Rui Ma [0:00]: Hey everyone, sorry for the long hiatus, but we’ve been working on some new stuff. First of all, we’ve soft launch Tech Buzz China Insider, which is a paid community for investors, operators and anyone else interested in getting into the weeds of China Tech. Our slogan is to build the smartest community for China Tech on the internet. And we’re offering timely news, insightful analyses, a panel of experts, members only events, and basically the most thoughtful place for learning about China Tech while building relationships with other like-minded individuals. If this sounds good to you, please go to techbuzzchina.com/insider to apply. We do have a brief entrance quiz, you must take and pass. But if you’re a regular Tech Buzz listener, we’re pretty sure you’ll pass it with flying colors. And we’re also releasing a series of live casts. As the name suggests, live casts are conversations that are recorded live and then edit it for your listening pleasure. Some of these chats are open to the public and we welcome you to join us. So be sure to follow us on Twitter or sign up for our mailing list to get notified. And today’s episode is actually one of them.
It’s an edited version of our webinar with BigOne Lab the leading alternative data firm in China. We have a quarterly partnership where they share with our listeners like yourself some of the most interesting data driven analyses on China tech. BigOne was founded in 2016 and has quite a complete suite of data products to track the operational metrics of the most followed companies in China, like the ones we will talk about today. That’s Bytedance Meituan, PinDuoDuo, and more. They’re regular advisor to my old employer, Morgan Stanley, China, and their data is truly superb. If you’re interested in learning more, please go to bigonelab.com for more information. In today’s episode, we had the pleasure of talking to Robert Wu BigOne’s Head of Research on the current state of China’s ecommerce, beginning with an update on community group buying, which we covered back in TechBuzz Episode 80. We then talk about the broader competitive landscape between the now six major ecommerce platforms that’s JD, Meituan, Alibaba, Pinduoduo, Douyin, and Kuaishou. And at the very end, we do a little bit of a deep dive into how Douyin and Bilibili work in their respective video businesses and how that shows up in their overlap of creators. Let’s get started.
Rui Ma [3:00]: Hi, everyone, we are Tech Buzz China by PanDaily powered by the Sinica podcast network by SupChina. We’re regular podcast focused and giving you a peek into what’s buzzing within the tech community in China. we uncover and contextualize unique insights, perspectives and takeaways on headline tech news that don’t always make it into English language coverage so you can be smarter about the world of China tech. Today, it’s just me, your cohost, Rui Ma, I really want to thank those of you who subscribe to our YouTube channel, which is located at www.youtube.com/techbuzzchina. And I also want to thank those of you who gave us a review on Apple podcasts; we’re now really, really close to 200 reviews, however, we’re still looking for more reviews so if you like today’s episode or any of our episodes, please take the time and leave us a review on iTunes or wherever you get your podcasts. Thank you so much.
Thanks, Rob. Thanks so much for being here.
Robert Wu [4:08]: Sure. Hi, everyone, this is Robert Wu, founder and head of research at Big One Lab we have more than 60 team members now most of us are data engineers, data analysts, scientists and product managers. We were founded by former members of other fintech and data company like YipitData, Bloomberg, BlackRock, etc. and we were also backed by S&P Global. We were actually the first and currently the only Chinese company ever invested by S&P Global.
Rui Ma [ 4:41]: Tell us a little bit more about the data products that you offer at Big One.
Robert Wu [4:46]: We offer data products in the format of terminal, a web-based terminal where you can log in and check out our data. We also have customized data offerings, API deliveries, we also offer reports and sessions like this to our clients. We mainly serve institutional investors at the moment, we also start to serve more and more of the corporates as well. We’re pretty strong in the ecommerce side. We cover not only household names T-Mall, JD, we also cover emerging ecommerce platforms like Douyin, Kuaishou, which I’ll refer to during this presentation. We also cover other ecommerce platforms across Asia Pacific area, like Macquarie of Japan, Sharpie and Tokopedia in Southeast Asia, and we also covering this whole sector, CGB, community group buying. We are pretty strong on the entertainment side, we cover all of the most relevant live streaming video, short and long video platforms in China, including Douyin and Kauishou.
Rui Ma [5:49]: So I know for community group buying, which is basically group purchase online, and next day self-pick-up of groceries and other items in rural China right now, we actually did a whole Tech Buzz episode on it, episode 80, for those of you who don’t know it. I know that you guys at Big One have a whole dashboard on it now.
Robert Wu [6:10]: We’ve been tracking this for more than one year ago actually. One advantage we have compared with some of our peers is that beyond public market funds, hedge funds mutual funds, we are also helping a lot of private equities and venture capital funds, mostly in China who are looking out for deals, looking out for emerging sectors. So that helped us to investigate the sector way before the wider public has pay attention to it, we’re actually lucky enough to join the due diligence efforts of top independent players like Xingsheng Select and Shihuituan. We’ve started track these data from pretty early on, from last September when all these major internet companies jumped into this game, we also start to track them instantly. Currently, we track nine of the leading platforms.
Rui Ma [7:00]: Okay, so what is the first piece of data that you think we should look at when it comes to community group buying?
Robert Wu [7:07]: First, when we look at this, because we cover all of these platforms, one thing we could do is to try to estimate what is the total addressable market of this sector. What is current the penetration rate, what kind of the room for growth for this sector. We sum up all the GMV that we’ve been tracking from all these platforms and we compare them with official statistics. The top two provinces right now are the southern province of Guangdong and central province of Hunan. These two provinces total CGB value account for between 2% to 4% of fresh grocery and FMCG sales. For most of the provinces, this penetration rate is either 1% or less, for most of them, they are actually smaller than 0.5%. So right now, we’re really at an early stage of this business model and we do see a lot of room for growth.
Rui Ma [8:00]: So yeah, it sounds like if you believe that rural Chinese people are going to be buying everyday products online via CGB-community group buying and this is a viable model, then I think you can definitely make the case that there is a really long way to go still for the sector. So okay, can you share with us what the Q1 market share data looks like? Especially geographically.
Robert Wu [8:24]: For Q1, the largest province is still Hunan but we start to see the gap between Hunan and Guangdong start to grow. Actually, if you just look at the end of March, Guangdong is now bigger than Hunan, as more and more players are investing a lot of their resources into this province; it’s one of the rich provinces basically. So, several things stand out to us. First, the independent players like Xingsheng, Shihuituan, and Tongcheng they still are big players in many of the key provinces. In the biggest province of Hunan, Xingsheng still dominates. Several months ago, when these internet companies started to invest heavily into this many people are afraid that due to huge subsidies programs and high-pressure strategies that Xingsheng may lose their strongholds in the provinces they started in especially in Hunan and Hubeil but it turns out Xingsheng is still pretty strong. If you look at the prices, if you look at ESP, the ESP of their goods only trend down slightly compared with, say half a year ago. Somehow, they managed to hold on to what they have built a tight knit networks of communities across different layers of regions, like from the provincial level to the city, to the prefectures to the counties to the villages, they managed to do that. And their community organizer to stay loyal and their consumers stay loyal in the provinces they were strongest and they’re still strong. Now similar things happen for often overlooked player like Tongcheng and Shihuituan, right now they among the leading players in the province of Guangdong. Among the top internet players, Meituan seems to have a significant share on most of the provinces it seems like this is where their networks of the distributors, their agents in the food delivery age, this is when the network start to show the value. Their growth is high quality across all the provinces. We also try to measure the efficiency of what they call the central warehouses, zhongxinchang. Each player uses a central warehouse to manage and to cover a group of cities. So the one that has the most central warehouses is Chengxin, they actually have 115 Central warehouses by end of February while Meituan had 84, PDD has 75. If you compare it per warehouse per day units sold Meituan is the highest close to 200,000 units sold per day per warehouse compared with only 124,000 for Pinduoduo.
Rui Ma [11:03]: Chengxin, by the way, is DD’s initiative. It’s like a play on words of their company color, which is orange and also it happens to be sort of a homonym for the word for sincerity. Tell us more about Chengxin.
Robert Wu [11:17]: Chengxin is actually the lowest, right now it is very heavily centered around selling fresh groceries, almost all of what they sell is fresh groceries, while other platforms are already progressing toward non-groceries sales. The more fresh groceries you have the more demand for warehouse capabilities, especially essential warehouses, and you become more fragmented. The one thing we look at is the SKU composition and we do see very different strategies between different platforms when it comes to what they sell. This is the top 20 selling SKUs in the last week of March for Xingxheng, most of the top selling SKU goods here are branded items, top Chinese meal products like MengNiu and Eli, and actually the 19th product here is a Maotai. This is actually remarkable because when you look at say when you think about CGB, you think about replacement for the wet markets. When in reality, the top selling goods now are usually the branded items and even some high-quality branded goods like Maotai, the Chinese top super premium baijiu brand. Meituan and PDD also have similar patterns, if you just compare with Chengxin this is also the top 20 selling goods for the last week of March. And you see the fresh goods everywhere, some top items include durian, shrimp, oranges, eggs are actually the most popular grocery items in all of these platforms. They’re just very standardized and they are very easy to be used as a marketing tool. So Didi’s Chengxin Select is still a pretty much a platform for selling fresh foods, it could be due to the fact that they are actually one of the weakest players in terms of their supplier resources. If you look at another independent player of Shihuituan, they are actually also one of the earliest CGB players, then you see a totally different pictures as well, among their top 20 is almost all iPhones. They still are selling fresh fruit, but you cannot even find fresh fruit among the top 20 selling goods.
Rui Ma [13:26]: This could be good or it could be bad. It could be good, because this is why everyone’s paying attention to community group buying right? It’s because starting with such high frequency purchases, groceries, and then of course, it’s probably going to be a stepping stone for broader ecommerce. Then there’s the question of whether this being done too early, or people going into broader ecommerce too early Is that something that should be resorting to at this very, very early stage in the process?
Robert Wu [13:57]: Whether it’s a rushing just to drive up the GMV, because in terms of units, sold on Shihuituan it’s like less than half even with like 1/3 in terms of the sizes, in terms of units sold compared with a top player like Meituan, but in terms of GMV it’s already has reached about 60% of the top player. When you look at all these players, they really have very different strategies. Personally, I think right now the best choice may actually be having a good balance of fresh groceries branded goods and not to just resort to any one segment or category at this particular stage, a very early stage.
Rui Ma [14:35]: Okay, so now we’re done with all the data on community group buying, by the way that surprised me quite a bit especially on those best-selling products. What do you have prepared for us on ecommerce generally?
Robert Wu [14:48]: Ecommerce, the competitive landscape has totally changed from about two years ago. We are not only seeing the CGB, we are seeing Pinduoduo, we’re seeing live streaming ecommerce disrupting the fields. We are also starting to see OTO platforms, the insta-sales platforms like Meituan and Daojia also disrupting the game. This is very exciting time for the sector. We also did this analysis comparing all of these platforms because we started track all of them. We track Pinduoduo’s branded sales, just one footnote here Pinduoduo is really notorious to track just give all of us a sense of what are the sizes of top emerging ecommerce platforms. And here I specifically pay attention to Douyin and Kuaishou because these are the two lesser-known ecommerce platforms. If you look at the Douyin ecommerce here, and this is the monthly GMV since we ever are able to track this since last July. In terms of monthly GMV, there has been a tripling growth from July to January so half a year to generate around 6 billion USD of GMV for that month. This is happening just within less than two years of the start-up’s venture into ecommerce. Kuaishou ecommerce also has similar sizes and Kuaishou ecommerce only count for the ecommerce and take place on Kuaishou itself. We’re not counting the sales from Kuaishou to Taobao yet. If we take that into account question right now Kauishou could be larger than Douyin.
Rui Ma [16:18]: For those of you listening in and you can’t see the slides basically for the month of February BigOne is estimating that Douyin is actually up to 19% of T-Mall revenues and Kuaishou is at 13%. This is just one month, but we can see that there’s a pretty clear growth trend and we also know from other reports generally that these platforms are just doing really sizable GMV now. I guess this data from BigOne really confirms that. But Rob, tell us what are people selling on there?
Robert Wu [16:50]: People are curious what they’re selling these live streaming ecommerce platforms. We compare the category composition with the more established platforms like T-Mall and JD. A few things stand out really for Douyin is still not very balanced structure, the top three categories clothing, FMB, personal care and makeup, they still account for over 60% of the total sales. This is because these categories are the categories that are easier to be shown in video. But if you look at Kuaishou, we have an even less balanced structure, the top one category is actually jewelry and gifts. If you look closely at it, it’s actually mostly gold bars, emeralds, it has become a key platform for trading jewelry in China now and there’s a big profit margin there. Whether that’s a healthy sign for an established ecommerce platform. that’s the question. We also are able to compare the brand on the different platforms. This is a brand count and brand count composition of three platforms PDD, JD and T-Mall. We pay all attention to the PDD here, they only have about 4500 brands right now on PDD. They’re really at an early stage of branded sales. On our platform, we could take out all of these brands and compare their sales across different platforms. And you could see that for some of the brands especially skincare and sportswear brands, Douyin start to account for very meaningful portion of the total sales. For February, they account for 25%, Fila 19% and that’s pretty sizable. Douyin starts to be this kind of platform where all these brands, companies are coming in to market themselves. And then if you look at Pinduoduo is almost non-exist for most of the brands. When you look at MilkDiary (MengNiu) starts to account for a big portion of the total sales, we are seeing more and more of the consumer staples to be sold on this platform. In terms of the prices, if you look at all these brands, ASP wise, they’re still the lowest. While Douyin, in terms of prices, yhey’re ranking pretty high; they are comparable to T-Mall and JD. So live streaming ecommerce doesn’t necessarily mean lower price goods. But if you are selling the same kind of brand on Pinduoduo, they are definitely lower right now, sometimes the price is 1/3 of the price you sell on T-Mall and JD.
Rui Ma [19:09]: The ecommerce platforms still have pretty different brand mixes and category segments that they dominated. It seems to be pretty different for each one. That’s to be expected, I think, but it’s so really interesting to see the differences. However, ecommerce doesn’t actually contribute that much revenue yet for Douyin or Kuaishou. Right now, I mean, these are still primarily entertainment apps. Rob, I know you actually have one last point on Douyin versus Bilibili that’s really relevant to this. It’s about their user engagement, right?
Robert Wu [19:42]: We are tracking most of the short video KOLs ongoing, we are tracking all of the Bilibili’s and we will start to track Tiktok by the same methodology. But here we just take a snapshot of what we’ve been doing. So here we’re taking out Bilibili’s top 1000 content creators and we found that close to 700 of them also have accounts and have posted content on Douyin. The number of followers they have on Douyin is way more than what they have on Bilibili, if you compare not only the number of followers, but compare with, say metrics, like number of shares, number of comments, number of likes, you gather from different platforms, it’s also the same picture. We found one big thing regarding Bilibili, it has a kind of a natural ceiling for their content creators, at least for the top tier KOLs therefore, Douyin poses a significant threat to Billy’s growth. That’s our conclusion.
Rui Ma [20:43]: But I see here that there’s a difference between the mid-level KOLs, the ones with the 500k to 1 million fans or even the ones with 100k to 500k fans and the so-called head KOLs, those with over 1 million fans.
Robert Wu [21:01]: For the mid-level KOLs, the ones with less than 1 million followers, Bilibili is actually not doing bad, they’re actually higher than Douyin in this. It’s only when you get to the top ones, the over 1 million followers and that’s about a few 1000 of them, but you start having a meaningful gap between Douyin and Bilibili. So that also explains a strength that Bilibili has, at least for the mid-level KOLs, the level of engagement of users with the contents more than compensate for the fact that they have a smaller fan base, the smaller traffic compared with Douyin. At least in that segment, in that mid-level KOL segment, it really is still quite competitive.
Rui Ma [21:42]: Okay. And now we have a few questions from the audience,
Audience Member 1 [21:48]: Thanks very much, Robert. super interesting. I was really surprised that jewelry and gifts was so high in Kuaishou, do you think of the audience or the user experience on the app?
Robert Wu [21:58]: What I learned is actually for the jewlery, so there used to be a very opaque business, it is exceedingly high margin business. In the off-line market, the margin could be as much as 90%, or 80%, there’s actually a big room for the merchants to just apply some discounts to themselves and they just quickly reach a broader audience. Then somehow there start a culture on Kuaishou where a lot of these Jade and jewelry merchants they come in and sell their goods. I think there’s partly because there’s room for them to still make a profit because of the high margin before. And also the fact that, you can imagine the user on Kuaishou most of them are people from the counties or villages, lower tier cities, where actually you know, for them you especially during their spring festival during the holidays for them to buy a piece of jewelry is actually a big thing. If you go to this video shows, you will see that usually a lot of gambling involved for example, you just keep dialing some number, or keep screenshot your phone. And if you catch some random number, you have a big surprise, a big prize. They talk a lot of these marketing tricks and just jewelry in general, it’s just easier to be shown. And it’s a great game, this easier to build a game around it and it’s very interactive.
Audience Member 2 [23:19]: Yeah. Thanks for doing this. I just had a quick question about how your data, does it try to reconcile for differences in GMV between different ecommerce players such as like order brushing or filled orders for some of these players or excluding versus including returns.
Robert Wu [23:34]: Yeah, that’s a very relevant question. When we state a GMV, we usually state the net GMV. Net GMV means a settled-transactions, excluding the returned goods. The financial reports usually have a biggest, the widest definition of GMV, for all of the metrics we report here we do net GMV. And we try to compare our figure with the official GMV to estimate what we call the natural growth ratio. For most of the time, that ratio is pretty stable, like for example, for the GMV we tracking on Baba, we usually have about 65% net to gross ratio compared with what we’ve been reporting.
Rui Ma [24:13]: Yeah, I’m sure it’s actually very different. Right? I actually have a question myself. I’m super interested because you compare Douyin into Bilibili for the top 1000 creators, right? What does it look like when you compare them with Kuaishou? Have you tried to do that?
Robert Wu [24:27]: Yes, that’s something we’re working on right now. There are actually four platforms we pay attention to a lot. There’s Bilibili, Douyin, Kuaishou, and there’s another video platform called Xigua, Watermelon video, also under Bytedance. The first exercise we did is compare Bilibili and Xigua but we found that actually the two platforms there look similar, but the target groups are actually quite different therefore there’s very small overlap between the two platforms.
Rui Ma [24:57]: Okay, we have one last question from that. About WeChat channels, the new video product from within WeChat. So how do you see that competing with Douyin and Kuaishou?
Robert Wu [25:09]: Yes, it does. But it’s just has very different logics. It’s a very distributed platform, the way they structure it will be never the same thing as say, Douyin or Kuaishou where they have a pretty centralized way of directing their traffic. On WeChat it’s all these random content creators, the creator with smaller ambitions, they can still create content for their friends for their close associates, and they could still have meaningful interactions. So it’s not a place for you to have a super kill-out, definitely, that’s for sure. I think, in general, it’s not a standalone service, it’s really part of the WeChat ecosystem that helps people to stay on WeChat and in terms of future commercialization efforts. I don’t think it’s ever going to be a standalone platform.
Rui Ma [26:01]: So interesting. You say that I actually came to roughly the same conclusion, when I did the Tech Buzz China live cast with Dan Grover, he’s a former WeChat PM, and it’s actually our first live cast episode. You guys should really try to listen to it because we spent a whole hour trying to understand and predict everything to do with WeChat channels. But yeah, the conclusion is that we don’t think it’s meant to be a Douyin competitor. But we’re out of time. Thank you, Rob, for giving us this excellent presentation today. I learned a lot. And thanks, also to everyone else for tuning in. We can’t wait for the next quarterly update to see what BigOne Lab will have to bring to us in terms of data driven analyses from China tech. Thank you again.
Rui Ma [26:54]: But we’re out of time today. Thank you so much, Rob, for giving us this presentation today, that’s usually only available to institutional investors. To summarize, today, we learned or maybe confirmed for those of you who’ve been closely watching the space that the independent players and community group buying or CGB are still pretty competitive. And in the case of Xingsheng, they’re actually still leading because they were just earlier than the big platforms. And they’re really good at designing what it looks like sustainable platform economics, so they’re really keeping their group leaders and their user base intact. However, the market is still super early, we’re talking about less than 0.5% penetration for most of the country. So I’d say it’s still anyone’s game. And I myself am personally Team Meituan because I see this as you know, operationally, especially logistically intense.
With today’s data, we actually confirm that Meituan is showing greater warehouse efficiency than some of the other players so giving them an early lead. However, if you want a more in-depth look, especially at what slightly late comers, JD and Alibaba are doing in CGB. If you want a more in depth look at what players like JD and Alibaba giants and ecommerce at what they’re doing and they’re slightly late efforts. Please join our insider community, I cover plenty of that in there. The next thing we learned is that true to what we predicted before on Tech Buzz and actually on Twitter. This isn’t really about groceries so much as it is about rural ecommerce in China. And the next thing we learned is that true to what we predicted before on tech buzz and on Twitter. This actually isn’t really about groceries so much as it is about rural ecommerce. And we can see that in the data of these best-selling items on the platforms that Robert shared. Then we also did go into some detail on how the new ecommerce platforms are doing relative to T-Mall Douyin and Kuaishou mainly and we saw that both of them are actually doing double digits as a percentage of total T-Mall GMV in the last couple months and that actually seems to be rising. So that’s definitely something you want to be keeping an eye on if you are interested in ecommerce. Of course, there might be some seasonality and promotional campaigns involved. But no doubt these are going to be really, really tremendous disruptive forces that Alibaba and all the other ecommerce giants are going to have to contend with. We saw that Douyin is actually competing more with Bilibili than I personally realized anyways, at least on the top layer of KOLs. Although Bilibili actually remains very strong with the middle layer of creators, those who have less than 1 million fans. I think that actually makes a lot of sense though, given what we know about the company. And if you’re unfamiliar with it, please go listen to our Tech Buzz China Episode 57 on Bilibili I think that’s it everyone. Thanks for tuning in, and we can’t wait for the next quarterly update.
Thanks for listening guys. Don’t forget to rate us that review for the podcast and definitely subscribe to our new Tech Buzz China live cast. If you have any questions or suggestions, email us We really enjoyed putting this together as always, and we’re always open to any comments or suggestions. You can find us on Twitter at the Pandaily at Tech Buzz China and you can find me at R-U-I M-A Rui Ma or Ying at YingLu2020. Tech Buzz China by Pandaily is powered by the Sinica podcast network on SupChina. Pandaily.com is an English language site that tells you everything about China’s innovation. Our producers are Bryce Ye and Kaiser Kuo. Thank you for listening.