Hello, ladies and gentlemen. Thank you for standing by for KE Holdings Inc. Q4 and FY 2025 earnings Conference Call. Please note that today's call, including management's prepared remarks, and question- and -answer session, will all be available in English. Simultaneous interpretation in Chinese is available on a separate line for the duration of the call. To access the call in Chinese, you will need to dial into the Chinese language line. At this time, all participants are in listen-only mode. Today's conference call is being recorded. I will now turn the call over to your host, Ms. Siting Li, IR Director of the company. Please go ahead, Siting.
Thank you, operator. Good evening and good morning, everyone. Welcome to KE Holdings Inc., Beike's Q4 and FY 2025 Earnings Conference Call. The company's financial and operating results were published in the press release earlier today, and are posted on the company's IR website, investors.ke.com. On today's call, we have Mr. Stanley Peng, our Co-founder, Chairman, and Chief Executive Officer, and Mr. Tao Xu, our Executive Director and Chief Financial Officer. Mr. Xu will provide an overview of our business updates and financial performance. Then Mr. Peng will share more on our strategic update and thinking. Before we continue, I refer you to our safe harbor statement in our earnings press release, which applies to this call as we will make forward-looking statements.
Please also note that Beike's earnings press release and this conference call include discussions of unaudited GAAP financial information, as well as unaudited non-GAAP financial measures. Please refer to the company's press release, which contains a reconciliation of the unaudited non-GAAP measures to comparable GAAP measures. Lastly, unless otherwise stated, all figures mentioned during this conference call are in RMB. Certain statistical and other information relating to the industry in which the company is engaged, to be mentioned in this call, has been obtained from various publicly available official or unofficial 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 English as the main language.
Please note that the Chinese translation is for convenience purposes only. In the case of any discrepancy, management statements in their original language will prevail. With that, I will now turn the call over to our CFO, Mr. Tao Xu. Please go ahead, Tao.
Thank you, Siting. Hello, everyone. Thank you for joining our 2025 Q4 and Full Year Earnings Call. To begin, I would like to provide a summary of our financial highlights for the fiscal year of 2025. In 2025, in response to evolving customer needs, we initiated a strategic pivot from sales-driven to efficiency-driven growth. To optimize our business model, better leverage technology, and improve our cost structure and unit economics, we implemented a series of initiatives. These efforts are laying the foundation for more sustainable growth while strengthening the stability and the flexibility of our earnings model. First, our full year revenue remained relatively stable amid market fluctuations, outperforming the broader industry trends. This performance was underpinned by a more diversified and countercyclical business structure. Revenue from non-housing transaction business accounted for record high of 41% of total revenue.
The internal structure of the housing transaction services also improved, with the existing home GTV accounting for 67.6% of our total GTV, reflecting our focus on market segments with greater structural growth potential. Notably, the GTV contribution from connected brands further increased to approximately 63% of our existing home GTV, indicating higher contribution of revenue with lighter business model. The existing home platform service revenue was basically stable year-on-year, also demonstrating the resilience of our platform business model. Second, our operational efficiency improved and cost structure was optimized, laying foundation for future profitability function. In our existing home business, fixed labor cost recorded a sequential decline for the four consecutive quarters throughout the year, significantly enhanced the profit elasticity. By the end of the year, we had the release of operating leverage with the contribution margin of existing home business rebounding sequentially in Q4.
In our new home business, both variable cost ratio and fixed personnel expenses decreased year-on-year, driven a 0.2 percentage points year-on-year increase in the full year contribution margin. The home renovation business significantly narrowed its operating losses, and home rental services turned profitable at operating level for the full year, with their contribution margins rising by 0.7 percentage points and 3.6 percentage points year-on-year respectively. The overall operational efficiency also continued to improve, with operating expenses ratio down 1.4 percentage points year-on-year. Third, we remain steadfast in our commitment to delivering the active shareholder return. In 2025, our total share repurchase reached approximately RMB 921 million, a year-on-year increase of around 29%.
Furthermore, we are pleased to announce a final cash dividend plan for 2025 of approximately $0.3 billion, bringing our full year total shareholder return to approximately RMB 1.22 billion, a year-on-year increase of around 9%. This accounts for approximately 170% of our 2025 non-GAAP net profit, far exceeding the proportion in 2024. Turning to our Q4 performance, due to the high base in the same period of 2024, our GTV and revenue saw a notable year-on-year decline. Our GTV reached RMB 724.1 billion, representing a decrease of 36.7% year-on-year. Revenue was RMB 22.2 billion, down 28.7% year-on-year.
As a result of the decline in transaction scale, our gross profit margin was 21.4%, a year-on-year decrease of 1.6 percentage point. Q4 GAAP net profit was RMB 823 million, down 85.7% year-on-year. Non-GAAP net profit was RMB 517 million, representing a year-on-year decline of 61.5%. It is important to note that our bottom line performance in Q4 was partially affected by one-off expenses related to our cost optimization initiatives, which is adjustment weighted on near-term profitability. They are helping us to streamline our cost structure and position the company with greater operating leverage going forward. With that overview, I'd like to provide some details on the financial performance of each business segment.
In our existing home business, due to the relatively higher base in the same period last year, the scale of our existing home transaction business declined in the Q4 . While the profitability improved, GTV from the existing home business reached RMB 482 billion in Q4, reflecting a 35.3% decrease year-on-year and a 4.7% decrease quarter-on-quarter. Revenue was RMB 5.4 billion, down 39% year-on-year and a 9.2% quarter-on-quarter. GTV outperformed the revenue year-on-year was mainly due to higher GTV contribution from the existing home transaction facilitated by connected agents, for which revenue are recorded on a net basis. On a quarter-on-quarter basis, GTV outperformed the revenue was mainly driven by the structural shift as the revenue contribution from the rental brokerage services decreased amid seasonal fluctuations, which have a relatively higher take rate.
In this segment, revenue from platform service decreased by 19.9% year-on-year, significantly outperforming the overall GTV decline and demonstrating the resilience of the platform model. Despite the year-on-year adjustment and the sequential decline in revenue, the contribution margin of the existing home business reached 40.4%, remaining stable year-on-year and rising 1.5 percentage points quarter-on-quarter. This resilience in profitability against external volatility is the direct result of our disciplined high control and our focus on organizational efficiency in 2025. For new home business, affected by high base, the scale declined year-on-year, with profitability improved. GTV reached RMB 207 billion in Q4, a year-on-year decrease of 45.7% and a sequential increase of 5.5%.
Revenue from the new home business was RMB 7.3 billion, a year-on-year decrease of 44.5% and a sequential increase of 9.4%. GTV outperformed the revenue year-on-year was mainly due to the higher base of the monetization rate. While revenue outperformed GTV quarter-on-quarter were primarily due to the seasonal factors. Even with the significant scale fluctuation, the contribution margin of the new home business rose to 28.3%, an increase of 2.6 percentage point year-on-year and a 4.2 percentage point quarter-on-quarter, benefiting from the cost structure optimization driven by the founder operations. For home renovation and the furniture services, revenue reached RMB 3.6 billion in Q4, a year-on-year decrease of 12% and a sequential decrease of 15.9%.
This temporary softening in revenue reflects our prudent balance between scale and risk as we proactively optimized our channel structure at a moderate pace of certain non-brokerage channels. Contribution margin was 28.8% in Q4, down 0.9 percentage point year-on-year and 3.2 percentage point quarter-on-quarter. Mainly because we made a provision for potential warranty costs of the home renovation orders still under warranty period at the end of 2025, based on the principle of prudence. Excluding this impact, our core cost structure continues to improve. This increase in centralized procurement has led to a sustained savings in material costs.
Turning to our home rental services, revenue reached RMB 5.4 billion in Q4, a year-on-year increase of 18.1%, with the profitability improved. The growth in revenue was mainly driven by the rapid growth in the number of rental units under management. By the end of Q4, we had over 700,000 rental units under management, a year-on-year increase of around 62%. On a sequential basis, revenue saw a slight decrease of 5.5%, mainly due to a change in accounting method brought by product model upgrade. The Carefree Rent business has continued to iterate towards a lighter and a lower risk product model, leading to an increase in the proportion of rental units with the revenue recognized on a net basis, which has a temporary impact on the revenue scale.
However, this does not change the robust growth trajectory of our management scale or the service capability. Meanwhile, the contribution margin from rental services was 10.4% in Q4, up 5.9 percentage points year-on-year and 1.7 percentage points sequentially, mainly driven by two factors. First, the structural improvements from the ongoing shift towards a lighter product model. As of the end of 2005, the proportion of rental units with the revenue recognized on a net basis has exceeded 30%. Second, operational efficiency gains that optimized our unit economy model. Since the process of restructuring and the professional role specialization, we have significantly improved the productivity of the property managers, leading to a notable optimization of labor costs.
In addition, the gradual penetration of AI technology across the entire operation value chain has laid the foundation for the large-scale expansion and the sustained profitability of the business. In Q4, our revenue from emerging and other services increased by 4.5% year-on-year and 60% quarter-on-quarter to RMB 459 million. Now, moving to other financial metrics in Q4, including other costs and expenses, profitability, and cash flow. Our store costs were RMB 702 million in Q4, the year-on-year decrease of 9.6%. This was primarily driven by the optimization of rental costs for our Lianjia stores and the refinement of our store structure. On a sequential basis, store costs increased by 7.2%, primarily due to one-off expenses from the store closures.
Q4 gross profit decreased by 33.7% year-on-year to RMB 4.8 billion, remaining relatively flat sequentially. The gross margin was 21.4%, a year-on-year decrease of 1.6 percentage point, mainly due to the declining revenue contribution of the existing home and the new home segments, which have relatively higher contribution margins. This impact was partially offset by the year-on-year gross profit margin expansion of the home rental business. Our gross margin was relatively flat quarter-on-quarter. In Q4, GAAP operating expenses were RMB 4.9 billion, a year-on-year decrease of 20.4% and a sequential increase of 13.3%. The quarter-on-quarter increase was mainly due to one-off expenses related to the cost optimization initiatives. Excluding this non-recurring impact, the trend of operating expenses is fully consistent with our efficiency improvement efforts.
These expense optimization initiatives position room for greater operating leverage going forward. To break down the component, G&A expenses were RMB 2.3 billion, down 23.9% year-over-year, mainly due to the reduced bad debt provisions and the lower share-based compensation. The 20.8% sequential increase was mainly due to the aforementioned one-off optimization costs. Sales and marketing expenses were RMB 1.9 billion, down 17.7% year-over-year, mainly due to the lower personnel-related expenses driven by operational efficiency improvements. The 11.7% sequential increase was mainly due to the seasonal marketing and the promotion expenses. R&D expenses were RMB 750 million, roughly flat year-on-year and up 2.3% quarter-over-quarter, mainly due to the aforementioned one-off optimization costs.
Moving to our bottom-line performance, our GAAP operating losses was RMB 147 million in Q4, compared with a profit of RMB 1.01 billion in Q4 of 2024 and RMB 608 million in Q3. The operating margin was negative 0.7%, a year-on-year decrease of 3.9 percentage points and a sequential decrease of 3.3 percentage points. The non-GAAP income from operations totaled RMB 323 million, decreasing 81.6% year-on-year and 72.5% quarter-on-quarter. The non-GAAP operating margin was 1.5%, a year-on-year decrease of 4.2 percentage points and a sequential decrease of 3.6 percentage points, mainly due to the increase in the operating expenses ratio.
Finally, GAAP net income totaled RMB 82 million in Q4, down 85.7% year-on-year and 89% quarter-on-quarter. Non-GAAP net income was RMB 570 million, falling 61.5% year-on-year and 59.8% quarter-on-quarter. Moving to our cash flow and the balance sheet, we generated net operating cash inflow of RMB 1.9 billion in Q4. In 2025, net operating cash flow was below the profit performance, mainly affected by the timing factors in working capital, including the payment of the accrued bonus from the previous year and the change in contract liability in our home renovation business due to order intake was moderated. Excluding the impact of the above timing factors, our net operating cash flow performance was broadly consistent with the profitability.
Our new home accounts receivable turnover days was 44 days in Q4, a sequential decrease of approximately 10 days remaining at a healthy level. In addition to spending approximately $246 million on share repurchase during Q4, our total cash liquidity, excluding customer deposit payable, remained at around RMB 68.7 billion. With a robust cash reserve, we place a high importance on shareholder returns. We spent approximately RMB 921 million on share repurchase for the full year of 2025, representing approximately 4.1% of total share outstanding at the end of 2024. Our track record reflects a consistent dedication to fulfilling our promise to shareholders.
Since the launch of our share repurchase program in September 2022, we have repurchased a total of approximately RMB 2.5 billion in shares at the end of 2025. A total reduction of approximately 12.6% of company's total issued shares prior to the program launch. On top of this robust shareholder return, we are pleased to announce a final cash dividend plan totaling approximately RMB 0.3 billion, which will be funded by surplus cash on our balance sheet. With this, our total shareholder return for 2025 significantly exceeded our non-GAAP net income, representing around 170% of our non-GAAP net income for the year.
Overall, in 2025, we placed a greater focus on improving operation quality and resources allocation efficiency, while continuing to optimize our business mix, cost structure, and expense discipline. Our current cost structure is more streamlined. The profit model is clear, and the profitability quality of each business segment has improved. We have also adopted a more comprehensive and proven approach for the pace of our emerging business, capital investment, and risk control, which has ensured a sound balance sheet. Looking ahead to 2026, we are willing to improve in the financial disciplines and strike a balance between efficiency and growth. We will continue to improve our earnings quality, optimize our capital and efficiency structure, while delivering our long-term competitive advantages, thereby creating sustainable value for our shareholders. Thank you.
Next, I would like to turn the call to our Chairman and CEO, Stanley.
Thank you, Tao. Good evening, everyone. Thank you for joining us today for Beike's Q4 and FY 2025 Earnings Conference Call. Over the past year, we have seen many changes in the market. For example, the transaction structure is evolving. The share of existing home transaction in China's housing market continues to increase. In 2025, the number of existing home transactions nationwide hit a historical high. The new home market is also seeing greater differentiation, with higher quality and new standard projects attracting stronger market demand. More and more young people are choosing to rent, while rental yields are gradually improving. Customer transaction behavior is also changing. Housing information is becoming increasingly abundant, yet the decision-making process is becoming more complex. Both buyers and sellers are thinking, taking longer to complete transactions.
The cost of making a mistake is much higher now, and consumers are more and more cautious. Buying a home used to be a relatively easy decision. Today, it's a balancing act that can require a careful reallocation of family assets. At the same time, some things have not changed. The overall demand for better living remains stable, and consumers' demand for safe, professional, transparent, and reliable services is still strong. By looking at what has changed and what has stayed consistent, we can tell two very important things. First, China's residential market remains the largest and most valuable housing market in the world. Second, the housing service industry has made a fundamental shift in its approach. Today's consumer needs more professional services that offer certainty in decision-making.
The industry is entering a new stage where core competency will no longer be defined by resource scale, but by service capability and operational efficiency. Ultimately, creating value for customers will be the only stable source of our long-term growth. In this trend, Beike continued to evolve in 2025. First, we improve our operational governance, creating more room for long-term strategic transformation and enabling us to continue driving progress across the residential service industry. Second, we upgraded our strategy. Leveraging data and AI, we are rebuilding our service logic around customer value. Through greater value creation, we aim to improve the platform overall customer coverage, resource conversion efficiency, and unit outputs. Our growth model is therefore shifting from one driven primarily by the scale of agents and stores to one driven by efficiency and value creation.
In the past, we focused on expanding the numbers of stores, listing coverage, and lead volume. Going forward, we will focus more on delivering greater certainty in transactions for customers, improving matching precision, and strengthening the unit economics. Specifically, we are working in four key areas. First, upgrading transaction services into full process decision support services, improving professionalism and certainty in the service process. Second, optimizing resource allocation through data and AI, so consumers can receive high quality, better matching services. Third, embedding AI capabilities into our service workflows, helping service providers and the platform deliver more professional, people-centric services. Fourth, building diversified service capabilities across a broader residential ecosystem to meet customers' full range of housing needs. Next, I will walk you through the progress of our major business segments in 2025 and share some of our thinking.
For existing home business, the platform facilitated RMB 2.15 trillion in GTV from the existing home transaction in 2025. Within that total, the number of existing home sales transaction increased by more than 10% year-over-year, reaching a record high. At the same time, transaction volume from platform connected stores increased by 15% year-over-year. These two figures highlights the resilient demand in the existing home market and the strengthening of our platform model. The overall scale of agents and stores on the platform remain stable, with more than 58,000 connected stores and over 445,000 agents at the year-end.
In terms of productivity, in 2025, the average number of existing home transactions per connected agents increased by 6% year-over-year, rising from less than two transactions per agent in 2022 to more than three. For our directly operated Lianjia business, we proactively optimized store networks and agent structure in 2025. We focused on high efficiency capability and deeper operations in core cities. Out of these adjustments, Lianjia's per agent productivity in core cities improved, indicating we are gradually achieving a healthier balance between scale, discipline, and productivity improvements. Operationally, we upgraded our lead allocation mechanisms and refined tier services, ensuring that high-quality clients receive services better matched to their needs. We also continue upgrading our service model to adopt a more consultative approach, moving beyond simple property tools and matchmaking through a deeper support for decision-making.
In today's market environment, customers do not lack information. What they lack is assistance in making judgment. AI is becoming a new productivity engine for our industry. Property transactions are not standardized commodity transactions. They involve both rational analysis and emotional judgment. They require both data support and real-world offline experience. In the past, the industry has not done a good job of structuring the rational parts of the process, nor has it placed the emotional aspects where they create the most value. In some cases, emotional judgment has even been used to replace decisions that should have been made rationally. When these two elements become intertwined, it inevitably leads to a loss of efficiency.
AI can make the rational parts of the process extremely rational while amplifying the value of the human and emotional aspects that must be handled by people. In our industry, machines can process data, but true judgment, explanation, and trust still need to come from people. AI cannot be ignored, nor can humans be replaced. This is why our strategy is to combine human expertise with AI capabilities. We are embedding AI directly into our core operational scenarios across the platform. For example, in our housing transaction business, AI marketing assistant help agents automatically generate marketing materials. AI simulation tools also help service providers practice real customer interaction scenarios and continuously improve their professional capabilities. Going forward, AI will act as a co-pilot for service providers across the entire customer life cycle.
This includes demand identification, precise matching between agents, homes and customers, pricing decision support, and process automation. Over time, AI will help prepackage the expertise of those top performing service providers so that these professional skills can be shared and used across the platform. New home business. Transitioning from channel-driven to structural efficiency. In our new home business, we are shifting from relying on channel distribution advantages to driving growth through structural efficiency improvement. In 2025, Beike facilitated RMB 890.8 billion in new home GTV. Despite a volatile market environment, we continue to outperform the broader market. Building on our growing listing support, supply, and channel sales scale, we are now driving sustainable growth by improving structural efficiency. This includes optimizing the mix of customers, projects, and service providers, as well as improving matching precision.
For home buyers, we are strengthening capabilities in customer demand identification, cross-project comparison, service providers matching, and decision support. For developers, we are beginning to provide early-stage project positioning insights while also offering integrated marketing and sales services in the latter stage of project sales growth. For service providers, we continue refine evaluation system and operational tools, and refine our resource allocation mechanisms, ensuring that agents with stronger conversion capabilities are matched with the right resources. Our goal is to upgrade the new home business from a model focused on traffic distribution to one that delivers greater certainty of results for all participants in the ecosystem. New business from scale exploration to profit quality and sustainable models.
Beyond brokerage services, our home renovation and furnishing and home rental business both enter a healthier stage of development in 2025. Across both business segments, we are placing great emphasis on profit quality and on building sustainable and replicable operating models. This is the foundation for this business to scale over time. In the home renovation and furnishing segment, full-year revenue grew by 4.4% to RMB 15.4 billion, while profitability improved meaningfully. Contribution margin increased to 31.4%, up 0.7 percentage points year-over-year, and operating losses narrowed significantly. Over the past year, we have focused on advancing product standardization and design digitalization through a system of packages and a modularized product offerings, as well as AI and BIM-enabled online design workflow.
We are gradually turning design capabilities into system capabilities. This helps reduce service variance and improve conversion efficiency. At the same time, we have been advancing supply chain integration and building standardization, standardized delivery systems while improving the customer experience. These efforts have also enhanced profitability while gross margin increased, losses narrowing significantly and the home renovation business is evolving from a project-based model relying on individual experience to a more scalable and replicable service model. We have also established a clear path toward long-term profitability. In our home rental services segment, the number of managed units exceeded 700,000 by year-end, representing a 62% year-over-year increase. The business achieved full-year profitability, with contribution margin improved to 8.6%, up 3.6 percentage points year-over-year, demonstrating a meaningful improvement in profitability.
We continue to upgrade the product structure toward lighter, more resilient, and more controllable models while strengthening unit economics at the individual property level. By redesigning our workflows and introducing specialized roles, the operational efficiency of core service providers continue to improve. AI capabilities are gradually being embedded into key areas, including property sign up, pricing support, leasing management, and operating strategy. This helps reduce operational risk, including, increase leasing efficiency and optimize cost structures. With this improvement, our rental business is forming a more stable profitability profile and a more consistent cash flow. Overall, our new business are moving from a phase of scale exploration into a stage focused on business model validation and profitability improvement. As these models mature and technology adoption increase, that will further diversify our revenue structure, strengthening resilience across market cycle and better serve our consumer broader residential needs.
At the organizational level, we are also advancing structural optimization and rebuilding capabilities. The purpose of our organization is not simple to manage metrics, but to continuously improve the customer experience. We are streamlining organizational structures, simplifying management layers that do not directly create customer value, and encourage managers to move closer to the front line to better understand and create customer value in real operating scenarios. In terms of capital allocation, while maintaining a strong cash positioning and the ability to invest for the long term, we continue to deliver meaningful returns to shareholders. In 2025, we repurchased approximately $920 million in shares, representing about 4.15% of our total shares outstanding at the end of 2024. We also announced a final cash dividend.
In total, shareholder return for the year was approximately $1.22 billion, significantly exceeding our non-GAAP net income for the year. I believe our long-term advantage lies in the combination of organizational efficiency and capability, capital efficiency. For 2026, we maintain a neutral market view. Given the scale of China's real estate market and the continued differentiation in demand structures, long-term value will not be driven by just buying traffic or adding more labor. Instead, it will be determined by how deeply we understand customer needs and by the systematic service capabilities we build around the entire customer life cycle. For Beike, 2026 will be a year of validating our decision support service model. We will focus on testing how this model improves conversion rates and unit economics. 2026 will also be a year of strengthening our service and organizational capabilities.
This capability will allow us to demonstrate greater operational resilience as the industry stabilizes. In a new cycle, true leadership will not come from scale, but from capability. The foundation of capability, we believe, ultimately lies in only one thing. That is continuously creating real and verifiable value for our customer. With that, we can now move to the Q&A session. Thank you.
Thank you. If you wish to ask a question, please press star one on your telephone and wait for your name to be announced. If you wish to cancel your request, please press star two. If you're on a speaker phone, please pick up the handset to ask your question. As a reminder, we only accept questions on the English language line. For the benefit of all participants on today's call, please limit yourself to one question. If you have additional questions, you can re-enter the queue. If you are going to ask the question in Chinese, please follow with an English translation. Your first question comes from Timothy Zhao with Goldman Sachs.
First, this strategic upgrade from scale-driven expansion to efficiency-driven growth is the natural and the inevitable outcome of the evolution of our platform business. What this transition really means is, upgrade in the way value is created. By creating great value for customers, we aim to improve the penetration of community-based residential services, increase the commercial efficiency of resources, and ultimately drive the business growth. This is exactly the opposite of a logic of simply cutting capacity or contracting the business. To understand this evolution, we need to ask a more fundamental question, what truly determines the c apacity in our industry.
What are the core production factors and the production function? What customer truly needs is not simply more agents or more stores, but higher quality and more reliable decision support. This includes more precise matching, more effective marketing solutions, and more comprehensive home buying planning. What we are doing is reallocating resources from nominal capacity to effective capacity, concentrating our organizational efforts on the areas that generally solves customer problems. Against this backdrop, in 2025, we have taken several steps around our agent and store network. First, with the lead business, we have concentrated resources on high-performing stores and agents to improve operational efficiency. Going forward, we will further flatten our management structure so that managers with the strongest customer service capability can stay closer to the front line to create value.
At the same time, while codifying and embedding the high-quality service capacities into platform and as a division of labor system rather than leaving them dispersed among individuals. Second, on the broader platform side, we continue to expand the scale of agent stores, but with a greater emphasis on the quality and efficiency. By the end of 2025, the number of active connected store and agents continued to grow significantly year-over-year, increasing by 27% and 7% respectively. At the same time, while optimizing the structure of the network by identifying and amplifying the value of high-performing, high-rated store and agents. In the Q4 , agents activity improved sequentially.
In cities, including Beijing and Shanghai, the lead-to-conversion rate for its in-home sales increased by around 8% quarter-over-quarter, while average per agent commission income from existing and the new home transactions increased by 2% sequentially, while also improving the efficiency towards more platform-based capability, including AI-driven tools. Data and AI are the most important drivers behind this evolution. By reaching our data and AI capabilities, while redesigning many aspects of the platform, including resource allocation mechanism, the division of growth among the service providers and the service process for both homeowners and the buyers. In many areas, this represents a systematic redesign of how the platform operates. Meanwhile, as the housing decisions become more complex for consumers, our opportunity to create value grow.
Helping customers reduce the possibility of regret in one of their most important decisions of their lives, selling or buying a home or improving the overall service experience creates seeking value for each individual customer. At the same time, it helps reduce the friction across the entire market and potentially increases market turnover, effectively responding to the signs of our market. Therefore, to answer your question, the future growth and earnings elasticity of the platform will not depend on who has the market, the higher count or store count. Instead, it will depend on the expansion of platform's capability boundary, as well as those of the service providers operating on the platform. Building stronger professional services and higher overall efficiency, we aim to earn the trust and choice of more customers. Thank you.
Your next question comes from Zen Guo with Guangfa Securities.
Let me translate my question. My question is about the new home business. The new home market is facing multiple pressures, including developers struggling with sales through declining profitability and the increasing market concentration among state-owned companies. Management mentioned innovations in marketing models in the new home business. How will this innovation change the company dynamics and the relationship with the developers? How will the performance of the new home business be sustained? Thank you.
Thank you, Zen. Our view on the new home business starts from the structural change in industry. The level of the digital penetration in the new home sector remains relatively low. In the past, our operating model for the new home business was largely based on the traditional channel sales logic. This involved allocating resources around commissions and traffic and then leveraging our massive channel traffic to solve developers' sales issue for the core projects. This model was effective during the market's function phase, but under current conditions, its boundary are more limited. It can only serve certain projects and certain buyers, and the value creation for developers, and especially for home buyers, is relatively constrained. We believe the new home market is entering a new stage.
For home buyers, the concern is not whether there is enough information, but whether they can be more certain about the purchase decision. For developers, the key concern is no longer simply gaining another sales channel, but whether they can achieve more predictable sales results within a constrained budget. Accordingly, we are upgrading the role of our new home business from a channel player to an integrated capability platform. Number one, the level of online integration and the digitalization in the new home segment remains relatively low. This represents a common pain point across the industry, but also a significant opportunity for upgrading. We are working to enhance online decision-making support in new home journey through the stronger data and the product capabilities, truly helping customers solve their most difficult decision-making pain points.
Number two, we will further optimize allocation of the traffic resources by leveraging our data and system capabilities to improve the structural matching efficiency between projects and the potential buyers. We aim to expand service coverage among home buyers, both at the top of our funnel and ultimately improve the conversion. Number three, we view developers as our long-term clients rather than merely the channel sales partners. We aim to provide developers with the integrated solutions covering the product positioning, customer acquisition, matching, and the sales pace management, helping improve overall project efficiency and address key pain points. In the long term, our goal is for the new home business to evolve beyond the transaction distribution layer and then become an efficiency-enhancing platform between developer and home buyers.
As this capability evolves, our service offering and the revenue stream in the new home segment will become more diversified, and our business model will become more resilient. We believe this evolution will be critical to sustaining our long-term competitiveness in the new home market. Thank you.
Your next question comes from Miranda Zhuang with BofA Securities.
Thank you for taking my question. My question is about AI. With the recent rapid advancement of AI, how does the company view the potential impact of AI on the real estate sector? For Beike company, how is AI being used to empower the different business lines, and what are the progress so far? Thank you.
Thank you, Miranda, for your question. Recently, there have been many discussions about whether AI will bring a revolutionary impact to real estate brokerage industry. My view is that the key question is not whether AI will replace real estate agents, but rather how it will reshape the division of labor, value creation, and organizational structure of the industry. A housing transaction is fundamentally not a short, standardized consumption decision. Instead, it's a long cycle, multi-stage, and a highly complex decision-making process. From searching for a property, to making a decision, to completing the transaction, and then to moving in, operating the property and improve the living experience. There are many stages along the way where AI can significantly improve efficiency, and in some cases, even automate the process.
For example, information gathering, demand matching, process reminders, document generating, preliminary risk check, and workflow coordination are all standardized and repetitive tasks governed by clear rules. In this area, AI can deliver significant productivity gains. We have already begun to see some very tangible changes internally. For example, in housing transaction services, agents previously spend a large amount of time organizing property information, creating marketing materials, and responding to repetitive inquiries. With AI, we can now automatically generate AI video explanations, property interpretations, and the communication materials for clients, allowing agents to focus more of their time on understanding customer needs and supporting transaction decisions. In our rental business, AI is also beginning to participate in property acquisition decisions, rental pricing recommendations, and leasing matching.
By analyzing historical transaction data, regional supply and demand, and property characteristics, AI helps our operators more quickly determine whether a property is suitable for acquisition, recommend a reasonable rental range, and improve leasing efficiency while strengthening risk identification.
Take it together, these capabilities essentially allow standardized tasks to be handled by the system, enabling service professionals to focus more on complex decision-making and client service. At the same time, they are part of this value chain that are not easily replaced by AI. In fact, these areas may become even more important as AI develops. For example, someone still need to determine whether what a client say they want truly reflects their underlying needs. Someone need to make pricing judgments to dynamically coordinate between buyers, sellers, mortgage providers, title transferring process, and fulfillment risks. Someone need to stabilize expectations and feeling at the final stage of a transaction. Ultimately, someone need to take responsibility. The core of this task is not simply information processing, but judgment, coordination, trust, and accountability. This is where human value continue to lie.
Therefore, our view is that AI will effectively split the workflow of this industry into two parts. One part will become highly automatic, with efficiency improving rapidly. The other part will increasingly concentrated on professional expertise, accountability, and high-value services. From this perspective, the value of the traditional information intermediary will diminish, while the value of transaction responsibility and housing service infrastructure will become even more important. For Beike, this does not mean that opportunity becomes smaller. It actually become larger. Because what Beike aims to build is not simply AI-driven efficiency. Our goal is to leverage AI to further upgrade ourselves into a comprehensive housing service infrastructure. On one hand, we want information matching processes and the collaboration to become far more efficient. On the other hand, we want transaction responsibility, fulfillment assurance, and service delivery to become more reliable.
There is also another important characteristics of this industry. First, demand on the consumer side is difficult to fully articulate. Many clients cannot clearly express what they truly want at the beginning. Second, supply is highly non-standardized. Homes are not fully standardized or commoditized products. Their pricing, suitability, and risk level all contain significant uncertainty. Because demand is difficult to articulate and supply is highly non-standardized, this industry inherently requires people to interpret, explain, match, coordinate, and ultimately take responsibility. Looking further ahead, AI's impact goes beyond this. As AI significantly improve the efficiency of standardized process, people will increasingly become the key variable that determine the upper bound of efficiency. In the past, inefficiency was often constrained by process, tools, and information.
As these constraints are optimized by AI, the ultimate limit of our organization will increasingly depending on the capabilities of the service professional themselves. Moreover, the improvement in service professionals capabilities is not linear. It has clear leveraging effects. As AI raises the efficiency baseline of the system, stronger service professionals can generate disproportionately greater marginal value. In other words, AI does not weaken service professionals. It differentiates them, amplifies the best ones, and makes the upgrading of the service capabilities itself one of the most important growth levers. For organization, the real question is no longer whether they have AI, but whether they can organize people together with AI. In such an environment, organizational capabilities, collaboration mechanisms, culture, and value become increasingly important.
In highly efficient, transparent, and collaborating systems, it becomes even more critical to have a stable set of value judgments, unified service standards, and trusted behavior norms to connect every service professionals, every operational steps, and every interaction with customers. From this perspective, the continued evolution of this industry will not be driven by a single force, but by four forces working together: the power of technology, which drives efficiency improvements and capability expansion. The professionalism of service provider, with which determines judgment and service quality in complex scenarios. Customer trust, which determine whether transaction can actually be complete and whether long-term relationship can be formed. Organizational culture and values, which determine whether the previous three forces can be continuously integrated into a stable, scalable, and evolving system. In this sense, AI will indeed reshape the industry.
It will eliminate information asymmetry, compress low value competitive work, compress and amplify the value of professional services, transaction responsibility, customer trust, and housing service infrastructure. Ultimately, what determine how far a platform can go is not simply whether it has AI, but whether it can truly integrate AI, professional service providers, customer trust, and organizational culture into a continuously evolving model. Thank you.
Your next question comes from John Lam with UBS.
Stanley , my question is regarding the new media. How does the company look at the new media? Also how does the company look at some of the KOL utilizing new media to facilitate the property transaction? Thank you.
Thank you, John, for the question. Regarding the influence of influencers and public accounts on the company. My view is that whenever a phenomenon continues to attract the attention of customers, it usually reflects some real demand. Rather than judging whether it is positive or negative, the more useful question is which needs it is actually serving, which customers it resonates with, in which situations and what needs may not have been well addressed before. In my view, this also reflects a broader shift in the industry. In the past, the real estate industry was largely centered around the property itself. At that stage, the key question for many customers was simply whether there was a suitable home available and whether they could buy it. The property was the primary scarce resource, and customer decisions often revolved around the house itself.
The personal needs and circumstances behind the decision were not always fully reflected in the process. Today, the situation is changed. The industry is moving from being centered on properties to being centered more on people. Customers are not just home buyers in an abstract sense. Each decision reflects a set of real-life considerations, including family structure, budget constraints, lifestyle preference, education needs, commuting patterns, risk tolerance, and plans for improving or relocation. Buying, renting, or upgrading a home may appear to be a real estate decision, but in many cases, it is essentially a decision about how people want to organize their lives. From this perspective, housing transactions have always involved complex decisions. For a long time, however, the industry handled them more as a relatively light matchmaking process.
As customer needs become more complex and personalized, the decision is returning to its original nature. It requires understanding, explanation, judgment, and trade-offs. Against this backdrop, the rise of self media, influencers, and public accounts is not simply about new media channels. What they provide is a different form of value. Their focus is not just on the property itself, but on the person behind the decision. Through content, perspectives, and explanations, they help customers better understand the market, compare options, and reflect on their own needs. In doing so, they can help reduce decision costs and anxiety in making decisions. Customers follow them, but not just to obtain more information, but because they hope someone can help them make judgment, compare alternatives, and weigh different trade-offs. For a company, this phenomenon is both a reminder and an opportunity.
The reminder that we should no longer think ourselves simply as an information platform that matches people with listings.
You say we need to become truly customer-centered and focus on understanding the person behind the transaction. The opportunity is that if we can combine content capabilities, professional service capabilities, execution capabilities, and customer trust, we may be able to build a more durable and deeper competitive advantage. Influencers can provide our perspectives and influence. In complex transactions, the responsibilities for execution, risk management, and service delivery ultimately depend on a professional service system. Fundamentally, I do not think. The key question is whether self media will negatively impact, or not, a company like ours. Rather, what this phenomenon reminds us is that customers today need more than property information. What they increasingly need is decision support, professional judgment, and trusted work services centered around the individual. Companies that can better meet these needs will be better positioned in the long run. Thank you.
Your next question comes from Eddy Wang with Morgan Stanley.
Thank you management for taking my question. My question is regarding the renovation and furnishing business. We see the business has experienced slower revenue growth in 2025, but its gross margin improved. What's the current status of the development in supply chain centralized procurement and the standardization execution? When shall we expect to see the inflection point for the profitability in the home renovation business? Thank you.
Thank you, Eddy. The slower revenue growth in 2025 was the result of our deliberate decision to control the pace of expansion. Home renovation is a delivery-intensive business. If the underlying unit economics are not stable, aggressive expansion itself becomes a risk. The liquidity challenges of certain industry players in 2025 further reinforced this pain point for us. As a result, our priority last year was repairing and validating the underlying profitability structure of the business. From the results we have seen so far, the contribution margin has improved and overall losses have narrowed significantly, which indicates that unit economics at the individual project level are becoming healthier. At the current scale of roughly RMB 15 billion in revenue, we break down improvements in unit economics into three main variables.
Product structure optimization, controlling explicit costs such as materials, labor efficiency and delivery efficiency, and the reduction of implicit costs, including rework, up-sell issues, and the reputation-related losses. In 2025, our primary focus was the cost side. On explicit costs, centralized procurement across the supply chain has helped optimize our material cost structure. We have completed the centralized national or regional procurement tenders for approximately 80% of our key materials and about 60% of our auxiliary materials. This has strengthened our buying power, improved long-term product quality stability, and reduced exposure to the price volatility. At the same time, substantial improvement in work order dispatch mechanism and also adoption of the digital design and the modular tools, along with the records, admission, certification, and the rating system, we are building a dedicated tool for high-quality delivery teams.
This has meaningfully improved the productivity of both project managers and designers. On implicit costs, we are even more focused on long-term fulfillment quality. Through the measures such as the fund escrow, service commitments, and the greater standardization of and the collision detection in design phase. We aim to reduce rework and delivery variability at the source, thereby improving the stability of the profitability at project level. Looking into 2026, as unit economics continue to improve and our delivery capability become solidified, we plan to widen our funnel for the scale expansion in a disciplined manner. Our core approach is not simply to increase traffic, but to improve the traffic conversion efficiency. First, we will continue to optimize our product portfolio so that our offerings more precisely match the needs of different customer segments.
Second, we will replicate the high conversion showroom model built around our selling centers, better connecting existing home transaction with the renovation product experience and creating scenarios with higher certainty for decision making. Third, we will deliver the neighborhood focus and operating model into more cities. Leveraging collaboration between the brokerage agents and the renovation service providers within specific constraints to improve the overall agent conversion efficiency. At the same time, as delivery becomes more standardized and increasing costs continue to decline, improvements in customer satisfaction and reputation will create a positive feedback loop, providing a stronger foundation for the future scale expansion. Of course, it will take some time for this improvement to be fully reflected in our financial statements. Over the next two to three years, we will further integrate data flow across design, construction, and operations.
By leveraging BIM and the modular components and library to build productized capability, we aim to gradually transform the home renovation from a project-based business into a predictable and scalable industrialized capacity system. Thank you.
Your next question comes from Brenda Zhao with CICC.
Stanley and Tao , thanks for taking my questions. My question is related to the home rental business, because in the past two years, the business has developed rapid profit growth, which has been a pleasant surprise. However, revenue has been contracted quarter-on-quarter due to the impact of the accounting treatment. I believe it may be fairer to assess this business from the perspectives of long-term unit economics. How does the company view the long-term UE trajectory and the potential for improvement in this business? Thank you.
Thank you, Brenda. Regarding our rental business, I'd like to clarify two key aspects, the trend in business scale and the profitability structure. First, from accounting perspective, the short-term revenue contraction mainly results from the change in accounting treatment for the new product offering of our Carefree Rent business, which moved from gross revenue recognition to a net revenue recognition. Under the net method, we only recognize the service fee income, which more accurately reflect our roles as an asset management service provider. This accounting treatment, along with the lighter operating model and a substantially reduced risk profile under the new product offering. Importantly, this adjustment does not have a negative impact on our cash flow or the unit level profitability per managed property. If we look at the underlying business fundamentals, the scale of our managed rental unit continues to grow rapidly.
Both historically and looking ahead, the core operating metrics for this business, the number of units under management has maintained a strong expansion. At end of 2025, our managed home units exceeded 700,000, representing a year-over-year increase of 62%. This growth reflects improvements in product competitiveness and the expansion of the market demand rather than any accounting change. Second, in terms of the profitability model, we focus more on the continuous improvement of the unit economics at the single unit level. In 2025, the rental business turned profitable for the full year after previously operating at a loss. The improvement in the profitability was not driven purely by the scale expansion. Looking ahead, we continue to see the room for the profit growth driven by the structural improvement in unit economics. This improvement mainly come from several factors.
First, workforce productivity improvements, particularly improvement for the productivity of our key role, the property manager. In 2025, the average monthly number of the unit acquired per property manager increased by 7% year-over-year, while the number of unit managed per person increased by 42%. Second, lower customer acquisition cost per unit. This is driven by the improvement in channel efficiency and the higher net conversion rates, as well as stronger post-rental service experience and a higher customer satisfaction, which lead to improved renewal rates among both landlords and the tenants, thereby reducing the cost of acquiring new customers. Third, optimization of the product structure. As a proportion of our lighter asset management product increase, risk-related costs decline significantly. In particular, the upgraded product structure has made our profitability model much less sensitive to rental price fluctuation.
Overall, what we are seeing is a business where scale continues to grow rapidly, unit-level profitability continue to improve, and operational risk continues to decline. Looking ahead, the long-term evolution of unit economics in the rental business will primarily be driven by product structure, upgrade that improve profitability, stability and risk resilience, workforce productivity improvements, better channel efficiency, optimization of the customer acquisition costs, and the continued benefit of scale expansion. From a financial perspective, we are building the rental business into a segment characterized by sustained scale expansion, improving profitability quality, and increasingly stable cash flow. Thank you.
We are now approaching the end of the conference call. I will now turn the call back over to your speaker host today, Ms. Siting Li, for closing remarks.
Thank you once again for joining us today. If you have any further questions, please feel free to contact Beike's investor relations team through the contact information provided on our website. This concludes today's call, and we look forward to speaking with you again next quarter. Thank you and goodbye.