KE - Earnings Call - Q1 2025
May 15, 2025
Transcript
Operator (participant)
Ladies and gentlemen, thank you for standing by for KE Holdings Inc's First Quarter 2025 Earnings Conference Call. Please note that today's call, including the management's prepared remarks and question-and-answer session, will all be 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.
Siting Li (Director of Investor Relations)
Thank you, Operator. Good evening and good morning, everyone. Welcome to KE Holdings Inc's First Quarter 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. Tao Xu, our Executive Director and Chief Financial Officer. Mr. Xu will provide an overview of our strategies and business developments on behalf of Mr. Stanley Peng, our Co-founder, Chairman, and Chief Executive Officer. Mr. Xu will discuss the financials in more detail. 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. 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's 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.
Tao Xu (Executive Director and CFO)
Thank you, Siting. Hello, everyone. Thank you for joining Beike's First Quarter 2025 Earnings Conference Call. In the first quarter, our business continued to deliver rapid growth. This expansion was partially based on the market momentum that was fueled by the supporting policies since last September. It was also consistently driven by our active growth strategy that we started in the second half of 2023. In the fourth quarter, GTV on our platform increased by 34%, and revenue rose by 42%, both on a year-over-year basis. Our business continued to outperform the market in the fourth quarter across multiple metrics. Notably, GTV for our in-state home transaction business increased by 28% year-over-year in the fourth quarter. According to Beike Research Institute data, the year-over-year growth of the national GTV in this segment was about 60%.
GTV for our new home transaction business increased by 53% year-over-year, versus a 0.4% nationwide decline year-over-year reported by the NBS data. While the top 100 developers' GTV for new home sales also fell by approximately 7% in the fourth quarter, we continue to see strong momentum in the growth of connected stores and agent-on-order platform. In the fourth quarter, the number of active stores surpassed 55,200, a record high, increasing over 12,600 from the same period one year ago. Of those, the number of connected stores increased by more than 12,300. On the agent side, the number of active agents grew by 23% year-over-year, representing a net addition of over 90,000 agents compared with the same period last year, with active agents year-over-year to reaching a record high. We also see a steady improvement in efficiency at the store and agent level.
In the fourth quarter, GTV per store and per agent rose by 8% and 14%, respectively, making the fourth consecutive quarter of year-over-year increase. For connected stores, GTV per agent was up by 18% year-over-year, translating into stronger revenue for both stores and agents. Our platform's operation support ratio remained high, with impressive year-over-year improvement. This year, we are focused on driving both skill and efficiency as two priorities of our growth strategy. In the fourth quarter, traffic leads for our in-state home transaction services hit new records. The in general active market helped with driving more traffic leads, with additional benefits from the higher customer satisfaction from the search results and the more personalized recommendations. By tailoring the broad browsing experience to each user's scenario and profile, we made it easier for people to explore home listing on our apps that fit their needs.
This improving performance also reflects users' current preference to view more home listings before making purchase decisions. For our new home transaction business, this year, we are focusing on optimizing our collaboration with developers to better support their sales through needs, while improving agent efficiency in matching customers with suitable new home projects. In the fourth quarter, we concentrated our efforts on high-end projects in the market. At the same time, we continued to drive greater participation from our stores in the new home business through the incentive mechanism. Our One Body, Three Wings strategy maintains stronger performance traction. For the home renovation and furniture business, we have adjusted our pace this year to strategically focus on reshaping our products and delivery capabilities. Our primary goal is to make them more customer-oriented while streamlining our organizational structure for greater efficiency.
On the product front, we significantly advanced the design of our new home group renovation products in the fourth quarter. On the delivery side, we rolled out the project management perfectionism program in 20 cities. This drove a 156% year-over-year increase in average monthly order intake per project manager, reaching 2.97 compared with previously 1.16 in 2024. We also carried out a worker sharing model. As a result, top-performing project managers have seen improved personnel income, enabling them to focus on more full-service delivery and quality. In the fourth quarter, over 4% of our total home renovation projects came from referrals by previous customers. In addition, our front-end organizational management efficiency improved markedly. The average month's order volume per home renovation designer increased by almost 33%, moving from 0.79 in 2024 to 1.05 in the fourth quarter of this year, an outpacing total order growth year-over-year in the home renovation business.
Our home rental services continued to achieve the scaled brick source in the first quarter, with more than 500,000 rental units under our management. We also made solid progress in improving both default management and increasing our renewal rate. Last quarter, Stanley shared some thoughts on our AI deployment plan. Next, I'd like to provide an update on our use of AI in the first quarter. In our housing transaction business, on service-to-end customers, we conduct testing of our AI-powered home seeking assistant, [Putin] Intensities, which is already accessible to 40% of our traffic on our homepage. [Putin] was developed based on DeepSeek R1 and our massive platform data sets and proprietary knowledge graphs, while actually building an industry vertical database based on the larger model to improve [Putin's] smart response accuracy and improve multimodal display capability to optimize the interactions between the service provider and customers.
We believe a smart AI assistant will empower both homeowners and buyers with more intelligent solutions for the homemaking and decision-making, while also helping service providers identify more accurate leads. In terms of AI tools for service providers, our agent service home buyers will introduce the Lyco, an AI-based agent assistant. Lyco offers a full suite of features, including customer acquisition, AI home selection, AI chat assistance, and smart follow-ups. These tools empower agents to activate customers, enhance their professional service capability, and improve their efficiency in connecting with customers. By the end of March 2025, over 200,000 agents nationwide have used Lyco, collectively managed over 2.5 million customers, with impressive efficiency improvements. The commercial rate, from leads to formal client mandates, increased by over 30%, and the mandate-to-transaction commercial rate rose over 10%.
Agents effectively using Lyco achieved a land-to-transaction commercial rate that was three times higher than those not using this product. For agents serving homeowners, we identify a common issue. Many home listings were not being properly maintained on the platform due to agents' limited time and attention, which reduced the sales-through efficiency. To solve this, we leverage the AI property maintenance assistant, which helps agents manage listings more efficiently and improves the experience for homeowners. Within the homeowner-dedicated AI service group, the assistant offers smart replies, market trend insights, report analysis, and intelligent voice-based promotion. As of the end of March 2025, the product has been piloted by 110,000 agents and has served 400,000 homeowners cumulatively. Home listings maintained with our AI assistant achieved a transaction conversion rate four times that of those without it.
Additionally, our digital partner, Xiaoyi, utilized AI capability to enhance critical operational workflow, from contract quality inspection to ultimate post-selling follow-ups. The solution has given measurable improvements in front-end service quality and efficiency, delivering over 30,000 cumulative hours in productivity saving. In our home renovation business, we launched the AI customer maintenance tool to strengthen follow-up and lead conversion during the most critical two-week window in home renovation marketing. The AI-based lightweight beam and the intelligent marketing solution have improved efficiency in both design and marketing. For our home rental service, our AI assistant, for post-rental support, Xiao Hui, has been tested online in 30 cities. It is already successfully handling 25% of tenant requests through intelligent automation, providing tenants with a smart, more responsive service experience. At the same time, it enhances efficiency through better collaboration among the various roles involved in the recent progress.
I shared lots of numbers on the total volume and the average efficiency rate of our business, but these are not the key items we focused on. We care deeply about every individual customer's experience, and we remain committed to enhance our service quality. Since 2024, we introduced the fund custody system in our home renovation business, giving customers greater control and peace of mind. Under this model, renovation funds are frozen in customer personal bank accounts and only released to us after project milestones have been completed and approved by customers, including plumbing and electrical checks based on renovation and final acceptance. This model shifts away the traditional pay-first, renovate-later approach in the industry. Through the system integration, customers can track their funds online in real time with full visibility and traceability, and the interest earned during the custody period is returned to the customer.
In 2025, we rolled out our renovation fund custody service in several cities, including Beijing and Wuhan. On top of that, we have developed a fund custody solution plan framework that can be utilized by other industry peers, underscoring our commitment to driving the industry progress. Finally, we are encouraged by China's technical advances and are closely watching the evolving of the external market environment. While we remain confident in our platform ability to deliver sustained growth over the long term under our One Body, Three Wings strategy, we are approaching the short term with cautious optimism. That is why we still continue to invest firmly in AI while taking a more measured approach to other investments this year. Following last year's rapid investment and invest-wide subsidies, we are now setting clear short- and middle-term ROI benchmarks to ensure the disciplined capital allocation.
This balanced strategy will help us better position ourselves to capitalize on both market recovery opportunities and AI-driven generation productivity leaps and safeguard our operational stability, all while protecting the interests of the shareholders who share our long-term vision. In line with that commitment, this year, we will continue with active shareholder returns. Thank you. Next, I will review our first quarter 2025 financials. Once again, thank you, everyone, for joining us. Before we dive into our Q1 performance, I'd like to briefly touch upon some updates in the housing market. In Q1, the market performance was very stable, perpetuating the continued positive influence resulted from policy implemented in September last year. The threshold that's caused for the home purchase was further lowered, exerting a stronger incentive effort on home buyers.
According to the National Bureau of Statistics, new home sales remained relatively flat year-over-year in Q1, better than the substantial year-over-year decline in the same period last year. Meanwhile, the existing home market remained at a high level in activity, excluding the impact of the holidays. Benefiting from the readily available nature of existing homes, according to the Beike Research Institute, in Q1, existing home GTV rose by around 16%, and the number of existing home transactions climbed by around 28%, both year-over-year. With the growth in the transaction volume, the overall supply-demand relationship improved, and the housing price showed a signal of bottom-out, instilling more confidence in potential home buyers to enter the market. Demand for the upgrades was even more robust. Among existing home sales in key cities, the share of three-bedroom and larger homes continued to rise year-over-year in Q1.
Turning to our Q1 financial performance, our total GTV was RMB 844.2 billion, representing a year-over-year increase of 34%. Net revenue reached RMB 23.3 billion, up 42.4% year-over-year. Gross margin declined by 4.5 percentage points year-over-year to 28.7%. GAAP net income was RMB 855 million, increasing 97.9% year-over-year. Non-GAAP net income reached RMB 1.39 billion, remaining stable year-over-year. Looking at our housing transaction services, revenue from existing home transactions reached RMB 6.9 billion in Q1, up 20% year-over-year and down 23% quarter-over-quarter. GTV was RMB 580.3 billion, rising by 28.1% year-over-year and declining by 22.1% quarter-over-quarter. GTV growth outpaced the revenue year-over-year, mainly due to a decline in the revenue share of the rental brokerage services and the high contribution from the existing home transaction service GTV facilitated by connected agents. The revenue and recorded as net revenues derived from platform services.
The contribution margin from existing home transaction services was 38.1% in Q1, representing a decline of 6.4 percentage points year-over-year, primarily due to the increased support and the improved welfare for the service providers. This is our long-term strategy to build a harmonious ecosystem. Sequentially, the contribution margin dropped by 2.3 percentage points, attributable to negative leverage influence due to the decline in revenue exceeding that in fixed labor cost. In terms of the new home transaction services, we still outperformed the market. SRC reports that the sales from the top 100 developers fell by around 7% year-over-year and 41% sequentially in Q1. In comparison, our new home GTV reached RMB 232.2 billion in Q1, up 53% year-over-year and down 34.6% quarter-over-quarter, once again outperforming the industry.
This was mainly due to the deepening of our collaboration with developers and our finely tuned operational capability and more sales confirmation from the partial subscription in last quarter. Revenue from new home transactions was RMB 8.1 billion in Q1, rising by 64.2% year-over-year and dropping by 38.2% from previous quarter. Revenue outperformed the GTV year-over-year, demonstrating our stronger maintenance capabilities. While GTV growth outpaced revenue growth sequentially due to the seasonality, the contribution margin from the new home transaction services rose by 1.1 percentage points year-over-year to 23.4% as we gained leverage from the revenue growth exceeding that of the fixed cost. Sequentially, the new home contribution margin declined by 2.2 percentage points, largely attributable to the seasonality effect. In Q1, SOE developers contributed around 54% of our new home sales revenue, increasing by around 4 percentage points year-over-year.
Revenue from home renovation and furniture, home rental service, and emergent art services grew by 46.2% year-over-year in Q1. It accounted for 35.9% of our total revenue compared to 35% in the same period last year. The contribution profit from this business accounted for 32.7% of our total gross profit. Revenue reached RMB 4-hour home renovation and furniture business. Revenue reached RMB 2.9 billion, increasing by 22.3% year-over-year, mainly due to the increased orders from the home renovation. Contribution margin for the home renovation and furniture business reached a record high of 32.6%, up 2 percentage points year-over-year and 2.8 percentage points quarter-over-quarter, mainly driven by the increased gross margin of our home renovation business. Our home rental services business continued to grow at an accelerated pace in Q1.
Its revenue reached a record high of RMB 5.1 billion, up 93.8% year-over-year, mainly benefiting from the rapid growth in the number of the rental units under management. By end of Q1, the number of the rental units under management exceeded 500,000, compared with over 250,000 in the same period of 2024. The contribution margin for the home rental services was 6.7%, up 1.2 percentage points year-over-year and 2.1 percentage points quarter-over-quarter, largely due to the improved gross profit of our carefully run business. As we continue to refine the carefully run business model based on the essence of the business contract, the revenue from some newly managed rental units was recorded as net revenues derived from the service fee in this quarter. In Q1, our revenue from emergent and art services decreased by 50% year-over-year and 28.3% quarter-over-quarter to RMB 350 million.
Next, let's move on to our after-cost expenses in Q1. Our store cost reached RMB 770 million, remaining relatively stable year-over-year and dropping by 8.8% quarter-over-quarter. The sequential decrease was mainly from the lower store rental cost. After-costs were RMB 547 million, up 44.4% year-over-year, primarily due to the increased tax and surcharge and the financial service reserve and the credit losses. Sequentially, after-costs declined by 26.7%, largely driven by the decreases in the tax and surcharge, financial service and reserves and the credit losses and the share-based compensation. Gross profit rose by 17% year-over-year to RMB 4.82 billion. Gross margin was 28.7%, down 4.5 percentage points year-over-year. The primary reason for the decline was decrease in contribution margin from existing home transaction services. Gross margin fell by 2.4 percentage points sequentially in Q1, mainly due to the structural reason as the revenue contribution of new home transaction service declined.
In Q1, our GAAP operating expenses totaled RMB 4.2 billion, up 2.9% year-over-year and down 31.3% sequentially. Lastly, G&A expenses were RMB 1.9 billion, decreasing by 7.2% year-over-year, mainly due to the reduced share-based compensation expenses. G&A expenses dropped by 36.7% quarter-over-quarter, primarily attributable to the lower personnel expenses and the decreased bad debt provision. Sales and marketing expenses increased by 9.2% year-over-year to RMB 1.8 billion, resulting from the increased expenses for the home renovation and furniture business. Quarter-over-quarter, sales and marketing expenses fell by 24.4%, mainly due to a decline in marketing expenses for home transaction services and reduced personnel expenses. Our R&D expenses were RMB 584 million, up 24.9% year-over-year, driven by higher personnel expenses and technical service fee. Sequentially, R&D expenses dropped by 27%, largely as a result of reduced personnel expenses.
In terms of the profitability, GAAP income from operation totaled RMB 591 million in Q1, a remarkable increase compared with the same period last year, and decreasing by 41.6% sequentially. GAAP operating margin was 2.5%, increasing by 2.5 percentage points from Q1 2024 and falling by 0.7 percentage points quarter-over-quarter. Non-GAAP income from operations totaled RMB 1.15 billion, growing by 19.6% from the same period last year and dropping by 34.6% sequentially. Non-GAAP operating margin reached 4.9%, down 0.9 percentage points from Q1 2024 and 0.7 percentage points from the previous quarter, mainly attributable to the gross margin decrease both year-over-year and quarter-over-quarter. GAAP net income totaled RMB 855 million in Q1, rising by 97.9% year-over-year and 48.2% quarter-over-quarter. Non-GAAP net income was RMB 1.39 billion, remaining stable year-over-year and increasing 3.7% quarter-over-quarter.
Moving to our cash flow and balance sheet, we realized a net operating cash outflow of RMB 4 billion in Q1. New home DSO reached 63 days in Q1, remaining at a healthy level, on top of approximately $139 million allocated to share repurchase during Q1. Our total cash liquidity remained at a high level of RMB 74.3 billion, which outscored the customer deposit payable. With our robust cash reserves, we continue to reward our shareholders who have grown reserves through the active share buyback, enhancing capital operation efficiency and sharing the benefit of our development with investors. In Q1, we repurchased around $139 million worth of shares, which accounted for around 0.6% of the company's total shares outstanding at the end of 2024. We have consistently delivered on our promise to reward shareholders.
Since the launch of our share repurchase program in September 2022, we have repurchased roughly $1.76 billion in shares at the end of Q1 2025, accounting for around 9.2% of our total shares outstanding before the program began. This year, our business will focus on efficiency improvements. In financial strategy, we will ensure that our investments are made more efficiently to improve personnel and store productivity. We will respect every penny and make sure the money spent yields visible results. While maintaining the discipline in the cost and expense control, we will continue to support the long-term business development by fully backing our One Body, Three Wings strategy initiatives and actively exploring AI technology. At the same time, we possess natural resources and the intention to consistently offer the stable and sustainable returns to our shareholders. This concludes my prepared remarks for today.
Operator, we are ready to take questions.
Operator (participant)
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 speakerphone, 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're going to ask a question in Chinese, please follow with an English translation. Your first question comes from Timothy Zhao from Goldman Sachs. Please go ahead.
Timothy Zhao (Analyst)
[Foreign language] Thank you, Megan, for taking my questions. I think my question is regarding the outlook for the property market going forward. I think we note a strong rebound of the property market transactions after twice new year. Just wondering, from second quarter and onwards, what is your outlook for the property market? Is it considering the latest macro dynamics as well as the impact from the U.S. tariff? Thank you.
Tao Xu (Executive Director and CFO)
Thank you, Timothy. For Q1 market performance, with the central government continuing to effort to stabilize the real estate market, the existing home market saw a relatively strong recovery after twice new year. The new home market also performed stable. Let me give some details. The existing home market rebounded after twice new year as expected. According to Beike Research Institute, nationwide, existing home GTV grew by 16% year-over-year in Q1, sustaining the momentum fueled by the September 26 policy stimulus. This growth was mainly due to the cumulative impact of the early stimulus policies. These policies substantially lowered the threshold and the cost of the home purchase, motivating more people to buy. Rising transaction volume also helped balance supply demand in the short term, narrowing the decline in the home prices year-over-year and bringing the cautious buyers back to the market.
Existing home prices have stayed most stable, dipping by 0.5% month-over-month in March. In first-tier cities like Beijing and Shanghai, as well as second-tier cities with strong net population inflows and a steady increase in the housing demand in recent years, such as Hangzhou and Chengdu, existing home prices have picked up slightly month-over-month. For new home market in Q1, it is also very stable. According to NBS data, Q1 new home sales were overall flat year-over-year, down 0.4%. GDP of 1,200 real estate developers dropped by 7% year-over-year in Q1. Notably, the sales by floor area increased by over 15% year-over-year. Since Q2, after twice new year, the market followed its typical seasonal pattern. The existing home transaction volume peaked in early March, then gradually declined through April. The month-over-month decrease of existing home prices responded somewhat with a 1.3% drop in April as transaction volume reduced.
From a supply demand standpoint, the total number of existing home listings on Beike rose in Q1, outpacing in Q4 of last year. This aligns with the seasonal trend of lower inventory at the year-end and higher inventory at the start of the year. It is also a natural result of the existing home market dysfunction and the lifting of the sales restrictions, which have released more housing supply. The faster supply of the nearly new existing home into the market has also improved overall listing supply quality and created better conditions for buyers looking to the home upgrades. Meanwhile, we observed that the increase in the market demand outpaced the increase in inventory. In April, the ratio of the home viewing to inventory was at 1.8, which is at the higher end of the historical range of 1.6-1.9.
This suggests a stronger buyer interest and plentiful demand, with the market able to absorb new inventory. However, the conversion from the home viewing to transaction has slowed, mainly due to the short-term uncertainties affecting buyer expectations, including external factors such as geopolitical tensions. This has led to some softening of the housing price expectations, which made buyers more hesitant to enter the market. For future market outlook, we believe the market outlook will depend on two main factors: the impact of the international trade frictions on housing transactions and the strength and timing of domestic countermeasures. On a neutral scenario, we expect a typical seasonal slowdown in Q2 on a sequential quarter basis. Year-over-year, however, the existing home market expects to see a slight increase in existing home transaction volume, although at a slower pace than Q1.
This is supported by the higher transaction volume in Q1, the growing supply of the high-quality, nearly new existing home, and the year-over-year increase in customer home viewing. As the market experiences a quarter-on-quarter decline in Q2, and if external trade pressure intensifies in Q3, indicators such as housing price, transaction volume, and development investment may weaken. This could create room for further supportive policy measures in the second half of the year, which would help improve both supply and demand in the property market and support stable market development. We are also closely monitoring the impact of the change in global trade on the real estate market. In terms of home listings, the overall number of new listings on Beike platform remained stable in April, with no signs of the homeowners rushing to sell.
The number of the home viewing for both existing and new home still showed a notable year-over-year increase in April. When categorizing the cities that we covered by high and low trade dependency, we observed that since the tariffs turned off in early April, cities with high trade dependency have shown weaker year-over-year and month-over-month home viewing performance compared to cities with low trade dependency. This indicates that while the trade frictions have caused some short-term disruption in buyer expectations in certain cities, there has not been a significant trend of divergence overall, and the homeowner sentiment remained stable. Moving forward, we will continue to monitor potential impact of the trade frictions on the housing market through the leading indicators such as home viewing, customer traffic, and listing volume.
We observed a notable de-escalation in recent U.S.-China trade tensions, which should help stabilize the business and the consumer expectations in the near term. In our view, we believe that the intentional trade frictions represent a long-term dynamic process with uncertainties and the potential for improvement. During the upcoming 90-day negotiation window, we will closely monitor the development, track the resulting impact, and assess the potential implications for both real estate market and our company on an ongoing basis. In the medium to long term, we maintain a cautiously optimistic outlook, trusting both China and the U.S. will continue to multilaterally support each other based on the positive progress made so far. Meanwhile, the continued implementation of domestic supportive policies is expected to further boost customer confidence.
Together, this is expected to mitigate the impact of the trade risk on the property market, helping to consolidate the initial stabilization of the existing home market and ease pressure on the new home market. Thank you.
Operator (participant)
Thank you. Your next question comes from Xiaodan Zhang from CICC. Please go ahead.
Xiaodan Zhang (Analyst)
[Foreign language] Thanks, management, for taking my questions, and my question is regarding the housing transaction services business. Could management elaborate on the expansion plan for this year in terms of housing agents and the agency stores? On top of that, how will you continuously improve the efficiency of those existing and newly connected agents and stores on the platform? Thank you.
Tao Xu (Executive Director and CFO)
Thank you, [Sophie]. Good to hear. We will continue to promote healthy growth, our agency network, to support the sustained expansion of our housing transaction services. At the same time, we will place greater emphasis on the cost-effectiveness of the stores' expansion. Our aim is to enhance the efficiency and the income of platform stores and agents, thereby increasing the stability of agent careers, providing better services to customers, and achieving more sustainable long-term growth for our platform. In terms of the agent and store network expansion, at the end of this Q1, the number of active stores on our platform increased by nearly 30% year-over-year, and the number of active agents grew by 23% year-over-year. The growth was mainly driven by the non-retail segment, with a 33% year-over-year increase in active non-retail store and a 24% year-over-year increase in active non-retail agents in our platform.
In Q1, several major brands joined our platform. This includes our collaboration with [Zhongyuan] sub-brand [Baiyuan] and [Kunjin] Brokerage Company in Kunshan and Suzhou. This shows the core value of our platform in the buy-side market, which is our stronger existing home business operation, agent connection network, and digital empowerment capabilities. In Q1, our efficiency efforts paid off in the stable market environment. The average number of transactions per agent rose notably in Q1. This helped offset the decline in average housing unit prices. As a result, existing home GTV per agent grew by over 9% year-over-year in Q1. The average number of agents also rose by 18% year-over-year. Together, this factor led to a 28% year-over-year increase in existing home GTV on our platform, clearly outperforming the 16% increase in national GTV as estimated by Beike Research Institute. Our platform's efficiency-focused mechanism also started to show results.
The share of the high-performing stores increased from 16.7% at the end of 2024 to 18.4% in Q1. On average, the stores were about 2.5x more productive than that of those stores on the platform in their respective cities. To improve the efficiency, we refined our internal measurements. We used digital tools such as online store owner workshop and AR property listing assistant, along with offline property listing sessions to facilitate the separation of home listings that accelerate sales. We also improved platform operation through the building mechanisms like point-based incentive programs and regional co-governance council. This encourages store owners to keep growing business and work more closely with each other. Our store retention rate remained healthy for both old and new stores. Our existing home retention rate dropped to 2.9% in Q1, down 6% sequentially and 38% year-over-year.
The six-month retention rate for newly connected stores in the first half of 2024 was 94%, showing the long-term value of our platform support. For the full year, we reasonably foresee that the number of net agents and stores will remain largely stable. Meanwhile, we expect a modest increase in the scale of non-retail agents and stores, with targeted expansion in certain key regions. On top of the stable agent and store networking, improving efficiency will be our core goal this year and beyond. This year, we will provide more targeted support to store owners to help them improve regional competitiveness. At the same time, under our points-based incentive system, we aim to develop more high-performing stores, aiming to upgrade the overall structure of our store network. In the long run, the large store model will be a key strategy for enhancing productivity.
In the future, our platform will host more high-performing large stores, each with over 10 agents. These stores will attract more top talents. This large store boosts high efficiency and strong staff retention, allowing store owners to achieve better income and stay in business longer. Those store owners can better support agents, ensure their income stability, and enable the owners to provide superior service to customers. The platform's various residential service will also offer agents diversified opportunities for additional income. Additionally, we firmly believe that Beike's tools in AI will present opportunities for transformative improvements in industry productivity. We have already developed a variety of AI applications to support our service and providers, and we will continue to accelerate development to redefine the capabilities of the quality service providers and drive their efficiency gain.
This year, in a volatile market, we aim to increase the average number of transactions per connected agent to maintain stable per capita commissions. Over the next two to three years, we plan to increase the proportion of large and high-quality stores. This store will have more stable, high-performing agents with high efficiency, with store productivity being two to three times of the current average. We anticipate within three years, this will lead to approximately a 20% improvement in efficiency of those connected agents on our platform. Thank you.
Operator (participant)
Thank you. Your next question comes from Jizhou Dong from Nomura. Please go ahead.
Jizhou Dong (Analyst)
[Foreign language] Beike's Home Renovation and Refurbishment business achieved 20+% year-on-year growth in the first quarter, with 2 percentage points improvement in contribution margin. Could management share more specifics of the segment's operation as well as the outlook on the margin in the future? In addition, management has shared a lot of the ideas of Beike using AI to drive its business and improve its service quality. Looking into the next few quarters, could management update us more on Beike's strategy and investment plan for AI from both the 2B and 2C perspectives? Thank you.
Tao Xu (Executive Director and CFO)
Thank you, Jizhou. Our home renovation and furniture business demonstrated excellent performance in this Q1. In terms of scale, the revenue amounted to RMB 2.9 billion, up 22.3% year-over-year. Cities such as Beijing, Guangzhou, and Zhengzhou performed especially well, and each achieved over 50% year-over-year growth in revenue. Regarding profitability, the contribution margin for the home renovation and furniture business reached 32.6% in Q1, an increase of 2 percentage points compared to the same period last year, reaching a record high and reflecting our capability of the refined operation and management. We believe that the AI has extensive application scenarios in the home renovation and furniture. We are also continuously deepening the application of AI, such as the contract conversion, construction process, and the internal management. Let me elaborate for details.
First, in the early stage of the contract conversion, previously, designers conduct the initial communication with customers through the two-dimensional black-and-white flow plans. The overly professional drawing reduces the customer's perception and affects the efficiency of the contract conversion. Currently, empowered by AI, when customers visit our offline store for the first time, designers can directly formulate an AI proposal based on the customer's preference for decoration style and the home layouts. This AI proposal encompasses the various types of the three-dimensional color-rounded design drawing, dynamic and static space analysis, storage, and smart device layout plan. This significantly enhances the experience of first-time store visitors and thereby boosts the contract conversion rate. Taking Wuhan as an example, the time period from first-time visiting the store to signing a preliminary contract would shorten from previously 10 days to within 6 days in March.
Secondly, in the construction process, we have developed an intelligent construction system. Real-time online inspections are realized by installing cameras on the spot. AI can also realize automatic measurements in core construction operation, such as the real name on the job certification, the inspection of the site cleanliness, and the noise recognition. In addition, by equipping staff with smart inspection devices, we assist in standardizing the home renovation acceptance process through AI recording technology. We can recontrast the acceptance process, enabling traceable and quantifiable evaluation of the construction quality. Taking Beijing as an example, the acceptance of accuracy rate has increased by more than 2 percentage points compared with before. Meanwhile, in terms of the internal management, we have multiple AI employees.
Sun Xiaosheng, as the operating management AI employee, enhanced the effectiveness of the team management by automatically summarizing and commenting on daily reports and reminding about the pending matters through AI. Sun Xiaosheng has collected and commented on more than 20,000 daily reports, saving the team more than 18,000 hours within half a year. [Xiao Caixun], as the order-following AI employee, realized functions such as information distribution and automatic order assignment through the information collection and AI analysis capabilities. [Xiao Caixun] had to distribute information over 5,000 times and send timely reminders over 10,000 times within half a year. In the future, our AI exploring for home renovation business will be concentrated on the following aspects: more accurate insights and analysis of the customer demands, and a more efficient design powered by AI.
We aim to achieve the better personality solution from demand to design and comprehensively enhance the professionalism and the efficiency of our services. Thank you.
Operator (participant)
Thank you. Your next question comes from John Lam from UBS. Please go ahead.
John Lam (Managing Director and Head of China and Hong Kong Property Research)
[Foreign language] Let me translate my questions. My question is regarding on Beihaojia. We see that Beihaojia has already participated in numerous new home projects. I just want to see how Beihaojia contributes to the new home development, and also regarding on the C2M business, how the business is being reflected. Thank you.
Tao Xu (Executive Director and CFO)
Thank you, John. Beihaojia's business model provides a Sichuan new home product solution for partners like developers. We use intelligent algorithms and a massive database to deeply understand our target customers' needs and preferences. Our tools help predict the type of home customers wanted and the price they expect. Developers use this insight to guide the project positioning and the product design, making their new home offering more catered to customer demands. So far, Beihaojia has participated in nine projects across different models. Through our self-operated project, aim at more comprehensive validating of C2M capabilities. Five projects involve equity partners, which Beihaojia primarily focuses on the product's positioning using C2M. The rest two projects are purely light asset model, where we do not engage in investment, but instead provide a product positioning solution to partners and charge a service fee.
Regarding the funding use across the seven investment-involved projects, the total investment has reached about RMB 2.3 billion. By the end of this Q1, we have recovered nearly RMB 500 million from [Chuosin]. Net investment from our own fund stands over RMB 1.8 billion. Among these, our first equity partnership, new home projects, which we collaborated with Power China Real Estate on in Beijing, Chang'an Huaxi Mansion in Wintergo district of Beijing City, achieved a complete sale-out of all initial units on the first day of launch. The project delivered an IRR nearly 30% at the shareholder level, demonstrating how our C2M service capability provides partners with enhanced sales and operation certainty. In building C2M capabilities, we have two key advantages. First, we have a deep understanding of the needs of potential consumers. This came from our unique database built on massive online and offline traffic of our platform.
It also draws from the rich customer interactions in our brokerage, home renovation, and rental services. Together, this forms the foundation of our core data infrastructure. With this data, we can analyze key indicators such as the source, quantity, purchasing power, and the specific product needs of potential customers in a more timely, intuitive, and competitive manner. This level of detailed customer insight helps developers make more accurate decisions and allocate resources more effectively in areas such as land acquisition assessment, unit mix planning, and product design, ultimately leading to greater operational certainty. Our second advantage is strong market knowledge and price ability. We use actual transaction price of the listing home along with the real-time and upstream data like homeowners' listing price and the price adjustment. We developed an algorithm model to build a valuation model for different geographic districts.
These models can more accurately estimate the project and sector value and update quickly based on the change in the market. This approach aligns more closely with the price formation logic in a buyer's market. Listing home prices are largely unaffected by policies or developers' strategies and instead result from decentralized and free negotiation between buyers and sellers. In a market dominated by listing home transactions, listing home prices often give a more accurate, multi-layered view of the market than new home market data. This helps us better understand the market trend. Based on these two advantages, we have refined our core C2M tools. We will continue to improve their accuracy and professionalism over time. We also support our C2M model with more innovative ways to reach customers.
While our agent network connects us with customers, we are also building an online community called Building a Better Home Together with our application. This enables users to directly participate in evaluation and core creation of new home product designs. For example, in our function project in Shanghai, users can visit Building a Better Home Together page. They can view and compare two home design plans, engaging in the design process of their future dream homes. This mode lets us connect with customers much earlier than traditional methods. It also makes their preference reflect in the new home products. We have already provided our first C2M product solution service in a Xi'an project, through which we earned a service fee, showing stronger market recognition from our business approach. In Xi'an project, Beihaojia provides a full set of product solutions.
This included customer surveys, product design positioning, cost optimization, price forecasting, and market services. We also give targeted project planning advice, such as optimizing the elevator-to-hospital ratio, enhancing the landscape, and adjusting unit size, trying to address the key developer pain points, such as fast capital recovery, product premium, and product competitiveness. Last but not least, as a newcomer to the industry, Beihaojia remains humble and respectful to the market. Although the business plan has been established for less than two years, we have already seen the promising results in several projects. This early stance has gradually validated our capability path and strengthened our confidence in continuous optimization and moving forward. Thank you.
Operator (participant)
Thank you. We are now approaching the end of the conference call. I will now turn the call over to your host today, Ms. Siting Li, for closing remarks.
Siting Li (Director of Investor Relations)
Thank you once again for joining us today. If you have any further questions, please feel free to contact the Consumer Relations team through the contact information provided on our website. This concludes today's call, and we look forward to speaking with you next quarter. Thank you and goodbye.