Earnings Call: H1 2017

Aug 24, 2017

Good evening, and welcome to Meitu's 2017 interim results conference call. I am Helena Chen from the IR team. Joining me today to discuss our results are Cai Wensheng, Chairman, Wu Xinhong, CEO, and Gary Ngan, CFO. I would like to take this opportunity to remind you that our remarks today may include certain forward-looking statements. A number of risks and factors beyond our control may cause the actual results to differ materially from those contemplated by this forward-looking statement. During this call, we will present both IFRS and non-IFRS financial measures. We may also discuss general market conditions for our industry, and such information may come from a variety of sources outside of Meitu's. For a detailed discussion of the risk factors we face and the non-IFRS measures, please refer to our public documents on corp.meitu.com. Now, I would like to turn the call over to Mr. Cai. He will present in Chinese, followed by translation. Thank everyone for joining our call this evening. Here are some key highlights for our interim results. For the six months ended June 30, 2017, total revenues increased by 272.3% year-on-year to RMB 2.18 billion. Non-IFRS adjusted net loss decreased significantly by 87% year-on-year to RMB 33.2 million. Strong revenue growth and associated operating leverage are the key drivers. According to unaudited monthly management accounts, the group generated positive net profits on March and May 2017, demonstrating our ability to achieve profitability. Our mission of making the world a more beautiful place means more than photos and videos in the virtual world. It covers making users more beautiful in real life, as well as more beautiful internally. We believe this involves the process of educating people about the concept of beauty. On the software side, our photo apps are already setting the trend of beauty. Meipai is also filling users with positive energies. On the hardware side, it is a way of materializing beauty. Smartphone is just the beginning. Meitu can and will expand to other products that are related to beauty, and there are huge market opportunities for these products. However, Meitu's products will always carry a few characteristics: gearing towards beauty, female, and mid to high-end products. We only monetize a specific target market segment, that is, the one with customers who are capable of and willing to pay a premium to become more beautiful. It is exactly because this model is highly replicable on our user base that is over a billion in size, we see there is a high margin of safety to invest significantly. It is also exactly why we are confident that we can make the world a more beautiful place. This is not just a slogan. What we are going to do. Also help user to become more beautiful in reality. [Foreign language] Now let me hand over to our CEO Wu Xinhong to share the latest developments of our company. 好的，谢谢文胜。大家好，在2017年的上半年，我们继续贯彻让我们用户变得更美的使命。作为一家数据驱动型的公司，我们正努力通过大量的数据和先进的技术来更好地了解我们的用户。 Thanks, Wensheng. In the first half of 2017, we continued to carry out our mission to empower our users to become more beautiful. As a data-driven mobile internet company, we are striving to gain better insight into our users through our large amount of data and advanced technologies. 我们活跃的用户基础是我们最为宝贵的资产以及数据的核心来源。2017年6月，我们的MAU达到了4.81亿，与2016年6月的4.46亿相比，同比上涨了7.8%。其中海外MAU同比增长35.9%，达到了9374万。 Our active user base is our most important asset as well as the key source of data. We are pleased to announce that our total MAUs reached 7.8% year-on-year growth to 481 million as at June 30, 2017, compared to 446 million in 2016. In particular, overseas MAUs reached 93.74 million, up 35.9% year-on-year. 我们的平台化工作取得了重大的进展。截止2017年6月30日，我们累计注册的美图账号MTID增长至大概2亿个。同时，我们首个平台产品美拍的用户增长趋势也令人鼓舞。美拍的MAU在2017年6月达到了1.52亿，与2016年的1.14亿相比，同比增长了7.9%。更重要的是，美拍应用类的MAU同比增长33.3%，意味着更高的用户粘性。 We are making great progress in our platformization effort, as reflected by the growth of our accumulated registered Meitu accounts, or MTID, to approximately 200 million as of June 30, 2017. Also, the user trend of our first platform product, Meipai, was encouraging. As of June 30, 2017, the MAU of Meipai increased by 7.9% year-on-year to 152 million, compared to 141 million in 2016. More importantly, Meipai's in-app MAU grew 33.3% year-on-year, indicating much higher user engagement. 我们十分重视美拍生态系统的建设。为了激励内容创作者制作更多的视频，首先我们升级过的推荐算法不仅能够找到他们感兴趣的视频，更能够预测并替他们发现新的兴趣点。其次，我们也展开了多项举措，协助内容创作者变现，从而形成一个正向的循环，刺激他们创作更多的内容。例如，我们推出的美拍M计划平台，可以高效地为网红和品牌广告之间进行匹配，让网红可以轻松地通过植入广告赚钱。另外一个举措是美拍的边看边买。这个功能使内容创作者能够将观众转化成顾客，从而增加其产品的销售量。其中一个让人鼓舞的例子就是这个功能帮助一个在美拍上售卖海鲜干货的网红，提升了400%的销售转化率。 We have been very focused on building Meipai ecosystem, aiming at incentivizing content creators to produce more videos. Firstly, we have upgraded our recommendation algorithms to not only match users with video of their interest, but also to predict and cultivate new interests for them. Secondly, we have also rolled out a number of initiatives to help content creators to monetize their talents. This in turn starts a virtuous cycle to incentivize them financially to create more content. For example, through the launch of Project M platform, KOLs, who are content creators with a large number of followers, are efficiently matched with brand advertisers, so they can make money by producing videos that incorporate marketing elements from those brand advertisers. Another initiative is video shopping, a feature that enables content creators to convert their audience into customers, hence increasing the sales volume of their products. One of the many encouraging examples for this new feature is that a KOL who sells dried seafood online managed to increase her sales conversion rate by 400% using this feature. 就我们的影像应用而言，美图秀秀及美颜相机的MAU同比增长率分别平缓至9.7%以及1.3%。这是因为在2017年的上半年，我们将大部分产品开发资源集中用于筹备这些影像应用的平台化，而不是推出多种工具功能来促进用户的增长。但是出于长期发展的考虑，我们认为这个阶段的投入是必须的。通过将这些影像应用平台化，我们能够提高用户的活跃度和留存率，以及更好地洞察用户的喜好，从而创造更多的商业化机会。For our photo apps, MAU year-on-year growth rate of Meitu and BeautyCam has moderated to 9.7% and 1.3% respectively because we have focused most of our product development resources in practice of their platformization, as opposed to launching multiple utility features to fuel user growth in the first half of 2017. However, we see this as a necessary investment phase for the long-term prospect of these apps. By transforming these photo apps into platforms, we will be able to increase the engagement and retention of users, as well as generate more monetization opportunities through better understanding of our user preference. 在技术方面，我们已经开始于不同产品融入AI技术。例如美拍类的百变背景特效，就采用了以AI为基础的视频分割技术，达到实时移除视频中用户以外的背景。这不仅让用户可以创作更多有趣和有创意的视频，亦可以为美拍带来更多的广告商业化的机会。这是有更多互动的品牌背景特效。此外，AI也帮助我们在美拍上对不同的视频打标签，以提高对内容分类以及分发的准确率，同时帮助我们搭建更全面的用户画像。我们目前已经拥有上百个标签，并打算通过深度学习将其扩展至上千个。另外，我们也计划使用我们在2017年初刚升级的推荐引擎，来推动我们日后的精准广告业务。On the technology side, we have started to incorporate AI technologies into various products. For example, the creative backdrop filter on Meipai employed an AI-based video segmentation technology to remove the background from a user-generated video in real time. This not only empowered users to create a lot of fun and creative videos, but also opened doors for Meipai to work with advertisers in creating branded backdrop filters for more interactive marketing. In addition, AI has helped us to label different videos on Meipai in order to improve the accuracy of content classification and distribution, as well as building a more comprehensive user profile. We now have hundreds of labels, and we aim at expanding them to over 1,000 through deep learning. Furthermore, we have upgraded our recommendation algorithm in early 2017, which were one of the key drivers for Meipai's significantly increased in-app traffic. We also plan to apply this recommendation engine to serve targeted ads in the future. 总的来说，我们在美图所做的每一件事情都围绕着用户，我们也会一直贯彻以用户为核心的策略。To sum up, everything we have done here at Meitu are focusing on our users, and we will continue with this user-centric approach. 现在我交给Gary来进行业务和财务数据的更新。好，谢谢。 Thanks, Xinhong. Good evening, everyone. I will take over and talk about financials. Results of the first half of 2017 were strong. Total revenue grew 272.3% to RMB 2.18 billion. IFRS net loss reduced by 94% to RMB 131.8 million, compared to RMB 2.19 billion in the first half of 2016. Adjusted net loss reduced significantly by 87% to 33.2 million. As our chairman mentioned previously, we have proven our ability to become profitable in March and May of 2017 on an audited basis. If we analyze a bit deeper, we believe these results have also cleared a few market misperception of our company. Firstly, revenue of our Internet service segment, which includes both advertising and IVAS, have grown over seven times to RMB 246.8 million. This segment now accounts for 11.3% of our revenues, compared to only 4.9% in the first half of 2016. Advertising revenues grew 219% to RMB 82.6 million, primarily driven by increase in the number of advertisers. However, we are still prioritizing quality over quantity of advertisers. We aim to further grow our advertising business, mainly through launching new advertising formats such as branded filters and stickers, introducing promoted feed ads to Meipai, and introducing targeted ad products to enable advertisers to reach specific audience groups. IVAS revenues grew more than 59 times to RMB 164.2 million. Average monthly paying users have increased to 324,000 from 58,000 in the first half of 2016. ARPPU or ARPU has also increased. We have continued to differentiate Meipai's live streaming content from other live streaming platforms. The strong growth from this segment indicates that our Internet business does have a very solid business model, and is highly scalable because of our massive user base. This is a very different performance compared to the misconception of the market that we don't have a business model for Internet. Now let me turn to our smart hardware segment. Revenue increased by 247% to RMB 1.93 billion. We sold 847,000 units in the first half of 2017, compared to 289,000 a year ago. Average selling price, or ASP, also increased to RMB 2,272 this year from RMB 1,900 last year. The ASP growth is mainly due to the introduction of T series, which is a higher priced model compared to the M series. Gross profit for the first half of 2017 grew about five times to RMB 435 million, representing a gross margin of 20%. This compares to RMB 74.5 million of gross profit and 12.7% gross margin in the first half of 2016. The key here is that our Internet services and other segment has turned profitable at a gross level, generating 15.8% gross margin in the first half of 2017, compared to a gross loss margin of 119.5% a year ago. If we break down the cost of revenues for this segment, we will see a number of costs are actually not that variable with revenues, such as bandwidth and storage-related costs and video content monitoring fees. These costs mainly associated with Meipai and generating a dollar of additional revenues from advertising doesn't really change these costs. As a result, we have seen very strong operating leverage for this segment. Going forward, the gross margin for this segment will mainly be driven by the business mix of advertising, IVAS, and e-commerce. For the smart hardware segment, gross profit grew 264% to HKD 396 million, representing a gross margin of 20.5%, which is one percentage point higher than that of first half of 2016. In terms of expenses, they are very straightforward. Expenses as a percentage of revenues decreased across the board. Majority of the expense grew due to an increase in headcount, personal related expenses. We ended the first half of 2017 with 1,715 employees, compared to 1,328 at the end of 2016. There was no significant fluctuation in our overall promotion and advertising expenses between the first half of 2017 and first half of 2016. As usual, vast majority of our user growth came from organic downloads as opposed to paid marketing. Loss for the first half of 2017 reduced significantly because a fair value loss on the convertible redeemable preferred shares of HKD 1.9 billion was recognized in the first half of 2016, but did not recur this year. All the preferred shares were automatically converted into ordinary share upon listing in December 2016. Touching a bit on the balance sheet, our cash and other liquid financial resources were HKD 5.3 billion as of June 30, 2017, compared to HKD 5.5 billion at end of 2016. We are very well capitalized, and we do not have any debt or borrowings. In terms of CapEx, we have spent about HKD 24.8 million in the first half of 2017, slightly higher than that of HKD 21.8 million a year ago. Let me finally talk a little bit about our outlook. Firstly, we said in the last result we will sell 1.5 million smartphone units this year, and now we expect to sell a bit more at 1.6 million for 2017. However, we note that proportion of sales of T series, which carries a high gross margin, will reduce in the second half of 2017 compared to the first half. Secondly, we believe it is still necessary to accelerate investment in our people, technology, and user growth, mainly in terms of expenses, for the second half of 2017. Therefore, we continue to increase our expenses across the board. As a result, we expect to incur a net loss for full year of 2017, an outlook that is unchanged from what we have discussed during the IPO. This is all our prepared remarks, and we will now open the call to Q&A. Thank you. Thank you, management. If you have a question, please press zero one on your telephone keypad now and you enter a queue. After you are answered, please ask your question. If you find that your question has been answered before it is your turn to speak, please press zero two to cancel the question. Once again, zero one on your telephone keypad now to ask a question. Our first question is Binyan Wang from Merrill Lynch. Please go ahead. Hi, thank you, management, for taking my question. First I have a question on Meipai. How is our average, in terms of ARPU, how is it trending? Also, can you just give us some color in terms of like, on Meipai, in terms of like, the differentiation you see against other platforms, like broadcasting platforms. My second question is that on our revenue growth. It seems as this half, like first half of the period, is still largely driven by the growth in our hardware sales. I understand that IVAS is actually increasing as a percentage of total. How are we seeing in terms of the three segments, advertising, the Meipai, and also e-commerce, how do we see each of those growth will trend? Any key metrics that you can share into these three platforms will be very helpful. Thank you. Okay. In terms of the ARPU, I think you can actually calculate from the announcement, but, as a trend, it has been increasing since we launched our service in the first half of 2016. In terms of the differentiation of Meipai, I think, number one, the content of Meipai is very lifestyle-oriented as opposed to most other live streaming platforms that are focusing more on the showroom business. The other key differentiation is Meipai fundamentally is focusing on short-form videos and live streaming, it's a secondary or support function. I think the main role of live streaming is to enable the KOLs to interact with their fans more easily, but the core is still short-form videos. To answer your question on revenue growth, I think the non-smartphone business has been growing very rapidly. Advertising growth is mainly driven by the number of advertisers. But as I said, overall strategy-wise, we are still focusing more on quality. I think the flip side of that, the implication of that is once we continue to build a business, you will see both the number of advertisers and also the revenue per advertiser will continue to grow in the next few quarters. Obviously IVAS is driven by both an increase in ARPU and paying users, and we are still seeing very strong growth coming from this segment. E-commerce is still at the testing stage, so we are now working on our product. I think the overall strategy of e-commerce is to find a point of differentiation, which will set us apart from different e-commerce services such as Taobao or JD.com. We are also very focused on incorporating some AI or artificial intelligence technologies into our e-commerce platform. I think you'll be able to see some of these products being launched in the second half of 2017 very soon. Okay, thank you. I just have one follow-up on the MTID, because I think management introduced it, talked about more in your opening remarks. How active are you talking about those MTID in terms of your 200 million, understand there is more registered accounts. How active are they? Then, in the longer term, do we see the cross-selling, or maybe the cross-users across our various products? Which I think that after we grow our users in certain products, then we can achieve cross-selling or maybe cross-users across those. Any color of that would be very helpful. Okay. I will take this question. MTID, if you look at the history of the MTID is mostly coming from the Meipai product. By definition, the Meipai users are obviously a lot more active than the utilities or photo apps users. To give you a sense, Meipai users spend over 30 minutes a day, whereas, most of the photo apps spend about 5-6 minutes a day. So you will be able to see the MTID activities. Currently, it is much higher than the non-registered users. But even for the non-Meipai users, the users using MTID obviously are more active than the non-registered user as well. Thank you. The next question is Kim Chong from CCBI. Please go ahead. Hi [Foreign language] Gary, thank you for taking my question. I have a question, when do we expect to see meaningful revenue contribution from e-commerce? Any updates on Meifu or MeituDIY would be very helpful. [Foreign language] OK。其实美图电商我们是从去年的第四季度才开始筹备，不管是从人员，包括产品。所以到现在半年过去，我们更多还是在一个试验阶段。所以你要看到一些真正的大的进步，应该是会在明年。我知道很多人对美图电商有很多的期待，但想一想，比如当时的淘宝跟京东，在前面几年几乎都没有收入，也没有利润。美图当然不会等待那么久，但是如果希望美图在短期几个月就看到一个大的进展，会有挑战。那我们要说的就回到美图。我们现在电商的团队组建得差不多已经有100多人，另外我们的确有几个产品，不管是美肤，还有美图定制，目前我们会倾向于主要主打两个，一个是美图定制，这个的确会跟我们产品非常好的结合。那么美肤这个方面，我们会调整一下，因为之前有点设得有点泛，我们接下来可能会主打美妆。美图美妆这些跟化妆品有关的东西。这个方面我觉得会在下半年就开始进入一个成长，但是真正的业绩体现我觉得会在明年。对。谢谢，我有一个follow-up。 Sorry, excuse me, let me just translate the answer. Yeah. The e-commerce business really started in Q4 of 2016, and six months later, we are now testing a lot of different concepts. I think in order to see really huge progress in terms of revenues will be taking us to 2018. We know that there is a lot of high expectation for our e-commerce business, but if you look at Taobao or JD.com, in their first few years, the business is actually quite slow. We do not think it will take as long as what they have taken, but if you want to see a great ramp up in recent months, that will be quite challenging. Right now we have already have a team of over 100 people and the two major businesses are Personalized services and Meifu. Personalized services is a very complimentary business to our user base, whereas Meifu, going forward, will be more focused in terms of the business model, especially on cosmetics and skincare products. We should be able to see revenues or business picking up in second half of this year. But to become a meaningful drivers of revenues, we will probably have to take it to 2018. 我有一个follow-up question，就是关于这个Google广告平台的合作的。我想问一下，现在目前只是在港台地区是吗？然后9月1号开始的话，对我们的业绩体现到今年会有四个月的影响，是这样吗？谢谢。翻译成什么？ Yeah. The question is the cooperation with Google, it is focusing on Hong Kong and whether it is focusing on Hong Kong and Taiwan exclusively, and whether we will have four months of full contribution from 1st of September. 我们跟Google的合作其实是分两个部分，一方面是从技术方面的AI，我们会跟他一个站内的合作。那另外一个方面的确就是广告平台的港台的合作。那对于美图来讲，我们在国内已经开始开卖广告，之前我们港台跟海外的并没有开卖，那这次跟Google的合作其实刚好可以弥补我们这方面的发展。当然在这个发展的过程，如果海外做得好，我们也会加入自己的团队来港台或者海外的广告。9月1号会开始测试，但是对于我们来讲，我们是把它定为一个surprise惊喜。如果说这个测试是好的话，那会增加很多，但是我们自己对它的预期更多也是一个弹性，因为之前海外的用户，我们并没有纳入一个广告系统，大概这样。 The cooperation with Google is two folds. On one hand, we have the advertising partnership with them, but also we are working something on technology and on the side of artificial intelligence. Going back to the advertising, right now the 1st of September is the start of a testing phase. If the result is going to be good, then we will also start to deploy more resources or sales resources ourselves into selling ads in these two jurisdictions. But I think overall speaking, we should really treat that as a surprise on the upside, because we have not really had guided that we will have much monetization in these two jurisdictions previously. 但有一点，因为那天我刚好也参加了在香港Google的发布会，我们现场差不多有将近100个各种品牌的奢侈品副总裁，大家对美图都是非常欢迎的。因为以前都找不到和美图的产品来做合作广告。所以那天我们从反馈来讲，我们还是觉得蛮欣慰的，蛮开心的。 I just so happened to also attended the Google event. In that event, there are over 100 advertisers from luxury brands and products. I think the reaction has been quite overwhelming and we are quite pleased with the reaction from the advertisers, because they do not really know that they can work with Meitu to advertise their products previously. Thank you. Our next question is Warren Wong from FSCSS Management. Please go ahead. 大家好，首先恭喜管理层，谢谢Gary对我们的支持。有几个问题，第一个就是互联网增值服务能不能讲稍微详细一点？因为我看到我们年初有公告，而且现在互联网增值服务这一块占比也很高，增长得也很快。能不能讲具体一点什么内容？为什么增长这么快？未来有什么展望？这个是第一个问题。这个我来回答一下。我们现在用户增值服务大部分都是直播的礼物打赏贡献的。那未来还会有两块用户增值服务，一个是软件里的应用类购买，就是某个功能需要用户付费后才能解锁。另外一块是VIP会员的服务。然后往后可能还会有一些针对用户的数据服务，也会一起纳入到增值服务里面。 The question is what is the differentiation and the outlook of the IVAS segment because it has been growing quite rapidly. Our CEO replied that right now most of the revenues are coming from the tipping of live streaming. Going forward, we will still have a couple other IVAS businesses such as the in-app purchases. For example, you will have to pay in order to unlock certain features of the apps. We will also start to work with VIP memberships as well as some data related services providing to the users. 好，谢谢。第二个问题就是，上半年还是超市场预期的，能不能对下半年收入还有手机的销量做一个展望？尤其是手机这一块，它产能、供求，还有定位，因为我们也看到好像拍照手机竞争日趋激烈。就是对下半年我们的业务还有手机这一块做个展望。谢谢。 The question was mostly related to the outlook of smartphone sales. I think we actually have said that in the prepared remarks that we will be shipping about 1.6 million units for the full year of 2017. That is our expectation for the full year. 但上半年好像势头应该是整个情况的收入和手机的销量都比预期要好。全年的目标你们现在会有调整吗？ The question is, because the first half has been very positive, are we raising our forecast of the smartphones shipment this year? I think the answer is yes, because previously we talked about 1.5 million units for 2017, and now we are revising that up to 1.6 million units. 好，这最后一个问题关于平台化，我们在图片社交上原先蔡总讲过，我们现在看到除了美拍之外，其他还有什么产品的考虑吗？类似Snap这种。应该这么说，我让我们CEO吴欣鸿说一下下半年我们可能会推出的一两个产品。现在平台化一直是美图最核心的战略。作为平台化的支撑，首先我们要把美图账号推广得更好，让更多的用户登录使用美图的产品。大家也看到我们目前比较成功的平台化产品是美拍。美拍是一个从工具到社区的一个典范。我们其他工具产品也会陆续地平台化，就比如说美图秀秀，应该说很快就能看到它从工具往社区去延展。具体的节点我可能还没有办法告诉你，但可以肯定地说，我们所有的产品未来都会往平台化。 The question is, platformization is one of your key strategies, so what are your product plans? Our CEO replied that the core of the strategy is to increase the number of users, proportion of users that log on to our apps using a registered account. If you look at Meipai, it is a very successful example of a utility app transforming into a social product. For most of the other apps, we are also adopting this similar strategy. For example, Meitu Social, it is also going to become a social platform in due course. Obviously, I cannot really disclose the timing at this moment, but I think the strategy is for most of our other apps to also adopt similar strategy. I just want to supplement with two points. One, all of our product going forward will be supported by AI technologies, and also will embrace some of the most popular features such as cloud-based beautification and augmented reality features. Okay. From a financial management point of view, we tend to be slightly more conservative because we want to be able to achieve the guidance that we are giving. But the first half of this year, we have launched a couple of new products and, if the products that we launch in the second half also beat our expectation, then obviously the end result will be higher than what we have guided today. Maybe we can talk about this offline because it is very detailed. Next question is Minnie Wong from Merrill Lynch. Please go ahead. Hi. Thank you for taking my follow-up question. I understand that advertising this quarter, one of the key IVAS has been growing fast, is also getting more meaningful. I understand that you are talking about that is the AI that potentially will have a lot of these branded backdrop features. Do you think this is something like what Snap Inc. has been doing? What are the major advertiser industries that we are seeing, and how do you think that this will trend, given that because our user base predominantly is still female-driven. So how does our advertising mix base now and how do you think that will trend? I guess across the three segments, do you think advertising, live broadcasting, e-commerce, which one of these, if you have to rank it, which one do you think our company would be focusing most of our efforts, or are they just going side by side along with each other? Sure. In terms of the advertisers, I think what we are seeing right now, it comes really across many different industries. Obviously, FMCG, luxury brands, skincare, cosmetics, they are the key pillars. But we are also seeing e-commerce business, some other apps, games. I think right now, because the user base is actually quite massive, even though it is sort of geared slightly towards female, but it does not change the fact that there is a huge user base and a lot of advertisers want to tap into that. I think the point I here want to make is, even if we just tap into the luxury, skincare, cosmetics, all the so-called female-related verticals, then it will be a huge opportunity for us already. That is the answer to your first question. I think second question, that is actually quite a good one. From a resource deployment perspective, we are looking at the three segments pretty much one segment. The reason is because the whole idea is to allocate our traffic to the highest return on investment business lines. As what the chairman has mentioned previously, for a certain business, the ramping up phase will take slightly longer than the others. For example, e-commerce, I think it will be one of the major, or even it could potentially be the largest sub-segment in the future. But in the near term, obviously, the ramp-up speed will be much slower than that of internet value-added services. I think the way I look at it is, you look at how many people are actually paying for the service, and then you will be able to understand how fast this business will ramp up. Because if you look at IVAS, the paying users is only a few hundred thousand people. Could become maybe 1 million people in the future, but still, it is a very small amount of people. Whereas e-commerce, you are talking about could be a few million or tens of millions of people paying for it. That is why you need time to slowly ramp up this business. But by the same token, e-commerce, because there are so many paying users, in fact, the revenue potential could be highest. But then also margin on that side would be lower. I would imagine advertising would have the highest margin, and then live broadcasting, and then e-commerce, based on what existing business model or existing companies are doing with this. Not necessarily. Not necessarily. Okay. Yeah, for Meitu. Yeah, I think advertising will definitely be the highest margin, but e-commerce versus IVAS, it will depend on the mix. It really depends on the mix. All right. Thank you. As a reminder, please press 01 on your telephone keypad now to ask a question. The next question is Chen Wang from TM Security. Please go ahead. The question is talking about the overseas users and what is the strategy and what is the distribution of the users. Which country has the higher overseas users, and what apps are these users using primarily? For both Chinese and overseas users, we have seen very good growth for this first half. Obviously, I think in order to fuel the growth further, we are focusing our user acquisition strategy in fourth- and fifth-tier cities. For example, like what Weibo has done previously. I think if you compare the Chinese user and the overseas users, the overseas users' activity level is not as good as China. I think there are some cultural differences and also definition of beauty that drive the differences. To also answer your question, I think most of the users are using the app called BeautyPlus in overseas countries. The three largest countries, in terms of users we have Hold on just one second. There are India, Indonesia, Thailand. They are the three largest countries. We also have quite a bit of users in Japan, Korea, and United States. The question is: what is the split between SME and KA currently? Because the perception is that the promoted feeds are mostly for SMEs, whereas the display ads are mostly for KAs. In fact, right now, all of our advertisers are KA advertisers. We haven't really started the promoted feeds business. We haven't really disclosed the number of advertisers, but if you look at the revenue size right now, most advertisers will have a budget of around RMB 500,000 to RMB 1 million per half year or per project. You can sort of back out a range of advertisers that we have at the moment. As a reminder, please press 01 on your telephone keypad now to ask your questions. This is the end of the Q&A section. Also, this is the end of the call. Thank you everyone for attending.