This conference call transcript was computer generated and almost certianly contains errors. This transcript is provided for information purposes only.EarningsCall, LLC makes no representation about the accuracy of the aforementioned transcript, and you are cautioned not to place undue reliance on the information provided by the transcript.
3/20/2024
Good day and thank you for standing by. Welcome to the Kings of Clouds fourth quarter and fiscal year 2020 free earnings conference call or webcast. At this time, all participants are in listen-only mode. After the speaker's presentation, there will be a question and answer session. To ask a question during the session, please press star 1 and 1 on your telephone. You will then hear an automated message advising your hand is raised. To withdraw your question, please press star 1 and 1 again. Please note that today's conference is being recorded. I would now like to turn the conference over to Nicole Shan, IR Manager of Kings of Cloud. Please go ahead.
Thank you, operator. Hello, everyone, and thank you for joining us today. Kings of Cloud's fourth quarter and fiscal year 2023 earnings release was distributed earlier today and is available on our IR website at ir.ksyun.com, as well as our global newsware services. On the call today from Kingsoft Cloud, we have our Vice Chairman and CEO, Mr. Zhou Tao, and CFO, Mr. Hongrui He. Mr. Zhou will review our business strategies, operations, and company highlights, followed by Mr. He, who will discuss the financials and the guidance. They will be available to answer your question during the Q&A session that follows. There will be consecutive interpretations. Our interpretations are for your convenience and reference purpose only. In case of any discrepancies, management statements in the original language will prevail. Before we begin, I'd like to remind you that this conference call contains forward-looking statements within the meaning of Section 21E of the Securities Exchange Act of 1934, as demanded and as defined in the U.S. Private Security Litigation Reform Act of 1995. These forward-looking statements are based on management's current expectations and current market and operating conditions. and related to U.S. study involve known or unknown risks, uncertainties, and other factors, of which are difficult to predict and many of which are beyond the company's control, which may cause the company's actual results, performance, or achievements to differ maturely from those in the forward-looking statements. Further information regarding these and other risks, uncertainties, or factors are included in the company's filings with the U.S. SEC. The company does not entertain any obligation to update any forward-looking statements. as a result of new information, future events, or otherwise, except for required under applicable law. Finally, please note that, unlike otherwise stated, all financial figures mentioned during this conference call are denominated in RMB. It's now my pleasure to introduce our Vice Chairman and CEO, Mr. Zhou. Please go ahead, Zhou Zhou.
Hello, everyone. Welcome to the 4th quarter of 2023 and the full-year business call conference. In 2023, it is a year in which the principles of comprehensive observation of high-performance sustainable development are achieved. In the whole year of 2023, after adjustment, the net profit rate reached 12.2%, which is 5.4% in 2022, which significantly increased by 6.8%. After adjustment, the net profit amount reached RMB 8.6 billion, which is almost half of the 4.5 billion in 2022. After normalization, the return rate is negative 3.4%, which is negative 8.9% in 2022. Over the past year, we have insisted on technical success, mouthpiece as king, management as steel, and good internal work, strategic adjustment of business structure, and positive embrace of the new AI era. For the future and in 2024,
Hello, everyone, and thank you all for joining TeamSubCloud's fourth quarter and fiscal year 2023 earnings call. In 2023, we continued to uphold the principle of high quality and sustainable development and accomplished significant achievements. For the full year of 2023, adjusted growth margin was 12.2%, a significant increase of 6.8 percentage points. up from 5.4% in 2022. Adjusted growth profit was RMB 860 million, almost doubling the amount of RMB 445 million in 2022. Normalized adjusted EBITDA was negative 3.4%, another significant improvement from negative 8.9% in 2022. During the year, we started building our success based on technology and innovation forging our reputation throughout the entire business process, enhancing our operations management, and building inner strength. We have been strategically adjusting our business mix and proactively embracing the new AI era, therefore laying solid foundations to long-term sustainable development in 2024 and beyond.
Next, I would like to focus on the performance of the fourth quarter of 2023. This quarter, Xinshanyun has achieved a growth rate and double growth in profit and profit capacity. Income reached RMB 17.2 billion, and interest rate increased by 6%. After adjustment, the interest rate reached 15.2%. Compared with the previous quarter, it increased by 3.1%. It achieved a steady increase of 6 consecutive quarters, creating a new high in history. Now, I will walk you through the business highlights of the fourth quarter 2023. This quarter, we achieved dual improvement in both our revenue and profitability measures.
In particular, total revenues reached RMB 1.72 billion, increasing 66.0% quarter over quarter. Adjusted gross margin recorded a major increase of 3.1 percentage points quarter over quarter to 15.2%, marking the sixth quarter of consecutive improvement. Adjusted gross profits reached RMB 262 million, increasing 55.8% year over year. Normalized adjusted EBITDA margin was negative 1.6%, representing a significant improvement of 8.5 percentage points year over year.
In terms of public cloud, this quarter, the revenue of 10.5 billion yuan has increased by 3.5%. In addition to CDN business, public cloud revenue has increased by 11.8%. In Xiaomi Jinshan Ecology AI business, CDN strategic management in three aspects. First of all, as the only strategic cloud platform of Xiaomi Jinshan Ecology, to serve the user needs of eco-friendly enterprises, especially in the context of Xiaomi's car WSAI, there is a high degree of visibility and a wide range of training and reasoning needs to grasp the opportunity in the wave of industrial structural development. This quarter, The revenue contribution from Xiaomi and Jinshan reached 16%, which increased by 5%. Secondly, we continue to develop our AI business and actively respond to the changes in the industry. In addition to the opportunity for AI business in Xiaomi Jinshan ecosystem, we fully cover a group of independent AI companies that are the most leading in China. To provide them with a long-term guarantee of high performance and long-term security AI income this quarter accounts for 8% of the total net profit and 82% of the net profit growth In addition, the company signed a loan agreement with Jinshan Software to obtain a loan amount of 1.5 billion yuan, which is dedicated to the development of AI business Third In terms of public cloud services, revenues were RMB 1.05 billion,
representing an increase of 3.5% quarter-over-quarter. Excluding CDN business, public cloud revenue increased by 11.8% quarter-over-quarter. We bear fruit on our three priorities for public cloud services, namely the Xiaomi and Kingsoft ecosystem, AI business, and CDN strategic adjustments. First of all, as the sole strategic cloud platform within the Xiaomi and Kingsoft ecosystem, we continued to serve the demands in the ecosystem well, especially with respect to the highly visible and tangible demands in training and inference from Xiaomi EV and WPS AI, seizing opportunities in structural industry trends. This quarter, revenues contributed by Xiaomi and Kingsoft ecosystem reached 16%, representing an increase of 5 percentage points year over year. Secondly, we vigorously developed our AI business with remarkable agility to emerging industry trends. Besides the AI opportunities from the Xiaomi and Kingsoft ecosystem, we have been fully penetrating leading independent AI companies in China, providing long-term, secured, high-performance computing power, which is highly sought after in the market. Our AI business represented approximately 8% to our public cloud revenues, or an increase of 82% quarter over quarter. Besides, we have signed a loan facility agreement with Kingsoft Corporation, obtaining a credit line of RMB 1.5 billion, dedicating to the development of AI business. Thirdly, we continued to push forward our strategic adjustments of CDN business. This quarter, CDN revenue decreased by nearly 10% compared to last quarter, and CDN revenue as a proportion of total revenue has decreased to approximately 23%. The revenue share of our largest CBN customer was 12% in this quarter. Top customer concentration has been fundamentally alleviated.
In terms of sales, we achieved a revenue of 6.7 billion yuan this quarter, a 10.2% increase in return. In the public service sector, we are actively taking advantage of the project opportunities of Zhengwu Yun and Guo Ziyun. In terms of the two types of business applications, with large model, large data, co-operative office as the base to carry out standardized operation. We use Beijing, Wuhan, Zhuhai as the base to build the benchmark project to replicate other large and medium-sized cities. This quarter, we successfully joined hands with Shenzhen State's Cloud to assist local state-owned enterprises in digitalization development. In the field of digital health, we continue to advance the five major models. to achieve new breakthroughs. We inherited and developed cooperation experience with Shanghai Ruijin Hospital, and expanded the medical digitalization capability in more famous hospitals, including Wuhan Xihai Hospital, Wuhan University Zhongnan Hospital, and China Medical Science Institute and Fuwai Hospital. In the financial field, we continue to deepen cooperation with large state-owned banks, Moving on to enterprise cloud services. Total revenues were RMB 617 million,
increasing by 10.2% quarter over quarter. In public service space, we actively seized opportunities of public services cloud and SASEX cloud. We implemented standardized operation and maintenance with core components such as large language model, big data, and workspace collaboration, targeting use cases in public service and enterprise application domains. With Beijing, Wuhan, and Zhuhai as our basis we have built benchmark projects radiating to other large and medium-sized cities. This quarter, we successfully joined hands with Shenzhen State-Owned Assets Cloud to facilitate the digitalization and intelligence evolution of local state-owned enterprises. In healthcare space, we continued to promote the five business models and make new breakthroughs. Inheriting and developing experience from the cooperation with Shanghai Rating Hospital, we replicated and scaled our medical digital capabilities to more renowned hospitals, including Wuhan Union Hospital, Zhongnan Hospital affiliated to Wuhan University, and the Fuwai Hospital affiliated to Chinese Academy of Medical Sciences. In financial services space, we continued to deepen our business cooperation with large state-owned banks. improving our big data products and service capabilities, and win the bids of big data platform projects for leading state-owned banks and trusts. Turning to Camelot, during the quarter, Camelot exhibited stable and healthy revenue and profitability, signing up four new customers while maintaining robust relationships with existing major clients.
In terms of product technology, we continue to maintain our technical expertise Focusing on the first-class customer experience of building core products, in the computing field, we continue to upgrade core products to improve the stability and sustainability of the domestic environment. This quarter, the release of the flexible metal EPCX-7, its computing, memory, network, and storage performance have been significantly improved. In the storage field, in order to better satisfy high-performance data processing scenarios, For example, storage separation, AI, animation rendering, and so on. We have launched the object storage extreme model, which has ultra-high performance, dynamic expansion, and product features such as unboxing and use. In the field of big data, Yunyuan Sheng Hu Cang One Platform will release a new version. The maximum increase in the number of tasks to be processed is up to four times the consumption rate, and it can better support automatic driving. AI, data model and other scenes. In the field of industry cloud, in the research report released by SIDI by Galaxy Cloud Platform, Siyouyun Market and Siyouyun System Platform market leaders were selected. This is a great recognition of Galaxy Cloud Platform's products and service capabilities. In this quarter, Galaxy Cloud will update the updated version to add six cloud products and more than 100 functions to improve the all-round industry cloud system of building cloud, uploading cloud, using cloud, and managing cloud. In terms of AI, Jinshan Cloud's MaaS mutual trust reasoning specialization program won the most innovative AI solution program award of the China Internet Economic Forum in 2023. Our AI product center
In terms of product and technology, we uphold our principle of building success based on technology and innovation, delivering best-in-class customer experience across our core product offerings. In computing space, we continue to upgrade our core products, in focusing on improving stability and domestic environment compatibility. This quarter, we launched our elastic bare metal compute product, EPCX7, with significantly improved computing, memory, network, and storage performance. In storage space, to better serve high performance use cases, such as decoupling of computing and storage, AI, computer graphic rendering, we have released the extreme version of object storage service, equipped with high-performance dynamic scaling and out-of-the-box features. In big data space, we upgraded our cloud-native Lakehouse platform with maximum throughput improvement by four times for batch processing tasks, better supporting use cases of autonomous driving, AI, and data models. In enterprise cloud space, our Galaxy Stack platform ranked in the market leader quadrant of both the private cloud market and the private cloud system platform market in the research report released by CCID Consulting, a great recognition of the products and services capabilities. This quarter, Galaxy Stack platform released another updated version, adding six cloud products and more than 100 functions. perfecting the comprehensive enterprise cloud system which powers our end-to-end service of cloud building, cloud migration, cloud usage, and cloud management. In terms of AI, our mass mutual trust dedicated zone solution was awarded the most innovative AI solution by China Internet Economy Forum 2023. Our AI product center deepened research cooperation with Kingsoft Office to strengthen model research capabilities and serve enterprise use cases.
In terms of talent development, the development speed of Wuhan R&D Center is fast. Since the end of 2023, the total number of people in Wuhan R&D Center is more than 500. It accounts for about one-third of the total number of R&D teams in the company. The number of specialists accounts for 50%. Our strategy of double R&D Center in Beijing and Wuhan The long-term development of the technical business of the company is to ensure the strength of the development and the effective management of the development cost. In general, our high-quality sustainable development strategy in 2023 will carry out all the effects. The level of profitability continues to increase. We look forward to the future. We insist on doing business that creates sustainable value for customers and society. Embrace AI opportunities. Moving on to talent strategy.
Wuhan Research Center has been quickly expanding. By the end of 2023, total employees in Wuhan exceeded 500, accounting for one-third of total R&D teams, and half of them hold graduate degrees. Our Beijing-Wuhan dual R&D center strategy attracts talent for our long-term development and ensures R&D intensity as we stick to our build success based on technology and innovations principle. while managing R&D expenses effectively. In summary, we have achieved great milestones in executing high-quality and sustainable development strategies, with our profitability measures improving continuously. Looking forward, we will keep our original aspiration and long-term strategy to create value to our customers and society through our business services, embrace AI opportunities, and continue to improve our profitability. leading with increasing management efficiency, we will keep imposing strict cost and expense control, enhance talent training, and expansion of Wuhan Research Center, and further improve the company's operating efficiency. I will now pass the call over to our CFO, Henry, to go over our financials for the fourth quarter and fiscal year 2023.
Thank you. Thank you. And I welcome everyone for joining the call. Now I will walk you through the financial results for the fourth quarter and the physical year 2023. Upholding the strategy of a high quality and a sustainable development, we are pleased to close the year 2023 with meaningful milestones on enhancement of revenue quality, margin expansion, and operating efficiency. For the fourth quarter, we deliver another quarter of steady profitability improvement. Our adjusted growth profit continued to grow for the sixth consecutive quarter and achieved 262.5 million RMB, increased by 55.8% year-over-year, representing adjusted growth margin of 15.2%, which is a record high for the company and significantly improved 3.1 percentage points compared with last quarter. Our normalized adjusted EBITDA narrowed from negative 216.3 million RMB in the same period of last year and a negative 44.1 million RMB in the last quarter to negative 27.7 million RMB this quarter. As a result, normalized adjusted EBITDA margin further narrowed from negative 10.2 in the same period of last year and a negative 2.7% in the last quarter to a negative 1.6% this quarter. Our total revenue were 1,722.5 million RMB this quarter, increased by 6% sequentially, of which revenues from public cloud services were 1,052 million RMB, representing an increase of 3.5% compared with 1,016.6 million RMB in the last quarter. The increase was primarily due to the expansion from AI-related revenues and partially offset by our strategic scaling down of our CDM business by approximately 10% quarter over quarter, contributing around 23% of total revenues. Revenues from enterprise cloud services were 670.3 million RMB, representing an increase of 10.2% from 608.5 million RMB in the last quarter, as more projects are scheduled for delivery towards the end of the year. We continue to enhance our cost control measures. Especially in Q4, we refined the procurement process of CDM business. Total cost of revenues decreased by 25.4% year-over-year to 1,469.3 million RMB. IDC cost decreased significantly by 30% year-over-year from 1,057.6 million RMB to 740.4 million RMB this quarter. The decrease was in line with our adjustments with CDN services. Depreciation and amortization cost decreased by 39.2% from 241.7 million RMB to 146.9 million RMB. The decrease was mainly due to previous impairments of our long-lived assets. Solution development and services cost increased by 8% from 465.8 million RMB to 502.9 million RMB this quarter. This increase was mainly due to business expansion of Camelot. Fulfillment costs and other costs were 9.4 million RMB and 69.7 million RMB this quarter, which are in line with our enterprise cloud project quality control strategy. Adjusted growth profits of this quarter increased by 55.8% year-over-year to 262.5 million RMB, representing adjusted growth margin of 15.2% this quarter, compared with 7.9% in the same period of last year and 12.1% last quarter, making another record high, as well as the sixth consecutive quarter of steady margin improvement. In terms of expenses, Excluding share-based compensation and impairment of long-lived assets, our total adjusted operating expenses were 494.8 million RMB, decreased by 32.2% year-over-year, and 1.9% from last quarter, of which our adjusted R&D expenses were 162.5 million RMB, decreased by 13.2% from last quarter. as we continue to focus on utilizing our Beijing-Wuhan dual research center and welcome new graduate campus recruiting employees. Adjusted selling and marketing expenses were 106.7 million RMB, representing a decrease of 6.5% from 114.1 million RMB last quarter. Adjusted G&A expenses increased by 11.1%, from 203.1 million RMB last quarter to 225.6 million RMB. The increase was mainly due to the year-end payment to vendors. As of December 31, 2023, our cash and cash equivalents and the long-term investments amounted to 2.3 billion RMB, providing us sufficient liquidity for operations. We have entered into a loan facility agreement with Kingsong Corporation with a cap of 1.5 billion RMB. It will dedicatedly support our AI business development and it will demonstrate the confidence and commitment of our ecosystem in bringing the future of AI. The capital expenditure for this quarter was 1,415.8 million RMB. as we invested in our infrastructure to build a sustainable AI business. Our operating cash flow once again recorded a net inflow reaching 16.8 million RMB. Since the second quarter this year, we have been generating net inflow for three consecutive quarters. It resulted from our margin improvements as well as our enhanced internal cash management. For the fall year 2023, our total revenue were 7,047.5 million RMB. Non-GAAP growth profit increased to 859.9 million RMB in 2023, almost doubled from 445.2 million RMB in 2022. Non-GAAP growth margin increased to 12.2% in 2023, from 5.4% in 2022. Such increases was primarily because of the optimization of revenue mix and our effective cost controls, testifying the success of our high quality and sustainable development strategy. Non-GAAP EBITDA was negative 242.1 million RMB compared with negative 726.2 million RMB in 2022. Non-GAAP EBITDA margin was negative 3.4%, compared with negative 8.9% in 2022. Looking ahead, we believe the positive trend in profitability will persist as we continue to pursue a high-quality and sustainable development strategy and unlock synergies within the Xiaomi and Kingsong Group ecosystems, as well as integrate our AI technology and ecosystem in the new era. Thank you.
This concludes our prepared remarks. Thank you for your attention, and we are now happy to take your questions. Please ask your questions in both Mandarin and English if possible. Operator, please go ahead. Thank you.
Thank you. As a reminder, to ask a question, please press star 1 and 1 on your telephone and wait for your name to be announced. To withdraw your question, please press star 1 and 1 again. Once again, please press star 1 and 1 on your telephone and wait for your name to be announced. Thank you. We are now going to proceed with our first question. And the questions come from the line of Sheldon Zhang from CICC. Please ask your question. Your line is opened.
Hey, Mr. Zhou, Mr. Henry, good evening. Thank you for accepting my question. I have two questions I would like to ask. The first question is about AI. We see that the company has been very active in setting up the direction of AI in the past year. I would like to ask Mr. Guan to share about the market competitiveness of our related products, including the solution. What is the expectation of the investment and income of this year's company in the AI field? Then my second question is about our interest rate. In fact, the increase in the interest rate of the company in the past year was very significant. In fact, part of it is also due to the optimization of our income structure, which is the continuous decline of the CDM ratio. So thanks management for taking my questions. And my first question is regarding the AI strategy. So the company has been actively investing in AI for the past year. So could management give us some color on the market competency of your AI-related products and solutions, as well as on this year's CapEx plan and revenue expectations? And my second question is on gross margin. So KC has achieved significant gross margin improvement last year, partially attributable to the change in your revenue mix. So as your CDM business adjustment is approaching the end, What do you think of the gross margin improvement pace going onward? Thank you.
Let me answer the first question and then Harry will answer the second one. As you said, we did a comprehensive layout for the entire AI business last year, so it was good in terms of the results. So the key question you just asked is, how is the market competitiveness of our AI products? I can't say that I'm better than others. I think it can only be said Uh, because of our own special This one, this neutral position, right? Made us get the market's trust Yes, we haven't talked about this in the past few queues, right? This is because of the competition of this big model Ah, this one, we don't do big models ourselves As an independent cloud provider Make me in this round of competition We still have a relatively obvious advantage In fact, from the result This, this, this, this, this, this, this, this, this, this, this, this, this, this, this, this, this Thank you, Jordan.
I think in terms of the margin expansion, I think there are three fundamental drivers for the margins. I think you point out correctly, the first one obviously is the continuous mixed change of the CDM business versus others. So as you see that the CDM business itself actually going through a cycle, but right now we are in a kind of more stable period that we are going to see a potential stability the CDA revenue contribution as well as the client usage patterns that will contribute naturally to a margin expansion. I think that's contributed one-third of the margin improvement. The second thing I also want to mention, probably you didn't notice, I want to point out is, so our measures and the new management initiatives, for example, including renegotiation, the supply chain contracts with our suppliers to cutting down the cost basis, That actually is not happening on one day or a single month of the year, as you can expect. It actually spent over, for example, the full year from last year. And those contracts and the repricing on the cost basis will be effective on a rolling basis from January to December of last year. And you're actually traveling those trends towards 2024. good benefits and the positive impact from the renegotiation and the lower the cost basis will actually carry the continuous benefits to our margin and cost into 2024. I think that part we are contributing the second one-third of the margin expansion. The third is, also you probably didn't notice, is really about, we actually, as Tozo mentioned in the prepared remarks, is getting the better quality of the projects. not only about the cloud project itself, but also on the enterprise side, as well as those verticals, including the financial service, healthcare, and certain high-quality public cloud projects as well. And those projects, as we sign into the contract, some of them are still in the backlog, and that will be unleashed and delivered in 2024. And in history, especially in 2023, we're actually seeing the good experience a good trend that those enterprise cloud projects, the single level contract revenue contribution has expanded from let's say single digits in history a few years ago to double digits and right now we are seeing those trends actually continue to be improved. So the enterprise cloud margin expansion and high quality initiatives and the strategy since 2022 has already seen some good results. So this part actually contributing the last one third of the margin expansion. So to conclude, I think we do see a positive trend, even though I understand Sheldon, your concern may be in Q4, you do see a very good Q on Q jump of the margin about three percentage points. But I think we are in the very good momentum and in a confident way that in Q1 and going forward, Given the three reasons I just mentioned, our growth margin will continue to be seeing a good result going forward.
So allow me to also translate for the first question. So the first question is about the competitiveness of our AI products. So as you correctly commented, indeed in the last year we have comprehensively worked very hard on expanding our AI business. However, it would be harder and less prudent for me to directly comment on the competitiveness of our AI products versus other of our peers. However, I would say that because of our unique positioning of our neutrality and independence, we have, if you look at it from the results, we have becoming a preferred choice of cloud service provider by a large amount of independent AI, large language model AI companies. And that is exactly because we don't do our large language model and build our large language model ourselves. And therefore, in this round of opportunities, we have seen as a result, we are already covering a significant number of independent AI companies becoming our customers. And you also asked about the future investment in terms of CapEx. I would say that the rule of thumb is we will be continue to investing based on the demands and the pipeline of the customers we have. However, the specifics as regards to the specific amount and specific dollar amount and the tempo of such investment it's relatively difficult for me to comment at this stage. Yeah.
Operator, please go ahead.
Thank you. We are now going to proceed with our next question. And the questions come from the line of Daily Life from Bank of America Securities. Please ask your question. Your line is opened.
Good evening, everyone. Thank you for accepting my question. I would like to ask two questions. The first one is still about AI. We mentioned before that demand is much greater than the ability to supply. What is the current demand and supply capacity like? If the supply is still short, how will the corresponding chip solve domestic and overseas purchases? What are the main growth points of the AI rental business in the future? For example, new customers or customers who are satisfied with the demand. The second question is about our profit rate. For example, in 2024, our EBITDA margin, adjusted EBITDA margin, I have two questions. The first one is about the AI business. Could you share some color about the supply and demand trend recently and how do we see the growth drivers for the AI data center AI and public cloud business going forward and the key drivers. And secondly, my second question about the margin trend. Looking into 2014, how do we see the margins such as the adjusted EBITDA margin and some costs like a share-based competition cost trend? Thank you.
Okay. Let me answer the first question first. Liu, you can go ahead and see if you have anything to add. Indeed, as I said before, the demand of AI for chips is greater than the supply. This is the main force of last year's whole year. So we have been actively satisfying our customers as much as possible. Well, in fact, to this day, everyone knows that some changes have occurred in the entire external environment last year, so it leads to this kind of supply and demand relationship. In fact, there is no fundamental change, or even some degree of deterioration. Of course, on the other hand, the rapid rise of the national carbon footprint itself. Well, I also talked about this point last year. In fact, we started to pay attention from the beginning of the previous year. In fact, to some extent, it has also played a certain relieving role, but in general terms, this market is still Uh, especially with the upgrade of this model parameter, the third one is getting bigger and bigger. This node node is getting bigger and bigger, so from our point of view, this is still a little bit of the demand in the entire market is greater than the attack. This is not completely solved, and our judgment is that this may be a long-term existence. As long as this continue to continue to develop this trend does not change the situation this I my own judgment also will not be too short-term will not be too fundamental change good so that in fact, in order to deal with this situation ah in fact, we also have several different strategies on the one hand, of course, this is what we have always been also predicted this form ah so we can with some of the domestic algorithm manufacturers also open a positive cooperation to build a national algorithm of this area In addition, we are also actively working with INVIDA on SP construction overseas. This is the current situation. The conclusion is that the development needs of AI are still abundant, but the supply and demand will still be under great pressure. So we will have various channels to find ways to satisfy our customers as much as possible. Then you talk about the factors of AI growth, or I will talk about it first, and then Liu Tao can add it. What I see is that the old user, with the improvement of the parameters of the training model itself, you can understand that this is the decision. It is the improvement of the requirements of the algorithm, so you can understand that some of the requirements of the old customer itself are expanded. Another part of it is actually at least from our side There are some new customers one after another These new customers are not for the sake of doing the big model But in fact, it can be understood that the ability of the big model They have business to do, so they also present this more and more abundant need In fact, some of the customers we have been dealing with in the past or are talking about do not even include some car customers Uh, it's true that this is actually happening, right? This is, uh, this is simply to say, in fact, it is the need for old customers to expand and some new ones. This is this, uh, ready to give the model to the new customers of the business. There are all of them, but at present, uh, the proportion of old customers is getting higher and higher. Yes, but my personal opinion is that this is with this. AI is getting more and more in the hearts of people. It's getting more and more recognized and accepted by everyone. Maybe the future growth of new customers is still very, very big. But I personally believe in this point. Okay, let's see if Liu Tao has anything to add. I would like to add a little bit of information about the growth points.
One is what Chairman Zhou just said. The old customer expansion is a very important point. The other is that we saw in the last quarter that the model application of Internet customers in the Internet company's demand for growth. So they will have this Internet company based on the model to do training to bring this training increase And they only have the model to use in their own IDC to increase the push power Another one is that it may include things like vision and video This includes the appearance of harassment and the increase in the production of this picture and video and Autonomous driving companies and car companies need to calculate the growth of the new technology used for autonomous driving This is the main growth point to see the translation of this first
So, yes, you mentioned that the situation of demand for GPU chips more than supply, which has been the key theme for last year. And that is why during the past year, we have also been trying our best to meet the clients, our customers' demand. Well, the situation this year, there's some changes, right? On one hand, the production of some made-in-China chips have been alleviating to some extent of the issue. However, we have also seen other factors, especially geographic tension factors, deteriorating the situation. So overall speaking, we see that since the AI market is still booming and demand continues The overall theme of the demand and supply balance is still that the demand is largely exceeding that of the supply. And we do not expect that kind of relationship to have any material change in the near term future. Now, in terms of our responses strategy to this situation, on one hand is that we're working with some of the firms that have computing powers within China to building a dedicated computing power zone together. And the second thing that we do is to follow up with new products that are within compliance premises being allowed to supply in China market. And in terms of growth drivers, there are basically two types of growth drivers. One is old customers, existing or let's say old customers, which are mainly characterized by the independent AI large language model customers that we have been serving recently. And the second type is new customers, which are typically not the independent AI large language model companies. However, those companies that are ready to leverage the capability of large language models to empower their existing business For example, EV customers. Currently, we see that the growth driver coming from existing customers to taking a higher proportion. However, we do expect that in the future, the potential for new customers' demand has higher growth potential. And Mr. Liu Tao, our SVP, also added that there are several types of growth drivers. One is the Internet companies training their own models and using their in-house models and by using the inference capability of the computer power to use their in-house models. And secondly is the advance and the launch of Sora that leading to a wave of demand coming from the video side of things. And the third part would be the new companies like Mr. Soto mentioned of EV autonomous driving kind of demand. Thank you.
Yeah, thank you. So I will take on the second question. I appreciate your notice on the expansion on the margin side. So I think there are a few kind of major directions we're trying to achieve going forward at the same time. First of all, we are aiming to keep a Q-on-Q expansion on the growth margin side. So this actually is driven by the cost-cutting on the supply chains and the better automation on the resources we have and cutting certain loss-making computing regions and certain disposal for certain equipment that we don't think are fitting to today's client requirements. So all the combination of actions were contributing to the margin expansion of the gross margin. And as I explained to the first question, those efforts are already in place, and we're going to see those impacts will be gradually delivered going forward in the next few quarters as well. So the first, as I mentioned, is really about expansion on a queue-on-queue basis on the gross margin. So the second part is really about narrow and better managing the expenses and operation expenses, including, for example, the better management of the human capital costs including the internal efficiency and the streaming line, all the internal management initiatives that were actually well contributing for the expansion of the EBITDA margin. So as you can see, our expenses between the line of the gross margin and the EBITDA margin continue to optimize and reduce as well in the past few quarters. So that has proven the management team has already got some good experience and practiced those skills pretty well. And we think we can carry on that in the next few quarters. So our second aim is to improve the EBITDA margin on a continuing basis. So we are confident to see as we make our efforts, our EBITDA margin will be getting to approaching the breakeven in a short term of time. And the third is we're also adding to another KPI for ourselves in terms of internal management. We're trying to not only making the EBITDA margin break even, but we're also trying to make the company with the better quality of revenue and a continuous optimized revenue mix and picking the right clients and also expanding the AI exposures. We'll be in a very good position to have the possibility to see a good trend of OP margin expansion as well. So I think these are the three goals we are trying to achieve in 2024. While we are not in a position today to give a clear guidance about the timing of the EBITDA margin breakeven as well as the profitability on OP margin side, we are confident to say that all the initiatives in place will see that in good results going forward in the next few quarters probably we can observe. And to a question with the SBC cost, one thing I want to note is that based on the accounting rules, the SBC cost was booked based on the share price at the time when the ESOP and the shares was granted to the employees, not on a vesting period of time. So when the share price was high, those costs will be booked and amortized in the next few years. So I don't think given the volatility of the capital markets for many of the shares today, the SBC line is reflecting the true value, i.e., the market value of the ESOP value we granted to the employees. So those values need to be revisited and recalculated if you want to do a model. But putting all things together, the company has adopted a very prudent way of granting and investing the shares, as you may see from the announcements. And also, we're extending third investing schedules to keep the company employee working a bit longer and with a better incentive. So as you can see, from 2022 to 2023, our SBC cost has declined from roughly 360 million RMB in 2022 to drop about half to 180 million RMB in 2023. So I think we are with a prudent attitude from issuing, granting, and investing the ESOP. And I think going forward, you're going to see a better and a narrow line between the GAAP and non-GAAP margin as well.
Thanks. Operator, will you mind the last question, please?
Thank you. We are now going to proceed with our last question. And the question comes from the line of view. It's Liu from Goldman Sachs. Please ask your question.
Thank you, Mr. Feng, Mr. Henry, and Mr. Clark for accepting my question. I'm Eunice Gao, a representative researcher at Unis. First of all, congratulations on the company's strong performance. I have two questions. The first is about the industry. You mentioned that in 2024, I will quickly translate myself. Could management talk about the strategic planning for enterprise cloud in 2024? And the second question is, we have seen some news on peer price competition in public cloud in early 2024. And could management tell us more about the latest competitive landscape and also looking for cloud's comparative advantages? Thank you.
Okay. Uh, I think there are a few new opportunities. Of course, the first one is not a new one. In fact, in the past two years, I have established a real estate company in various places. And then this is the end of the season. In fact, there is an opportunity to bring this country to the cloud. In fact, since last year, we have been going back and forth. I just introduced some of them at the phone conference. I think this is an opportunity. But I think by this year, I think there is even a greater opportunity, even this opportunity. We have recently had a few discussions with this part of the Han Yun's department. Anyway, the smaller it is, the more excited it is. In fact, what is it? Everyone knows that in 2024, this data road map, this is called the original year. The data is the real estate road map of the original year. Last year, the Ministry of Finance released this draft of the data road map. Yes, I want this thing. Uh, this one, this one also includes us recently with the plastic lock of Shanghai, the plastic lock of Shanghai, this one has done some deep communication around this one, including the recent one, including some other audit industries in China, and also do some communication. Anyway, how to say that this may be a very big opportunity, but this may be said to have something to do with the cloud. Our, our point of view is that this tree must run on the cloud. This must be a cloud tree. It is impossible to be locked in a hard disk. So in the future, of course, he has to make it clear. What does this mean? Simply put, this number has become money. If the number is money, he will let his production flow, trade, etc. This is all opportunities. And these are what we think must be running in the clouds. Yes, so this is what we think is a new opportunity for this industry. Of course, we have been doing some planning and layout around this. In the near future, we will also hold some closed-door meetings around this and do some consultations. This is a relatively big opportunity that we have seen in the near future. Of course, this number table does not mean much to enterprises listed abroad. It is mainly aimed at domestic listed enterprises, because this regulation should be China's current speciality. But you can also say that you understand that you are looking for a relatively cheap one. This is what I think is the one that I saw this year in Hanyun. In fact, it has nothing to do with any specific industry. It's just that it's just to highlight this tree and become valuable and become money. Yes, it's good. So this is a trend of sharing these two industries. So I think this is a very good question. And we have been conducting internal discussions
around exactly the question you raised. So a few opportunities I would like to share with you. The first one is you might have noted that recently a large number of digital asset companies across China have been established, and that basically led to a wave of state-owned asset companies migrating onto cloud and using more cloud. And the second, which I think is probably a large opportunity that is happening this year, is the the measure of digital assets showing on the balance sheet. And we should call this year of 2024 to be the first year of this opportunity because of the government policy recently promulgated. We have also been making communications with the Shanghai Digital Asset Exchange and also with professional parties, for example, like auditing firms. We do think that it represents a significant opportunity for cloud service companies and for us because all of these digital assets will be running on the cloud infrastructure and digital assets and those data assets cannot run by itself. So in our words, it's the integration of cloud service and digital assets. So all this whole chain of data data production of data transaction, et cetera, et cetera, are going to be supported by the cloud. And therefore all of this represents opportunities for us. So it's not opportunity with respect to any particular industry, but a general, but a probably larger opportunity that we can expect to see in recent years.
Okay. Well, what you're talking about is the price drop. To be honest, this is my point of view. The core point is two points. I'm the first one. This is the price drop year after year. It's not this year either. This is from the beginning of our cloud. This is not to say that in the past two years, this price drop, all kinds of price drops, this year, this price drop, tomorrow, that price drop. In fact, there was a small drop last year. This is the same as the previous year. Anyway, my point is that the last one is actually not a big change for the entire industry development. This is from our own development perspective, right? We also have six consecutive seasons. We continue to increase the profitability of our business. Business is also getting more and more stable. So this is the result. This is a real talk. So we don't think this will have a real impact on the industry. This is a second one, which is the reason for the analysis. My own point of view, especially last year, we were looking for this one for this quality improvement. After this customer bought one, the customer should say that our customer is paying attention to the first point, but the price must be sensitive, but he still pays more attention to the entire product service. You say you are at this price. If the service is not good, the product is very bad. It doesn't make sense either So so that's what I want to express That's the key to the price But not to say This is the first element At least I'm from my client This is for these years Anyway, this is the first point I want to make The price goes up year after year But in fact it didn't Because of this to us The impact of the nature of the industry From the perspective of this question All right So it's actually Of course, I also understand I also learned from my colleagues In fact This price also has some of the ingredients for the price. In fact, it is based on its list price. In fact, the real transaction, many of its customers are lower than this discount. So it may be for some new customers. So I said at the end that the price is more of this kind of salesman. Anyway, I mentioned this point last year. It's a kind of salesman. In fact, this is for existing customers. Anyway, that's what I was thinking last year. Anyway, there's not much impact. All right.
Let's see if Liu Tao has anything to add. I just want to add one thing, which is the price reduction strategy you mentioned. Because there's a lot of news about it. The way the other party lowered the price is actually based on annual payback. On the traditional discount basis, they gave an additional discount. So he's targeting small-scale annual payback customers. This customer group is actually with the customer group of Jinshan Yun. There is no overlap.
So the second question relates to the price competition in the market. So I would like to share with you two of my core views. The first one is we have actually been experiencing price pressure since the day of our founding. Every year we have been seeing different kinds of price pressure coming from various of our peers in different kind of products. However, at the end of the day, we do not really see any of those price cuts by our peers have any material or significant impact to the market. And from our own performance results, we have also achieved for six consecutive quarters of profitability improvement Now, I would like to say that we think the reason for that is pricing obviously is one important factor for customers to consider. However, it is far from the most important factor. And from our own experience, and also as evidenced by our achievements in the past six quarters, focusing on the satisfaction of customers is actually the most important thing. And I would also like to say that my view about this kind of large amount of price cuts by peers is more towards the end of marketing and PR measures. In fact, some of these price cuts are based on catalog price. However, a lot of the transaction price that cloud service providers have agreed with potential customers are already deeply discounted and already way below the price cut level that we are seeing today. And also, our SVP Liu Tao also added that this round of price cuts from up here mainly focuses on the customers, which are relatively small in scale and are paying their file usage fees on an annual basis. And this group of customers actually do not overlap with the customers that PinkSoft Cloud enjoys. And therefore, it does not have any material impact to our pricing strategy or our business performance. Thank you.
Thank you, Professor.
We have no further questions at this time. I would like to hand back to Nichol Shan for closing remarks.
Thank you once again for joining us today. If you have any further questions, please feel free to contact us. Look forward to speaking with you again next quarter. Have a nice day. Thank you.
Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect your lines. Thank you and have a good day. Thank you. you Thank you. music music Thank you. Good day and thank you for standing by. Welcome to the Kings of Clouds fourth quarter and fiscal year 2023 Earnings Conference Caller webcast. At this time, all participants are in listen-only mode. After the speaker's presentation, there will be a question and answer session. To ask a question during the session, please press star 1 and 1 on your telephone. You will then hear an automated message advising your hand is raised. To withdraw your question, please press star 1 and 1 again. Please note that today's conference is being recorded. I would now like to turn the conference over to Nicole Shan, IR Manager of Kings of Cloud. Please go ahead.
Thank you, operator. Hello, everyone, and thank you for joining us today. Kings of Cloud's first quarter and fiscal year 2023 earnings release was distributed earlier today and is available on our IR website at ir.ksyun.com, as well as our global newsware services. On the call today from Kingsoft Cloud, we have our Vice Chairman and CEO, Mr. Zhou Tao, and CFO, Mr. Hongrui He. Mr. Zhou will review our business strategies, operations, and company highlights, followed by Mr. He, who will discuss the financials and the guidance. They will be available to answer your question during the Q&A session that follows. There will be consecutive interpretations. Our interpretations are for your convenience and reference purpose only. In case of any discrepancies, management statements in the original language will prevail. Before we begin, I'd like to remind you that this conference call contains forward-looking statements within the meaning of Section 21E of the Securities Exchange Act of 1934, as demanded and as defined in the U.S. Private Security Discretion Reform Act of 1995. These forward-looking statements are based upon management's current expectations and current market and operating conditions. and related to U.S. study involve known or unknown risks, uncertainties, and other factors, of which are difficult to predict and many of which are beyond the company's control, which may cause the company's actual results, performance, or achievements to differ maturely from those in the forward-looking statements. Further information regarding these and other risks, uncertainties, or factors are included in the company's filings with the U.S. SEC. The company does not entertain any obligation to update any forward-looking statements. as a result of new information, future events, or otherwise, except for required under applicable law. Finally, please note that, unlike otherwise stated, all financial figures mentioned during this conference call are denominated in RMB. It's now my pleasure to introduce our Vice Chairman and CEO, Mr. Zhou. Please go ahead, Zhou Zhou.
Hello, everyone. Welcome to the 4th quarter of 2023, and the full-year performance call conference. In 2023, It is a year in which Jin Tianyun's full-scale observation of high-capacity sustainable development has achieved excellent results. In the whole year of 2023, after adjustment, the interest rate reached 12.2%, while the 5.4% in 2022 significantly increased by 6.8%. After adjustment, the interest rate reached RMB 8.6 billion, which is almost half of the 4.5 billion in 2022. After normalization adjustment, the return rate is negative 3.4%, which is negative 8.9% in 2022. Over the past year, we have insisted on technical business, mouthpiece as king, management as steel, practice internal work, strategic adjustment business structure, and actively embrace the new era of AI. For the future of AI in 2024,
Hello, everyone, and thank you all for joining King's Hub Cloud's fourth quarter and fiscal year 2023 earnings call. In 2023, we continued to uphold the principle of high quality and sustainable development and accomplished significant achievements. For the full year of 2023, adjusted growth margin was 12.2%, a significant increase of 6.8 percentage points. up from 5.4% in 2022. Adjusted gross profit was RMB 860 million, almost doubling the amount of RMB 445 million in 2022. Normalized adjusted EBITDA was negative 3.4%, another significant improvement from negative 8.9% in 2022. During the year, we started building our success based on technology and innovations forging our reputation throughout the entire business process, enhancing our operations management, and building inner strength. We have been strategically adjusting our business mix and proactively embracing the new AI era, therefore laying solid foundations to long-term sustainable development in 2024 and beyond.
Next, I would like to focus on the performance of the fourth quarter of 2023. In this quarter, Xinshan Yun achieved a growth rate and double growth in profit and profit capacity. The income reached RMB1.7.2 billion, and the interest rate increased by 6%. After adjustment, the interest rate reached 15.2%. Compared to the previous quarter, it increased by 3.1%. It achieved a steady increase of 6 consecutive quarters, creating a new high in history. Now I will walk you through the business highlights of the fourth quarter 2023. This quarter, we achieved dual improvement in both our revenue and profitability measures.
In particular, total revenues reached RMB 1.72 billion, increasing 66.0% quarter over quarter. Adjusted gross margin recorded a major increase of 3.1 percentage points quarter over quarter to 15.2%, marking the sixth quarter of consecutive improvement. Adjusted gross profits reached RMB 262 million, increasing 55.8% year over year. Normalized adjusted EBITDA margin was negative 1.6%, representing a significant improvement of 8.5 percentage points year over year.
In terms of public cloud, this quarter, we achieved a revenue of 10.5 billion yuan, which increased by 3.5%. In addition to CDN business, public cloud revenue increased by 11.8%. In Xiaomi Jinshan Ecology AI business, CDN strategic adjustment in three aspects. First of all, as the only strategic cloud platform of Xiaomi Jinshan Ecotourism, to serve the needs of ecotourism companies, especially in the context of Xiaomi's car WS AI, the need for visible and comprehensive training and reasoning, to grasp the opportunity in the wave of industrial structural development, The revenue contribution from Xiaomi and Jinshan reached 16%, which increased by 5%. Secondly, we continue to develop our AI business and actively respond to the changes in the industry. In addition to the opportunity for AI business in Xiaomi Jinshan, we fully cover a group of independent AI companies that are the most leading in China. to provide them with high-performance long-term security for market shortcomings. In this quarter, AISolo accounts for 8% of the total net profit and 82% of the net profit growth. In addition, the company signed a loan agreement with Jinshan Software to obtain a loan amount of 1.5 billion yuan, which is dedicated to the development of AIO. Third, In terms of public cloud services, revenues for RMB 1.05 billion
representing an increase of 3.5% quarter over quarter. Excluding CDN business, public cloud revenue increased by 11.8% quarter over quarter. We bear fruits on our three priorities for public cloud services, namely the Xiaomi and Kingsoft ecosystem, AI business, and CDN strategic adjustments. First of all, as the sole strategic cloud platform within the Xiaomi and Kingsoft ecosystem, we continued to serve the demands in the ecosystem well, especially with respect to the highly visible and tangible demands in training and inference from Xiaomi EV and WPS AI, seizing opportunities in structural industry trends. This quarter, revenues contributed by Xiaomi and Kingsoft ecosystem reached 16%, representing an increase of 5 percentage points year over year. Secondly, we vigorously developed our AI business with remarkable agility to emerging industry trends. Besides the AI opportunities from the Xiaomi and Kingsoft ecosystem, we have been fully penetrating leading independent AI companies in China, providing long-term, secured, high-performance computing power, which is highly sought after in the market. Our AI business represented approximately 8% to our public cloud revenues, or an increase of 82% quarter over quarter. Besides, we have signed a loan facility agreement with Kingsoft Corporation, obtaining a credit line of RMB 1.5 billion, dedicating to the development of AI business. Thirdly, we continued to push forward our strategic adjustments of CDN business. This quarter, CDN revenue decreased by nearly 10% compared to last quarter, and CDN revenue as a proportion of total revenue has decreased to approximately 23%. The revenue share of our largest CBN customer was 12% in this quarter. Top customer concentration has been fundamentally alleviated.
In terms of sales, this quarter we achieved a revenue of 6.7 billion yuan, a 10.2% increase in return. In the public service sector, we are actively grasping the project opportunities of Zhengwu Yun and Guo Zhiyun. In terms of the two types of business operations scenarios, with large model, large data, and co-operative office as the base to carry out standardized operation. We use Beijing, Wuhan, and Zhuhai as the base to build the benchmark project and replicate other major cities. This quarter, we successfully joined hands with Shenzhen State University to assist local state-owned enterprises in digitalization and development. In the field of digital health, we will continue to advance the five major models. to achieve new breakthroughs. We inherited and developed cooperation experience with Shanghai Ruijin Hospital, to duplicate the medical digitalization capability in more famous hospitals, including Wuhan Xihe Hospital, Wuhan University Zhongnan Hospital, and China Medical Science Institute, Fuwai Hospital, etc. In the financial field, we continue to deepen cooperation with large state-owned banks, Moving on to enterprise cloud services. Total revenues were RMB 617 million,
increasing by 10.2% quarter over quarter. In public service space, we actively seized opportunities of Public Services Cloud and CESAX Cloud. We implemented standardized operation and maintenance with core components such as large language model, big data, and workspace collaboration, targeting use cases in public service and enterprise application domains. With Beijing, Wuhan, and Zhuhai as our basis, we have built benchmark projects radiating to other large and medium-sized cities. This quarter, we successfully joined hands with Shenzhen's state-owned assets cloud to facilitate the digitalization and intelligence evolution of local state-owned enterprises. In healthcare space, we continued to promote the five business models and make new breakthroughs. Inheriting and developing experience from the cooperation with Shanghai Rating Hospital, we replicated and scaled our medical digital capabilities to more renowned hospitals, including Wuhan Union Hospital, Zhongnan Hospital affiliated to Wuhan University, and the Fuwai Hospital affiliated to Chinese Academy of Medical Sciences. In financial services space, we continued to deepen our business cooperation with large state-owned banks. improving our big data products and service capabilities, and win the bids of big data platform projects for leading state-owned banks and trusts. Turning to Camelot, during the quarter, Camelot exhibited stable and healthy revenue and profitability, signing up four new customers while maintaining robust relationships with existing major clients.
In terms of product technology, we continue to maintain our technical expertise Focusing on the first-class customer experience of building core products, in the computing field, we continue to upgrade core products to improve the stability and sustainability of the domestic environment. In this quarter, the release of flexible aluminum EPCX7, its computing, memory, network, and storage performance have been significantly improved. In the storage field, to better satisfy high-performance data processing scenarios, such as pure separation, AI, and dynamic rendering. We have launched the object storage, extreme speed type, with ultra-high performance, dynamic expansion, and open-box use, and other product features. In the field of big data, Yun Yuan Sheng Hu Cang One Platform will release a new version, with a maximum increase of up to four times the volume of task processing, and better support for self-driving. such as AI, data model, etc. In the field of industry cloud, in the research report released by SIDI by Galaxy Cloud Platform, the market leaders of Siyouyun Market and Siyouyun System Platform appeared. This is a great recognition of Galaxy Cloud Platform's products and service capabilities. In this quarter, Galaxy Cloud will update the updated version to add six cloud products and more than 100 functions to improve the all-round industry cloud system of building cloud, uploading cloud, using cloud, and managing cloud. In terms of AI, Jinshan Cloud's MaaS mutual trust reasoning specialization program won the most innovative AI solution program award of the China Internet Economic Forum in 2023. Our AI product center and deep development cooperation with Jinshan Office to strengthen the model development capability used for corporate service and business scenarios.
In terms of product and technology, we uphold our principle of building success based on technology and innovation, delivering best-in-class customer experience across our core product offerings. In computing space, we continue to upgrade our core products, in focusing on improving stability and domestic environment compatibility. This quarter, we launched our elastic bare-metal compute product, EPCX7, with significantly improved computing, memory, network, and storage performance. In storage space, to better serve high-performance use cases, such as decoupling of computing and storage, AI, computer graphic rendering, we have released the extreme version of object storage service equipped with high performance, dynamic scaling, and out of the box features. In big data space, we upgraded our cloud native Lakehouse platform with maximum throughput improvement by four times for batch processing tasks, better supporting use cases of autonomous driving, AI, and data models. In enterprise cloud space, our Galaxy Stack platform ranked in the market leader quadrant of both the private cloud market and the private cloud system platform market. In the research report released by CCID Consulting, a great recognition of the products and services capabilities. This quarter, Galaxy Stack platform released another update version, adding six cloud products and more than 100 functions. perfecting the comprehensive enterprise cloud system which powers our end-to-end service of cloud building, cloud migration, cloud usage, and cloud management. In terms of AI, our mass mutual trust dedicated zone solution was awarded the most innovative AI solution by China Internet Economy Forum 2023. Our AI product center deepened research cooperation with Kingsoft Office to strengthen model research capabilities and serve enterprise use cases. In terms of talent development, the development speed of the Wuhan Research and Development Center is fast.
Since the end of 2023, the total number of people in the Wuhan Research and Development Center has exceeded 500. It accounts for about one-third of the total number of research and development teams in the company, and accounts for about 50% of the total number of people in the company, and accounts for about one-third of the total number of people in the company, and accounts for about 50% of the total number of people in the company, and accounts for about 50% of the total number of people in the company, and accounts for about 50% of the total number of people in the company, and accounts for about 50% of the total number of people in the company, and accounts for about 50% of the total number of people in the company, and accounts for about 50% of the total number of people in the company, and accounts for about 50% of the total number of people in the company, and accounts for about 50% of Moving on to talent strategy.
Wuhan Research Center has been quickly expanding. By the end of 2023, total employees in Wuhan exceeded 500, accounting for one-third of total R&D teams, and half of them hold graduate degrees. Our Beijing-Wuhan dual R&D center strategy attracts talent for our long-term development and ensures R&D intensity as we stick to our build success based on technology and innovations principle. while managing R&D expenses effectively. In summary, we have achieved great milestones in executing high-quality and sustainable development strategies, with our profitability measures improving continuously. Looking forward, we will keep our original aspiration and long-term strategy to create value to our customers and society through our business services, embrace AI opportunities, and continue to improve our profitability. leading with increasing management efficiency, we will keep imposing strict cost and expense control, enhance talent training, and expansion of Wuhan Research Center, and further improve the company's operating efficiency. I will now pass the call over to our CFO, Henry, to go over our financials for the fourth quarter and fiscal year 2023.
Thank you. Thank you. And I welcome everyone for joining the call. Now I will walk you through the financial results for the fourth quarter and the physical year 2023. Upholding the strategy of a high quality and a sustainable development, we are pleased to close the year 2023 with meaningful milestones on enhancement of revenue quality, margin expansion, and operating efficiency. For the fourth quarter, we deliver another quarter of steady profitability improvement. Our adjusted growth profit continued to grow for the sixth consecutive quarter and achieved 262.5 million RMB, increased by 55.8% year-over-year, representing adjusted growth margin of 15.2%, which is a record high for the company and significantly improved 3.1 percentage points compared with last quarter. Our normalized adjusted EBITDA narrowed from negative 216.3 million RMB in the same period of last year and a negative 44.1 million RMB in the last quarter to negative 27.7 million RMB this quarter. As a result, normalized adjusted EBITDA margin further narrowed from negative 10.2 in the same period of last year and a negative 2.7% in the last quarter to a negative 1.6% this quarter. Our total revenue were 1,722.5 million RMB this quarter, increased by 6% sequentially, of which revenues from public cloud services were 1,052 million RMB, representing an increase of 3.5% compared with 1,016.6 million RMB in the last quarter. The increase was primarily due to the expansion from AI-related revenues and partially offset by our strategic scaling down of our CDM business by approximately 10% quarter over quarter, contributing around 23% of total revenues. Revenues from enterprise cloud services were 670.3 milliRMB, representing an increase of 10.2% from 608.5 milliRMB in the last quarter, as more projects are scheduled for delivery towards the end of the year. We continue to enhance our cost control measures. Especially in Q4, we refined the procurement process of CDM business. Total cost of revenues decreased by 25.4% year-over-year to 1,469.3 million RMB. IDC cost decreased significantly by 30% year-over-year from 1,057.6 million RMB to 740.4 million RMB this quarter. The decrease was in line with our adjustments with CDN services. Depreciation and amortization cost decreased by 39.2% from 241.7 million RMB to 146.9 million RMB. The decrease was mainly due to previous impairments of our long-lived assets. Solution development and services cost increased by 8% from 465.8 million RMB to 502.9 million RMB this quarter. This increase was mainly due to business expansion of Camelot. Fulfillment costs and other costs were 9.4 million RMB and 69.7 million RMB this quarter, which are in line with our enterprise cloud project quality control strategy. Adjusted growth profits of this quarter increased by 55.8% year-over-year to 262.5 million RMB, representing adjusted growth margin of 15.2% this quarter, compared with 7.9% in the same period of last year and 12.1% last quarter, making another record high, as well as the sixth consecutive quarter of steady margin improvement. In terms of expenses, Excluding share-based compensation and impairment of long-lived assets, our total adjusted operating expenses were 494.8 million RMB, decreased by 32.2% year-over-year, and 1.9% from last quarter, of which our adjusted R&D expenses were 162.5 million RMB, decreased by 13.2% from last quarter. As we continue to focus on utilizing our Beijing Wuhan dual research center and welcome new graduate campus recruiting employees. Adjusted selling and marketing expenses were 106.7 million RMB representing a decrease of 6.5% from 114.1 million RMB last quarter. Adjusted G&A expenses increased by 11.1% from 203.1 million RMB last quarter to 225.6 million RMB. The increase was mainly due to the year-end payment to vendors. As of December 31, 2023, our cash and cash equivalents and the long-term investments amounted to 2.3 billion RMB, providing us sufficient liquidity for operations. We have entered into a loan facility agreement with Kingsong Corporation with a cap of 1.5 billion RMB. It will dedicatedly support our AI business development and it will demonstrate the confidence and commitment of our ecosystem in creating the future of AI. The capital expenditure for this quarter was 1,415.8 million RMB. as we invested in our infrastructure to build a sustainable AI business. Our operating cash flow once again recorded a net inflow reaching 16.8 million RMB. Since the second quarter this year, we have been generating net inflow for three consecutive quarters. It resulted from our margin improvements as well as our enhanced internal cash management. For the fall year 2023, our total revenue were 7,047.5 million RMB. Non-GAAP growth profit increased to 859.9 million RMB in 2023, almost doubled from 445.2 million RMB in 2022. Non-GAAP growth margin increased to 12.2% in 2023, from 5.4% in 2022. Such increases was primarily because of the optimization of revenue mix and our effective cost controls, testifying the success of our high quality and sustainable development strategy. Non-GAAP EBITDA was negative 242.1 million RMB compared with negative 726.2 million RMB in 2022. Non-GAAP EBITDA margin was negative 3.4%, compared with negative 8.9% in 2022. Looking ahead, we believe the positive trend in profitability will persist as we continue to pursue a high-quality and sustainable development strategy and unlock synergies within the Xiaomi and Kingsong Group ecosystems, as well as integrate our AI technology and ecosystem in the new era. Thank you.
This concludes our prepared remarks. Thank you for all your attention, and we are now happy to take your questions. Please ask your questions in both Mandarin and English if possible. Operator, please go ahead. Thank you.
Thank you. As a reminder, to ask a question, please press star 1 and 1 on your telephone and wait for your name to be announced. To withdraw your question, please press star 1 and 1 again. Once again, please press star 1 and 1 on your telephone and wait for your name to be announced. Thank you. We are now going to proceed with our first question. And the questions come from the line of Sheldon Zhang from CICC. Please ask your question. Your line is opened.
Hey, Mr. Zhou, Mr. Henry, good evening. Thank you for accepting my question. I have two questions I would like to ask. The first question is about AI. We see that in the past, companies have been very active in setting up the direction of AI. I would like to ask the management to share about the current market competitiveness of our related products, including the solution. What is the expectation of the investment and income in the AI field this year? Then my second question is about our interest rate. In fact, the increase in the interest rate of the company in the past year was very significant. In fact, part of it is also due to the optimization of our income structure, which is the continuous decline of the CDM ratio. So thanks management for taking my questions. And my first question is regarding the AI strategy. So the company has been actively investing in AI for the past year. So could management give us some color on the market competency of your AI-related products and solutions, as well as on this year's CapEx plan and revenue expectations? And my second question is on gross margin. So KC has achieved significant gross margin improvement last year, partially attributable to the change in your revenue mix. So as your CDN business adjustment is approaching the end, What do you think of the gross margin improvement pace going onward? Thank you.
Xiaodan, I will answer the first question. The second one is for Henry. Chenglong, you said that we did a comprehensive layout for the entire AI business last year. So from the result, it is also good. So the key question you just asked is how is the market competitiveness of our AI products? I can't say that I am better than others. I think it can only be said Uh, because of this kind of special one of our own This is the neutral position It made us get the market's trust Yes, we haven't talked about this in the past few queues, right? This is because of the competition of this big model Ah, this is that we don't do big models ourselves As an independent cloud shareholder Ah, it makes us in this round of competition We still have a relatively obvious advantage In fact, from the result This, this, this, this, this, this, this, this, this, this, this, this, this, this, this, this, We will also make a matching investment in this rhythm. It is not convenient to disclose the specific rhythm and amount of investment at the moment. Okay, let's see if there is any supplement. The second question is that Harry will answer it.
Thank you, Xiaodan. I think in terms of the margin expansion, I think there are three fundamental drivers for the margins. I think you point out correctly, the first one obviously is the continuous mixed change of the CDM business versus others. So as you see that the CDM business itself actually going through a cycle, but right now we are in a kind of more stable period that we are going to see a potential stability of the CDA revenue contribution as well as the client usage patterns that will contribute naturally to a margin expansion. I think that's contributed one third of the margin improvement. The second thing I also want to mention, probably you didn't notice, I want to point out is, so our measures and the new management initiatives, for example, including renegotiation, the supply chain contracts with our suppliers to cutting down the cost basis, That actually is not happening on one day or a single month of the year, as you can expect. It actually spent over, for example, the full year from last year. And those contracts and the repricing on the cost basis will be effective on a rolling basis from January to December of last year. And you're actually traveling those trends towards 2024. good benefits and the positive impact from the renegotiation and the lower the cost basis will actually carry the continuous benefits to our margin and cost into 2024. I think that part we are contributing the second one-third of the margin expansion. The third is also you probably didn't notice is really about we actually have mentioned in the prepared remarks is getting the better quality of the projects. not only about the cloud project itself, but also on the enterprise side, as well as those verticals, including the financial service, healthcare, and certain high-quality public cloud projects as well. And those projects, as we signed the contract, some of them are still in the backlog, and that will be unleashed and delivered in 2024. And in history, especially in 2023, we're actually seeing the good experience a good trend that those enterprise cloud projects, the single level contract revenue contribution has expanded from, let's say, single digits in history a few years ago to double digits, and right now we are seeing those trends actually continue to be improved. So the enterprise cloud margin expansion, high quality initiatives, and the strategy since 2022 has already seen some good results. So this part actually contributing the last one third of the margin expansion. So to conclude, I think we do see a positive trend, even though I understand Sheldon, your concern may be in Q4, you do see a very good Q on Q jump of the margin about three percentage points. But I think we are in the very good momentum and in a confident way that in Q1 and going forward, given the three reasons I just mentioned, our gross margin will continue to be seeing a good result going forward.
So allow me to also translate for the first question. The first question is about the competitiveness of our AI products. So as you correctly commented, indeed in the last year we have comprehensively worked very hard on expanding our AI business. However, it would be hard and less prudent for me to directly comment on the competitiveness of our AI products versus other of our peers. However, I would say that because of our unique positioning of our neutrality and independence, we have, if you look at it from the results, we have becoming a preferred choice of cloud service provider by a large amount of independent AI, large language model AI companies And that is exactly because we don't do our large language model and build our large language model ourselves. And therefore, in this round of opportunities, we have seen as a result, we are already covering a significant number of independent AI companies becoming our customers. And you also asked about the future investment in terms of CapEx. I would say that the rule of thumb is we will be continue to investing based on the demands and the pipeline of the customers we have. However, the specifics as regards to the specific amount and specific dollar amount and the tempo of such investments, it's relatively difficult for me to comment at this stage. Yeah.
Operator, please go ahead.
Thank you. We are now going to proceed with our next question. And the questions come from the line of Daily Life from Bank of America Securities. Please ask your question. Your line is opened.
Hi, everyone. Good evening. Thank you very much for accepting my question. I would like to ask two questions. The first one is still about AI. I would like to ask, as we mentioned before, about the demand. is far greater than our attack capability. I would like to ask, what is the situation of the demand and our attack capability? If the supply is short now, how will our chip solve domestic and overseas purchases? Where will the growth of the AI rental business be in the future? For example, new customers or customers who may not be satisfied with the demand. The second question is about our profit rate. For example, in 2024, our EBIT margin, adjusted EBIT margin, I have two questions. The first one is about the AI business. Could you share some color about the supply and demand trend recently and how do we see the growth drivers for the AI data center AI public cloud business going forward and the key drivers. And secondly, my second question about the margin trend. Looking into 2004, how do we see the margins such as the adjusted EBITDA margin and some costs like a share-based competition cost trend? Thank you.
Okay. Let me answer the first question first, and see if there is anything to add. Indeed, as I said before, the demand of AI for chips is greater than the supply. This is the main force of last year's whole year. So we have been actively satisfying our customers as much as possible. In fact, up to now, everyone knows that some changes have occurred in the entire external environment last year, so it leads to this kind of vacancy relationship. In fact, there is no fundamental change, even to a certain extent, it is still deteriorating. Of course, on the other hand, the rapid rise of the national carbon footprint itself, I also talked about this point last year. In fact, we have been paying attention to it since the beginning of the previous year. In fact, to some extent, it has also played a certain relief role, but in general, this market is still Uh, especially with the upgrade of this model parameter, the third one is getting bigger and bigger. This node is getting bigger and bigger. So from our point of view, this is still a little bit, or the demand in the entire market is greater than the attack. This is not completely solved, and our judgment is that this may be a long-term existence. As long as this continue to continue to develop this trend does not change the situation this I my own judgment also will not be too short-term will not be too fundamental change is good so that in fact, in order to deal with this situation ah in fact, we also have a few different strategies on the one hand, of course, this is what we have always been also predicted this form ah so we can with the domestic some of the business owners also opened a positive cooperation to build a national business of this this area In addition, we are also actively with that because everyone is related to SP this this construction has also been in this this this This is the current situation. Anyway, the conclusion is that the development needs of AI are still abundant, but the supply and demand will still be under great pressure. So we will have all kinds of channels to find ways to satisfy our customers as much as possible. And then you talk about the factors of AI growth, or I'll talk about it first, and then Liu Tao can add it. What I see is that the old user, with the improvement of the parameters of his own model training, you can understand that this is the decision. He is improving his algorithm requirements, so you can understand that some of the requirements of the old customer itself are expanded. Another part of it is actually at least from our side There are some new customers one after another These new customers are not for the sake of doing the big model But in fact, it can be understood that the ability of the big model to make them have business So it also presents this more and more abundant need In fact, some of the clients we have been dealing with in the past or are talking about do not even include some car customers It's true that this is actually happening, okay? This is, uh, this is simply speaking, in fact, it is the demand for old customers to expand and some new ones. This one is going to give the model to the new customers of the business. There are all of them, but at present, uh, the proportion of old customers is getting higher and higher. Yes, but my personal opinion is that this is with this. AI is getting deeper and deeper into people's hearts. It's getting more and more recognized and accepted by everyone. Maybe the future growth of new customers is still very, very big. But I personally believe in this point. Okay, let's see if Liu Tao has anything to add.
I would like to add a little bit of information about the growth points. One is what Chairman Zhou just said. Old customers and new customers are a very important point. Another is that we saw in the last quarter. The model application of Internet customers in the Internet company's demand for growth. So they will have this Internet company based on the model to do training to bring this training increase And they only have the model to use in their own IDC to increase the return power Another one is that it may include things like visual and video This includes the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance of this picture and video generated by the appearance
So, yes, you mentioned that the situation of demand for GPU chips more than supply, which has been the key theme for last year. And that is why during the past year, we have also been trying our best to meet the clients, our customers' demand. Well, the situation this year, there's some changes, right? On one hand, the production of some made in China chips have been alleviating to some extent of the issue. However, we have also seen other factors, especially geographic tension factors deteriorating the situation. So overall speaking, we see that since the AI market is still booming and demand continues The overall theme of the demand and supply balance is still that the demand is largely exceeding that of the supply. And we do not expect that kind of relationship to have any material change in the near-term future. Now in terms of our responses strategy to this situation, on one hand is that we're working with some of the firms that have computing powers within China to building a dedicated computing power zone together. And the second thing that we do is to follow up with new products that are within compliance premises being allowed to supply in China market. And in terms of growth drivers, there are basically two types of growth drivers. One is old commerce, existing or let's say old customers, which are mainly characterized by the independent AI large language model customers that we have been serving recently. And the second type is new customers, which are typically not the independent AI large language model companies. However, those companies that are ready to leverage the capability of large language models to empower their existing businesses. For example, EV customers. Currently, we see that the growth driver coming from existing customers to taking a higher proportion. However, we do expect that in the future, the potential for new customers' demand has higher growth potential. And Mr. Liu Tao, our SVT, also added that there are several types of growth drivers. One is the internet companies training their own models and using their in-house models and by using the inference capability of the computer power to use their in-house models. And secondly is the advance and the launch of Sora that leading to a wave of demand coming from the video side of things. And the third part would be the new companies like Mr. Soto mentioned of EV autonomous driving kind of demand. Thank you.
Yeah, thank you. So I will take on the second question. I appreciate your notice on the expansion on the margin side. So I think there are a few kind of major directions we're trying to achieve going forward at the same time. First of all, we are aiming to keep a Q on Q expansion on the growth margin side. So this actually is driven by the cost cutting on the supply chains and the better automation on the resources we have and cutting certain loss-making computing regions and certain disposal for certain equipment that we don't think are fitting to today's client requirements. So all the combination of actions were contributing to the margin expansion of the gross margin. And as I explained to the first question, those efforts are already in place, and we're going to see those impacts will be gradually delivered going forward in the next few quarters as well. So the first, as I mentioned, is really about expansion on a queue-on-queue basis on the gross margin. So the second part is really about narrow and better managing the expenses and operation expenses, including, for example, the better management of the human capital costs including the internal efficiency and the streaming line, all the internal management initiatives that were actually well contributing for the expansion of the EBITDA margin. So as you can see, our expenses between the line of the gross margin and the EBITDA margin continue to optimize and reduce as well in the past few quarters. So that has proven the management team has already got some good experience and practiced those skills pretty well. And we think we can carry on that in the next few quarters. So our second aim is to improve the EBITDA margin on a continuing basis. So we are confident to see as we make our efforts, our EBITDA margin will be getting to approaching the break even in the short term of time. And the third is we're also adding to another KPI for ourselves in terms of internal management. We're trying to not only making the EBITDA margin break even, but we're also trying to make the company with the better quality of revenue and a continuous optimized revenue mix and picking the right clients and also expanding the AI exposures. We'll be in a very good position to have the possibility to see a good trend of OP margin expansion as well. So I think these are the three goals we are trying to achieve in 2024. While we are not in a position today to give a clear guidance about the timing of the EBITDA margin break even, as well as the profitability on OP margin side, we are confident to say that all the initiatives in place will see that in good results going forward in the next few quarters, probably we can observe. And to a question with the SBC cost, one thing I want to note is that based on the accounting rules, the SBC cost was booked based on the share price at the time when the ESOP and the shares was granted to the employees, not on the vesting period of time. So when the share price was high, those costs will be booked and amortized in the next few years. So I don't think given the volatility of the capital markets for many of the shares today, the SBC line is reflecting the true value, i.e., the market value of the ESOP value we granted to the employees. So those values need to be revisited and recalculated if you want to do a model. But putting all things together, the company has adopted a very prudent way of granting and investing the shares, as you may see from the announcements, and also extending third investing schedules to keep the company employee working a bit longer and with a better incentive. So as you can see, from 2022 to 2023, our SBC cost has declined from roughly 360 million RMB in 2022 to drop about half to 180 million RMB in 2023. So I think we are with a prudent attitude from issuing, granting, and investing the ESOP. And I think going forward, you're going to see a better and a narrow line between the GAAP and non-GAAP margin as well.
Thanks. Operator, will you mind the last question, please?
Thank you. We are now going to proceed with our last question. And the question comes from the line of view. It is Liu from Goldman Sachs. Please ask your question.
Thank you, Mr. Feng, Mr. Henry, and Mr. Clark for accepting my question. I am Eunice Gao, a representative researcher at Unis. First of all, congratulations on the company's strong performance. I have two questions. The first is about the industry. The company was founded in 2024. I will quickly translate myself. Could management talk about the strategic planning for Enterprise Cloud in 2024? And the second question is, we have seen some news on peer price competition in public cloud in early 2024. And could management tell us more about the latest competitive landscape and also things of cloud's comparative advantages? Thank you.
Okay. Uh, I think there are a few new opportunities, but the first one is not a new one. In fact, in the past two years, I have set up a digital company in various places, and then this is the post-pandemic. In fact, this is an opportunity for the country to rise. In fact, since last year, we have been going back and forth. I also introduced some of them at the phone conference just now. I think this is an opportunity, but I think by this year, I think there is a greater opportunity, even this opportunity. We have recently discussed this with this Han Yun's department. Anyway, the more you think about it, the more excited you are. In fact, what is it? Everyone knows that in 2024, this is the data route. This is called the original year. The data is the original year of the production route. Last year, the Ministry of Finance released this draft of the data route. Yes, I want this thing. Uh, this, this also includes, uh, we recently with the plastic lock of Shanghai, the plastic lock of Shanghai, this is all about this piece of communication that has been done very deeply, including the recent one, including some other audit industries in China, and also do some communication. Anyway, how to say that this may be a very big opportunity. Uh, but this may be said to have something to do with the cloud. Our, our point of view is that this tree must be running on the cloud. This must be a cloud tree. It is impossible to be closed in a hard disk. So in the future, of course, he has to make it clear. What does this mean? Simply put, this number has become money. If the number is money, he will let his production flow and trade, etc. This is all opportunities. And these are what we think must be running in the clouds. Yes, so this is what we think is a new opportunity for this industry. Of course, I don't want this. We have been doing some planning and layout in the past. Uh, uh, we will also hold some of these closed meetings in the near future. I want to do some research on this, okay? This is what we have seen in the near future. It's a big opportunity. Of course, this table shows that it's not a big deal for companies listed abroad. It's mainly for companies listed in China. Because this regulation should be China's current special, right? But you can also say that you understand that you are looking for a relatively cheap one. This is this is what I think is the one in Hanyun that I saw this year. In fact, it has nothing to do with any specific industry. It's just that it's just to highlight this tree to become valuable and to become money. Yes. All right. So this is to share these two industry trends. So I think this is a very good question. And we have been conducting internal discussions
around exactly the question you raised. So a few opportunities I would like to share with you. The first one is you might have noted that recently a large number of digital asset companies across China have been established, and that basically led to a wave of state-owned asset companies migrating onto cloud and using more cloud. And the second, which I think is probably a large opportunity that is happening this year, is the the measure of digital assets are showing on the balance sheet. And we should call this year of 2024 to be the first year of this opportunity because of the government policy recently promulgated. We have also been making communications with the Shanghai Digital Asset Exchange and also with professional parties, for example, like auditing firms. We do think that it represents a significant opportunity for cloud service companies and for us because all of these digital assets will be running on the cloud infrastructure and digital assets and those data assets cannot run by itself. So in our words, it's the integration of cloud service and digital assets. So all this whole chain of data data production of data transaction, et cetera, et cetera, are going to be supported by the cloud. And therefore, all of this represents opportunities for us. So it's not opportunity with respect to any particular industry, but a general, but a probably larger opportunity that we can expect to see in recent years.
Okay. Well, what you're talking about is the price drop. To be honest, this is my point of view. The core point is two points. I'm the first one. This is the price drop year after year. It's not this year either. This is from the beginning of our cloud. This is not to say that in the past two years, this price drop, all kinds of price drops, this year, this price drop, tomorrow, that price drop. In fact, there were a lot of cases last year. This one, this one, this one, this one, this one, this one, this one, this one, this one, this one This is a second one, which is the reason for the analysis. My own point of view, especially last year, we were looking for this one for this quality improvement. After this customer bought one, the customer should say that our customer is paying attention to the first point, but the price must be sensitive, but he is still more concerned about the whole product service. You say you are at this price. If the service is not good, the product is very bad. It doesn't make sense either So so that's what I want to express That's the key to price But not to say This is the first element At least I'm from my client This is these years anyway This is around the first point just now The price goes up year after year But in fact there is no Because of this to our business Have a fundamental influence From the perspective of this question All right So it's actually Of course, I also understand I also learned from my colleagues In fact This price is also a part of the cost of the price. In fact, it is based on its list price. In fact, the actual transaction, many of its customers are lower than this discount, so it may be for some new customers. So I said at the end that the price is more of a marketing hand. Anyway, I raised this point last year. It's a marketing hand. In fact, this is for existing customers. Anyway, I was the same last year. Anyway, there is not much impact. All right, let's see if Liu Tao has anything to add.
I would like to add one point. The strategy of the price drop you mentioned is also very big news. The way the other party prices are actually based on annual payback. On the traditional discount basis, there is an additional discount. So he is targeting small-scale annual payback customers. This customer group is actually with the customer group of Jinshan Yun. There is no overlap.
So the second question relates to the price competition in the market. So I would like to share with you two of my core views. The first one is we have actually been experiencing price pressure since the day of our founding. Every year we have been seeing different kinds of price pressure coming from various of our peers in different kind of products. However, at the end of the day, we do not really see any of those price cuts by our peers have any material or significant impact to the market. And from our own performance results, we have also achieved for six consecutive quarters of profitability improvement Now, I would like to say that we think the reason for that is pricing, obviously, is one important factor for customers to consider. However, it is far from the most important factor. And from our own experience, and also as evidenced by our achievements in the past six quarters, focusing on the satisfaction of customers is actually the most important thing. And I would also like to say that my view about this kind of large amount of price cuts by peers. It's more towards the end of marketing and PR measures. In fact, some of these price cuts are based on catalog price. However, a lot of the transaction price that cloud service providers have agreed with potential customers are already deeply discounted and already way below the price cut level that we are seeing today. And also our also added that this round of price cuts from up here mainly focuses on the customers, which are relatively small in scale and are paying their sub-usage fees on an annual basis. And this group of customers actually do not overlap with the customers that Kingsoft Cloud enjoys. And therefore, it does not have any material impact to our pricing strategy or our business performance. Thank you.
Thank you, Dr. Reza.
We have no further questions at this time. I would like to hand back to Nicole Shan for closing remarks.
Thank you once again for joining us today. If you have any further questions, please feel free to come here. Look forward to speaking with you again next quarter. Have a nice day. Thank you.
Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect your lines. Thank you and have a good day.