along; so when we do our reviews, we split out the steps already completed, the shifts happening within the company, and what we've achieved from a business perspective. Third, we need to strike a balance between changes—meaning how quickly versus how gradually they happen. The Kolo store worked out well since they had six months these days. If we push these near-term metrics harder, innovation will be destroyed. That's the reason we created a safeguard for this gradual pace, allowing practice sufficient time to observe, giving partners room to develop, and we must gather plenty of data to confirm our principle. So I realize this attracts attention, because the outside market wants results quickly. On our side, these new offerings need sufficient time to develop, so striking the right balance matters. To start with, we should safeguard a solid financial runway while also ensuring TIM's advances are visible to outside observers. Let me now move to my second topic—organizational culture—and the way we establish enduring direction. Transformation is fundamentally a matter of redefining value; as shifts take place, we must remain united around our principles, around whom we serve and the worth we deliver. When tensions emerge, what guides our choice? On November 11, we marked our 25th anniversary. We introduced fresh culture principles for our mission—they represent service providers and improved living; the vision is to support every stage of home with expertise. We had four values. The first was to take on what is hard—challenging but right. The second was to keep consumers at the center. The third was to help professional service providers succeed. The fourth was to seek shared success through working together. We also introduced some company spirits: authentic, open, resilient, and optimistic. One of these was actually new—let professional service providers win. The word "let" highlights the platform's responsibility. The three cases I mentioned all reflect this: the designer Luton joined viewing since collaboration was rewarded; the partner was there to discuss the noise because we set no rigid metrics; and the chairs at Gola were designed successfully to comfort seniors because Jon had time to sit in the store. For the platform, we need to build a system supporting professional growth across 5 directions. Number one, rewards. We need to allow professionals to earn commissions directly from each closed deal. Number 2, organization; we need collaboration across dis plans to create complete consumer value. Third, institutions. We need to set up a credible professional certification that lends credibility to the professional participants involved. Fourth, visibility. Experts ought to be discoverable by customers and chosen via content and branding efforts. Number five: technology. We are bringing in AI so that high-level or elite expertise becomes trainable, scalable methods — meaning more individuals can pick up and acquire those capabilities, with this fresh cultural shift guiding things rather than relying on particular people. Each of these decisions carries a benchmark offering long-term certainty for changes. So, having offered several of my practice-related observations, let me turn back to the business. Figures for Q3 reveal a rising count of consumers selecting our services, and we are boosting the platform's productivity. These practices really give us confidence about what's ahead. As we assess whether customer issues are resolved more effectively, whether professionals encounter fewer repetitive chores, and whether such enhancements can reach numerous other use cases and produce greater efficiency improvements, our fresh capabilities are converging. Thus, from our goals to how we assist customers in selling, a lengthy opportunity lies ahead of us. Even so, this remains our most vital route, and it will continue to yield improved business returns. Thank you. That concludes my comments. We are now prepared for questions and answers.
I'd like to ask about the recent policy developments. So here's a quick question. What shifts have you observed in both the new and existing home markets since the recent policies came in? And regarding price stabilization, do you think that will carry on? And what's your outlook for the market beyond Q4 and 2027?
Thanks, Eddy. From what we've observed, the degree of stabilization in the existing home market has increased since the beginning of the year, particularly in core cities. Let's start with pricing. Nationwide existing home prices that we track were largely flat during the first months. In Tier 1 cities, cumulative price growth reached 4.1%, and Shanghai and Beijing rose by approximately 12% and 3.9%, respectively. Despite some short-term ups and downs, core cities are already showing encouraging signs of a recovery. Transaction volumes have held up well too. According to the National Bureau of Statistics, online-registered floor area for existing homes nationwide was up 10.7% year-over-year in the first line. After the latest policy subsidizing mortgage interest rates, October activity remained brisk, with unit volume in the 50 cities we track increasing roughly 31% year-over-year. For existing homes that met the subseas flow area and total price criteria, average daily transactions were 10.6% month-over-month, compared with 5.8% for other homes. We also saw a larger share of transactions in the lower price range, and that's a positive signal. That said, this is still an early data point. It will take additional time to determine if that incremental demand actually shows up. In the meantime, structural adjustment is still underway in the new home market. Across the country, new residential sales fell 13.7% year-over-year in the first 9 months. Core cities and selective quality projects performed better, although a wider recovery will require time. More important than shifts in volume and price is this: after years of investment, several key contents for rebalancing housing supply and demand have improved materially. Looking at demand, the first point is that the need for people to actually move remains fairly steady, and a growing share of it is being satisfied by previously owned properties. When you combine new and pre-owned houses, yearly transactions across the country stayed broadly in the 13.5 million to 15 million unit range between 2022 and 2025 — about 10 purchases per 1,000 residents each year. Though the overall figure held steady, its composition changed a lot. Pre-owned properties accounted for 30% of deals in 2022 and climbed to 52% in 2025, with roughly 57% anticipated for this year. Since the new-build segment has shrunk, housing demand keeps getting fulfilled via existing homes. The second point: cheaper house prices alongside reduced interest rates have brought down the bar to purchase and made long-term ownership more economically attractive. Cheaper prices have brought down both down payments and principal amounts for homes, and reduced rates are softening financing expenses across the country. Newly originated residential mortgage rates declined from 5.63% at the close of 2021 to roughly 3.1% by June of this year. For first-home provident fund loans with terms beyond 5 years, rates also fell from about 3.25% before 2022 to 2.6% as of today. Meanwhile, annualized rental yields across those 50 cities moved from 1.9% at the end of 2022 to 2.8% in September this year, which is now close to commercial mortgage rates and about level with first-tier fund rates. On the subsidized portion of an eligible commercial mortgage, the rate stands at 3.06%. During the subsidy window, the contract rate works out to an effective burden of roughly 2.06%. For long-term owner-occupiers whose down payments fit within their means and who face low financing costs, mortgage interest on certain homes may already sit below the rent charged for comparable properties. The downward pressure on rents is easing as well. In October, rents in Tier 1 cities were up 0.3% year-on-year, and Shanghai came in at about 2.1%. By helping to cover underlying living needs, we intend to turn the comparison between rental and financing costs in our business into lasting purchase demand. Turning to supply, on the development side. To start, there are fewer existing homes being listed for sale, and both active listings and their volumes are going to sit under earlier highs; during the first nine months, fresh listings throughout the cities climbed 14.3% year over year, which includes 12.2% in Beijing and 16.8% in Shanghai, with transactions holding up well. Listings in total across the 50 cities likewise came down in August as well as September. That change becomes even more evident in core cities by September. Active listings in Beijing declined 19.3%, going from an earlier peak of 163,000 units down to 132,000, whereas Shanghai fell 29.5%, from 165,000 to 116,000. Together, fewer new listings, demand that is steady and inventory that is falling demonstrate that supply-demand pressure is easing, which supports price stabilization in core cities. Next, new home supply has shrunk sharply following a year of adjustment. Nationwide, residential new starts fell about 70%, going from roughly 1.46 million square meters in 2021 down to 430 million in 2025, and then dropping a further 25.4% year-on-year during the first nine months. The volume of new supply actually brought to market is shrinking as well, with residential floor area launched across 10 key cities declining 11.8% year-on-year in the first eight months. That also means incoming new supply exerts less pressure, and more demand gets satisfied through the existing housing stock. Demand that has held up, better buying economics and reduced supply pressure are now combining to rebalance supply and demand. In core cities, volume, price and inventory are already showing positive shifts, which gives market stabilization a firmer footing. Turning to the fourth quarter of 2027, the move toward sales of completed homes and adjustments to the presale framework could influence the supply of new homes and the prices at which transactions occur, pushing part of the demand toward existing homes while postponing other purchases. How quickly things recover will continue to differ from city to city. We'll keep a close eye on how resilient existing home transactions are, whether price concessions get smaller, and whether transactions continue to absorb active listings, along with household purchasing power and the destocking of new home inventory. In light of the supply and demand dynamics today, we remain confident about sustained fundamental demand, the resilience of existing homes and a gradual stabilization of prices. Thank you, Eddy.
I believe your results have been quite strong. So what I'd like to ask is, in what ways might the completed home sales reform take shape across the industry? And how will you adapt your strategy? And what are the implications for growth and profitability when it comes to dealing with the new changes in the market?
Thanks, Chen. Now, regarding the completed home sales reform, we're monitoring it closely. I'd like to share 2 early observations. First, on market size: although new home supply could swing around over the coming few years, we do have professional third-party marketing and sales services, and those are quite steady. And when it comes to the existing home market, it should keep expanding further. Second is creating an end-to-end offering that runs from market — from marketing through to project sales organization; so by answering who the buyers will be, how we reach them, and how we help them make their decision. That's why we brought in developer money effectiveness, and we're able to seize bigger business opportunities. So beginning with the market developers, the total marketing budget for will contract as new supply declines, and how that budget gets allocated is moving this effect. For one thing, when budgets get allocated, third-party marketing as a profession enables developers to reach buyers earlier and communicate project value as a whole in language that's friendly to consumers, with efficiency and cost filings, which creates room for deeper service penetration. For another, developers may cut fixed spending on in-house sales centers and their teams, directing budgets toward third-party professionals. As a result, we see both a broadening scope for professional services and a growing per-packet share of marketing budgets, which offsets the decline in overall industry budgets. We also have < > So the capabilities and the projects that were experienced from < > can be reused in these services too. At the same time, the existing home market has room to grow, particularly for high-quality new existing homes; many by, yes, who favor new homes actually prize the in-house, community environment and living quality of new homes. And with qualifying new supply chains, comparable new system homes at the right price could become a viable purchase. By serving these 2 markets, we're able to help consumers compare according to actual needs while capturing this incremental demand. Turning to our strategies: for new homes, we're integrating targeted marketing, sales agency and distribution channels to boost developers' overall sales efficiency. What developers really focus on is concrete questions: where the buyers are, what makes them pick one project over another, and how interest gets turned into an actual purchase. Basically, they're after higher sales volume at a lower overall selling cost—or quicker sales at roughly the same cost—all while holding the line on price. That's precisely what our services are built to deliver. Beike's goal is to provide developers with an end-to-end set of services, all the way from finding purchasers through to running a project's sales effort. And when the partnership goes deeper, we get involved sooner on the production side too—covering product positioning, unit-mix layout, design refinement, and pricing. Marketing, then, addresses who the buyer is, the reasons behind their choice, and how to generate awareness. Channels locate and attract those customers, while on-site sales teams assist them in making decisions once they arrive, and the integrated service offering coordinates every stage so that developers achieve better sales results. In marketing specifically, the industry depends on broad brush campaigns, while we concentrate on precision marketing, drawing on multidimensional user market data and historical transaction data to identify likely buyers' motivations and their concerns. We then convert project features into clear, comparable content delivered through the big application, agents, private networks, off-line stores and external media. We also examine inquiries, registrations, views and subscriptions in order to redefine content and ad placements; for buyers, this brings needed information earlier. On the developer side, it brings in real demand, generates leads that convert at a higher rate, and links marketing spending more tightly to actual sales. Stronger channels and better on-site conversion are what effective marketing depends on, and by drawing on our stores and agent network we assist buyers in comparing new versus existing homes and matching projects, while also improving on-site reception, presentations and follow-up. Developers are able to buy individual modules or mix them together to strike a better balance among pricing, sell-through speed and selling costs. These models are already producing results in practice. Across the Suzhou marketing projects, average daily views during the peak promotion period increased by roughly 34%, and the view-to-transaction conversion rate climbed to 60%, which confirms that our leads are accurate. Separately, for Guangzhou Guadu Urban Construction Investment Group — our first integrated service project — the new homes and B Hotel teams came in during early customer research. They helped reshape core products, positioning, lot mix and overall pricing so it matched local demand. Those insights then shaped marketing content, agent training and on-site sales. Initial launch subscriptions came in above 90%, and the integrated package sharply lowered the developer's marketing expense ratio. That's why we're confident that weaving customer insights through product design and sales delivers clearly fuller value. Moving ahead, we'll continue honing our service model and confirming outcomes project by project. Changes in the market only strengthen how sure we are about this path: when we provide marketing and sales services that work well, linking IS with home and assisting developers in selling, we can secure additional partnerships, handle wider scopes of work and achieve better returns, which opens up lasting growth and profitability in the market. Thank you.
Nice work on everything you've achieved. I've got a marketing question: during Q3, we can see non-in transaction volume on a per-person basis. So looking ahead, how do we improve efficiency, and what do we expect the next phase of development to look like?
After the changes we made earlier, our costs have come down, and that shows up in the financial reports too; going forward, we'll push the conversion rate as well as service efficiency. In practice, that means more customers choosing Beike for our professional service while resources don't grow at the same pace; and we can steer clear of scaling the resource. Deciding to buy or sell a home is one of life's biggest choices. Buyers and sellers expect a partner who understands their circumstances, makes the options clear, and stays with the deal until it closes. Having already built a network that links buyers, listings, and agents, our task now is to use it to deliver service more effectively. Going forward we have three priorities: raising how professional our service is and strengthening problem solving; moving beyond merely connecting listing buyers and agents toward matching them together; and extending proven best practices to more people. In Q3, home sale transactions per AI agent rose 31.7% year-on-year, chiefly because of staffing adjustments, while sales per connected store were up 16.8% year-on-year, largely on the strength of gains in retail stores. Looking ahead, we intend to grow the service brand and adapt by improving service and sharpening matching. To start, we're creating more specialized, need-specific professional services for particular customer requirements. First off, making expertise scalable is essential; we're urging managers and leading agents to put selling points, ongoing market insights, and lived experiences—including recent ones—into digital form. This presents know-how to buyers while converting what an individual knows into shared, reusable knowledge. In Beijing, for instance, our core community seller agent content spans more than 4,000 accounts and 140,000 posts, measured by expense metrics. That content has expanded beyond property listings into community reviews, living experience, and buying advice, so clients can see the expertise before they even view a home. Second, we're expanding what our services include. Take renovation as an example: our layout redesign proposals reshape the home search, letting buyers influence future upfront costs, alongside cross-project and new-resi-versus-existing-home comparisons. These tackle genuine customer needs that went underserved before. Last, we're piloting clear row specialization—assigning specialists to buyer needs, sellers, deep project research, and complex transactions—plus offerings such as community open days, and we're committed to sell organized workflows through and collaborations on to complete solutions. Professionalism can't rest on individual importation; it also demands dedicated roles and structured service methods. Second, we're boosting matching and collaboration across listing, buyers, and agents to maximize our professional capabilities. In some cases, our front-end online service assistance team evaluates what the customer needs, and prior to pairing them with suitable agents, lead allocation increasingly relies on service quality expertise and a proven conversion track record, while making sure there is continued follow-up for leads that were previously under-nurtured in Beijing, for instance. As of October 29, prospective buyers whom OFC had identified and nurtured during Q3 had achieved a cumulative conversion rate of 2.7%, which is more than 8x the platform average over that same period. That shows the value of identifying needs early, nurturing continuously, and handing off seamlessly to agents. To determine whether these efforts genuinely boost agent and store productivity, we track three things. First, are teams and agents reaching and serving a greater number of qualified buyers? Our priority is acquiring new clients via our content and China channels, while also drawing on our current listing and customer pools to capture demand we previously missed. What we track is growth in overall network coverage, not simply clients or buyers moving from one agent to another. Second, can we turn a bigger portion of that same customer pool into completed business? That depends on whether stronger need discovery, matching advice and execution settle decisions where clients were previously sold or lost to competitors. Third, can we finalize deals while reducing repetitive work of low value? After research on a layout property or a transaction issue is done, agents shouldn't need to redo it — the platform and AI can take care of routine questions and arrange the data, which frees professionals to focus on tasks requiring high judgment. Our target growth model centers on serving more clients, recent conversion and reducing duplicated effort. And of course, other business lines also show clear paths to better profit; for example, our new homes, our precision marketing channel and agency services deliver wider margins than the pure distribution model. Second, in rentals, the market is vast through ongoing product diversification and optimization. Our goals are to increase carefree rent penetration, improve renewal rates, lengthen tenancy periods, enhance agent productivity, and reduce repeat acquisition costs along with per-unit operating expenses. Third, in home renovation, where we maintain consistent delivery quality, we seek to improve conversion through pre-transaction renovation proposals and end-to-end sales design and delivery capabilities. During Q3, rental services accounted for roughly 11% of the company's gross profit, while renovation brought in approximately RMB 300 million in operating profit before headquarter expenses. These two business lines are set to become major profit drivers; and < > is also actively exploring, together with our new home team, the creation of integrated developer solutions. Looking ahead, we'll check two things: whether these practices scale consistently across teams, and how much net incremental profit is left once incremental investment is subtracted. That's how we convert professional service and customer value into sustainable profit growth. Thank you.
Well, congratulations on the strong results. My question is about our renovation business. We saw that contracted value has come back to growth. Could you give us a bit more color? Where are we in terms of our strategic adjustment? And regarding whether contract value growth can be sustained, how do you see that? How do you — or will you — balance scale, investment and profitability going forward? That's my question.
Thank you, Timothy. So, the decline in revenue for this quarter is mostly a reflection of those contract losses we saw earlier, plus the time gap between when we sign something and when we can actually book the revenue. On a positive note, though, the value of renovation contracts came back to year-over-year growth this quarter, and that was helped by a higher year-over-year conversion of referrals coming from our core brokerage operations. This comes from tighter synergy between home transactions and renovation, which improves efficiency all the way from spotting the need and making the referral through to fulfilling the service. In Q3, the conversion rate for existing Home 2 renovation — that's renovation orders divided by existing home transactions — hit 4%, up from 3.2% in Q2. A major factor was referring and bringing more existing home buyers into renovation programs. At this front-end stage, we concentrated on 3 areas. To start, we're clarifying how brokerage haven and renovation teams coordinate and what rules apply between them, while aiming incentives at agents who convert well so referrals grow. Next, we're adding renovation customer managers who specifically handle agent referrals, follow up quickly, and connect clients to specialist teams—this lowered drop-off among customers after transactions. As of the close of Q3, customer manager headcount was up roughly 300 net from where it stood at the beginning of the year. Third, we're placing renovation showrooms in the same locations as transaction signing centers. That way buyers can look at renovation options immediately after they sign the contract, which cuts the time agents spend traveling and coordinating. Getting clients into the showroom is step one; turning them into buyers once they're in the store matters just as much. So with that in mind, we've just refreshed our showrooms—adding more tangible displays, quicker design generation, and actual completed case studies—all to give clients confidence in what they'll ultimately receive. These upgrades center on three areas within the showroom. First: transforming vague ideas into concrete requirements. Clients frequently arrive with just a general style or a lifestyle concept. Through style displays, real-life setups, and model rooms, we let clients experience the layout, storage, lighting, and coordination firsthand—clarifying their functional and aesthetic needs while helping designers grasp them accurately. Second: swiftly turning needs into concrete design proposals. The showroom styles, themes, and models are backed by our product catalogs. With the floor plans and spending limits that customers who've committed to the tour provide, our owners are able to match products rapidly into a particular design and bundle, reducing the back-and-forth involved in choosing and giving customers a transparent view of their budget. Last, and crucially, we present genuine finished projects, so folks have confidence that we will come through. We take completed or prior projects, together with the corresponding materials, products, and craftsmanship, and display them. This lets customers witness the real home experience. Since our supply chain manages sourcing and delivery, this ensures that what you see is what you get, which pushes more conversions. Consider the Shenghuo Tiandi new-build project in Wuhan, where the starting price sits in the high teens; there we carried out a model renovation, adapting the primary floor plans to what U.S. buyers require. Between the project's launch and early November, preliminary renovation contracts were signed for over 30% of new home subscriptions. That distinguishes us from conventional home improvement or renovation retailers. Our showrooms went beyond merely displaying materials—they offer full living solutions matched to each client's actual home, needs, and budget. Looking ahead, we will stay anchored in what customers need, enhancing the experience via professional design and quality delivery so as to drive high-quality renovation growth. Thank you.
I'd like to ask about the rental side of our business. Switching to a different accounting principle has had some effect on us. So setting aside the impact of the net basis counter, how is the rental business really doing? In other words, what does the medium- to long-term route to unlocking operating leverage look like?
Thanks for the question. To begin with, during Q3 both rental scale and operating efficiency continued to improve. At the close of Q3, the total number of rental units under management reached 830,000 units, a 25% increase year-over-year. Carefree rental renewal rates rose steadily. As of the end of September, landlord renewal at exploration came to 77.9%, 7.6 percentage points higher year-over-year, and tenant renewal stood at 56.3%, up 3.5 percentage points year-over-year. Turning to finances: our switch to net us accounting caused Q3 rental revenue to fall 25% versus a year earlier, landing at RMB 4.27 billion. Contribution margin rose 7.9 percentage points from the prior year to 16.6%, largely a reflection of Net basis products carrying more weight. Net basis units accounted for 65.9% of total units by the close of Q3, compared with 24.5% twelve months before; in dollar terms, Q3 rental contribution profit reached RMB 708 million, roughly RMB 210 million above the year-ago figure. The main forces behind this were expansion of our free rent scale plus continued cost efficiency in operating staff channels and post leasing services. Starting with scale growth: during Q3 we raised leasing speed for newly signed units. First-time leasing success on a monthly basis averaged a peak of 88%, which is a gain of 2.9 percentage points compared with a year earlier, allowing us to stabilize units more quickly, enhance the landlord experience and underpin inventory growth. Second, staff costs on a per-unit basis keep getting better thanks to row, specialization, digitalization and a standard of workflow, and we achieved a reduction in repetitive tasks along with gains in liquid management efficiency. During Q3, the number of units managed per property manager came in above 170%, an increase of more than 20% from the prior year, while monthly units leased per manager hit 22%, 19% higher year-over-year. Third, our channel cost ratio came down through proactive renewal management, centralized teams and AI tools — we identify renewal opportunities and churn risks early, which lifts retention and lowers repeat acquisition costs. In addition, repair and cleaning costs after leasing showed further improvement as a result of supply chain optimization. Moving on to making the rental system profitable in a careful way—our number one priority is to lift renewal rates and keep tenants longer, and we're doing that on three fronts. First, we're improving how renewals are managed by setting up proactive outreach, figuring out what people need, and creating follow-up mechanisms around the key pre-exploration models milestone. Second, we're offering flexible service packages for different tenant needs to raise satisfaction and retention; for instance, affordable basic packages for tenants who are price-sensitive, and hassle-free home switching for people relocating for jobs. We're also looking into pet insurance for pet owners. An early buyer need identification linked to our brokerage business for an integrated and rent-to-buy journey. Number three, elevating the in lead living experience. We are testing light touch-ups and partial renovations to upgrading unit quality and comfort. At the same time, role specification lets us have staff dedicated to lead service and maintenance, which ensures front-line and effective resolution. Building on that, we plan to explore a variety of long-term rental options aimed at different landlords and tenants, in order to boost appeal and expand quality supply. Alongside this, ongoing efficiency gains will help us convert growth at scale into profit growth more efficiently, which unlocks extra operating leverage. Thank you.
That wraps up the Q&A portion for today. Thanks again for being with us. Should you have any additional questions, you're welcome to reach out to Beike's Investor Relations team using the contact details listed on our IR website. That brings today's call to a close, and we hope to talk with you once more next quarter. Thank you, and goodbye.
Read the full transcript via the API
You're viewing the first half of this call. Get the complete KE Holdings Inc. transcript - plus 131,000+ transcripts from 7,000+ companies, speaker segments and full-text search - through the EarningsAPI REST API or hosted MCP server.
Get an API key View API docs →For developers and AI pipelines
Programmatic access to KE Holdings Inc. earnings transcripts and 131,000+ others is available through the
EarningsAPI REST API and the hosted MCP server.
Quarterly plans from $162 - full transcripts, speaker segments, full-text search,
and the /api/v1/transcripts/recent polling endpoint for ETL pipelines.