BingoBox’s Zilin Chen and The Future of ‘New Retail’ in China

Zara Zhang: Hi everyone. We wanted to get to know you our listeners better so that we can serve you better. Please visit our website to fill out a very brief questionnaire. By filling out this survey, you could be chosen to appear on the podcast as a guest host, and ask your own questions to our guests, as well as Hans and me, on an episode of the show. If you’re looking to join a startup, we can also help recommend opportunities to you based on your response. I look forward to hearing from you.

Also I wanted to tell you about one of my favorite podcasts, Acquired, a show about tech acquisitions and IPOs. On the show hosts David Rosenthal and Ben Gilbert go behind the scenes of these exits. Every episode is full of great history lessons and insights into some of the most important companies in the world. It’s my go-to show during my commutes and I highly recommend checking out if you’re into tech or business in general. On the latest episode, Acquired covered the Xiaomi IPO, and Hans and I appeared on the show as guests. You can check out the episode by visiting Acquired.fm or searching “Acquired” in any podcast app.

Hans Tung: Hi there. Welcome to the show, where we interview movers and shakers of China’s tech industry, as well as tech leaders who have a U.S.-China cross-border perspective. My name’s Hans Tung. I am the managing partner at GGV Capital, and have been working at startups and investing in them in both the U.S. and China for the past 20 years.

Zara Zhang: My name is Zara Zhang. I’m an investment analyst at GGV Capital and a former journalist.

On the show today we have Zilin Chen, founder and CEO of BingoBox, China’s first scalable, 24-hour, unmanned convenience store. It’s a small, lightweight, box-like shop where you can purchase pretty much everything that you may find in a 7-Eleven. But unlike traditional convenience stores, no human staff is required inside.

Hans Tung: To enter a BingoBox, you must scan a QR code to register on its website. Then to check out, you place the items on a counter, scan the QR code and then pay with WeChat Pay or AliPay which is a mobile payment. As you head out, a sensor at the door can automatically detect any unpaid items. The door will only unlock once all the items have been paid.

Zara Zhang: BingoBox is a pioneer in a phenomenon known as “new retail” in China, commonly understood as using technology to transform offline retail. BingoBox was launched to the public in August 2016 and now has over 300 boxes in almost 30 cities in China. GGV led BingoBox’s Series A investment in July 2017. The company raised another $80 million in Series B this January, which GGV also participated in.

Hans Tung: Hi Zilin, thank you and welcome to the show.

Zilin Chen: Hi Hans.

Hans Tung: Tell us about your background before you founded BingoBox. What motivated you to decide to solve this particular problem? Why do you think online convenience stores have a future in China, and how is that different from the traditional ones we know, like FamilyMart, 7-Eleven, and so forth?

Zilin Chen: Well, before I founded BingoBox, I was working in Ogilvy in advertising and I helped my clients to solve marketing problems. I helped them to sell products to consumers and diagnose channel problems. At that time, I knew that channels were very important to manufacturers. Early in 2013, I began to think about developing a new channel for products where clients could bring their products to consumers in a highly efficient way, at very little cost. At the end of 2013, I started BingoFresh. It’s an e-commerce platform selling fruit through WeChat, a very popular application in China.

Hans Tung: That’s right.

Zilin Chen: Yes. It took one year’s effort to build up our selling network in South China, before I started BingoFresh. It’s an online, e-commerce platform selling fruit online. We deliver our fruits to customers and we built up our own logistics teams. That business worked quite well and we raised our pre-A finance in August 2015. But after pre-A financing, I felt that the business was hard to grow quickly because we were not the leading company in China in that category.

At the same time, I was thinking about how I could lower the delivery cost for the e-commerce business. At that time, delivery cost was already low, but I wanted it to be extremely low, to make BingoFresh outstanding and competitive with other leading companies. I also wanted to raise another round of financing. So, at the time, my initial idea was to was to put a refrigerator inside the community. We could put refrigerator downstairs for our consumers. Then we would deliver our fruits to the refrigerators and they would go downstairs to pick up their fruits by themselves. This would lower my delivery costs. We did a pilot that was very successful.

At the same time, I thought “why not sell some things that were easy to stock with a long shelf life?” And we could sell some beverages like Coke, and some biscuits to get more return on investment and lower cost. Then I came up with an idea. I thought, since I want to sell products in the refrigerator, why don’t we just do it bigger and make a community store. That was the initial idea for BingoBox.

Hans Tung: How was that community store going to be different from a FamilyMart or 7-Eleven, which are located pretty close to these communities anyway? As we all know, in China, each community may have 20 to 50 towers. Each tower has maybe 20 to 50 floors, so there may be 10,000 people in a community. You can find FamilyMart or 7-Eleven outside of those communities and still be very close to them. What makes your solution more efficient?

Zilin Chen: Because you know we are closer to the consumer than 7-Eleven or FamilyMart.

Hans Tung: How so?

Zilin Chen: Because you don’t need to walk out of the residential area, your community–

Hans Tung: It’s smaller? So you can put it inside the community itself?

Zilin Chen: Yet that’s not the most important thing. The key thing is that you only can put a machine inside a community, and not a store.

Hans Tung: So you want to become a machine, and not have to pay any rent by being inside these communities. That’s the trick.

Zilin Chen: Yes, that’s the trick. But we did pay our rental. We just paid one-tenth of–

Hans Tung: –rental cost of a retail store.

Zilin Chen: So you can see, we have no staff cost, also we have an extremely low rental cost, and that makes our total cost much lower. I mean it is one-fifteenth of the traditional operating cost than the outside stores. So that’s very high economic efficiency.

Hans Tung: And the delivery costs will be cheaper than e-commerce, because you don’t deliver by order. You deliver in bunches to the machines, so you have synergy on the delivery cost as well.

Zilin Chen: So yes, here comes our mission. Our company’s mission is we are working to enhance society’s retail efficiency. All our efforts are to make this happen.

Hans Tung: So I’m going ask one more question before I let Zara jump in on the next one. You were an advertising executive before. This business model, BingoBox, has a novelty angle, to build something unmanned. How did you find a partner to work with, to come up with a technological solution?

Zilin Chen: I have to say I have some IT background, even though I’m not a technical guy. But I am a geek. When I was in university, I did coding myself, and I know most of the technical stuff. I know how it works. I find the solution has been in my mind for years. This kind of technology is very mature and I know the basics. So I just pick up some very creative guys in my team, in my technical development team, and I tell them the direction.

And so then we just work it out. I have to say the team is really very creative and they work very hard. And they’re very smart. So they found a lot of solutions. They talked to a lot of experts in this industry. And they used our “Internet mind” to combine all the solutions. I must say, the first version of BingoBox is only innovative at the application level. I don’t think the technical point of view is very high tech.

Hans Tung: But that’s the first version.

Zilin Chen: We created the–

Hans Tung: The MVP, the minimally viable product, first.

Zilin Chen: Yes. I can do that because I have a good team. We have survived running BingoFresh in the e-commerce platform. The team trusts me, and they trust this direction. I have to say the team is really good. I have a good team.

Zara Zhang: So obviously, Amazon also has its own version of the unmanned convenience store, called Amazon Go. They opened a store in Seattle, which opened to the public this January. And many people refer to BingoBox as the “Amazon Go of China”. So, I was wondering, when you started BingoBox, did you have Amazon Go in mind at all? And how do you think about this different version of what you’re doing?

Zilin Chen: If you notice that the launch time, when I did BingoBox, it was 2016. I remember very clearly that the video of Amazon Go was released in December 5, 2016. And at that time, our first BingoBox had already been operating for four months. So there’s no way I had any idea of Amazon Go when we were developing BingoBox.

But actually I think, in my point of view, we are very different from Amazon Go. From the very beginning, we were aiming to solve a different problem. With BingoBox, we are creating a new channel, a higher efficiency channel. We created a new shop and a new environment for consumers, a new way for you to buy products.

And Amazon Go is all about enhancing the efficiency of their old shopping environment. There is no shopping mall. And they are not an unmanned store. To be correct, it is a cashier-free store. They also have people to do the merchandising. From my understanding, Amazon Go is trying to solve that problem that in the United States, the cashier is very highly paid.

Hans Tung: And you have to wait a while to get through.

Zilin Chen: They have a pain point–that is the peak time and low time. They pay a lot of rental for that store, and when people come through at the peak time, they see a long line. And people give up because they have a lot of choices. They can drive to another store.

Offline stores are very developed in the U.S. and consumers have a lot of choices, so they can go to another store. They’re trying to solve this problem. If you can have no line for checkout, you can save a lot on rent. I think in that way, if the cashier-free model can work, and also the cost–when the cost can be low enough, or reasonable, then they’ll benefit a lot for the old, traditional channels. But we have different directions and the target is different.

Zara Zhang: How is a target different? Who do you target?

Zilin Chen: Actually, we are a convenience store and our target is all kinds of people. When you have a lease, the thing above you will give you an offer.

Zara Zhang: Yeah. So a seven-person team on the backend can watch 45 stores at the same time right? For BingoBox?

Zilin Chen: Yes, a seven-person team can maintain 40 or so stores. And the efficiency is getting higher now, and in the future, a seven-person team will maintain 50, 60, even 80 boxes. That’s our plan.

Zara Zhang: So could you walk us through the technology that you use to track items in your store? I think you started out with radio frequency identification, or RFID, which uses a tag that contains electronically stored information. But later on you also integrated computer vision. How did you make these particular choices and in this particular order? How much resources are you spending on R&D?

Zilin Chen: Well this order is not a choice (laughter). I think it’s a necessity. Because in the beginning we had no research power to do computer vision. It takes a lot of money for research. So we just felt that if we used RFID, as RFID is a cheap and mature technology to build up our unmanned community store system. I mean, in the total solution, product recognition is only the first step of the unmanned convenience store operating system.

First step, we need to recognize the product. Then we need to process the payment. And after that, in the back end we have a maintenance operational system to support this. As you can see, the result is a seven-person team that can maintain 45 BingoBoxes. So, again, the first step is product recognition, and we used RFID, a very easily-accessed technology, to do that. At a large scale, it is not cheap enough, because every tag has to be attached manually. You have to tack on the labor and tech cost to do that.

Definitely, computer vision is the future. So at the same time we use RFID to build our entire system and help us to prove our understanding of this business model. But at the same time, we invested lots of money, I mean millions of US dollars, to do the research. We had our team of over 100 research engineers to do the computer vision, and our product is called 小范FAN AI. We can use the computer vision to do the recognition of the products that we can read off our ID. So our operational cost is 30% lower. That means we are all ready for large scale recognition.

Zara Zhang: What is the accuracy rate of your computer vision solution?

Zilin Chen: Technically, the accuracy rate should be over 99%. Frankly, there’s no technology that can achieve 100% accuracy. 99% is already very reliable. Reliability is very important in the payment, it is more important than the time of a transaction. People don’t want to save one more minute, but they want the feeling of security. So when you pay all of our products from our “cashier”, 99% of the product list is correct. But out of the one percent, you need to move the product around to make the adjustment. After that, the checkout list should be correct.

Hans Tung: If it’s not accurate for whatever reason and the customer can’t correct it by moving the item around, what can they do? Can they call someone to help?

Zilin Chen: Yes they can call someone to help. We have 24/7 visual service staff. You can easily push the button on the cashier station, then you can see our customer-service face to face. We have a video conference system embedded in BingoBox. But that is unlikely to happen. When you move the product that you still doesn’t work–that is unlikely.

Zara Zhang: Another difficult thing about an unmanned system is how do you make sure people pay? So what is the shrinkage rate for BingoBox, or the percentage of goods that are stolen?

Zilin Chen: Actually we don’t put any effort to track the consumers or prevent consumers from stealing. Because when you walk inside BingoBox, there are very, very few people who steal, because you are under surveillance.

Hans Tung: It’s recorded.

Zilin Chen: Every movement is recorded. And also, our camera is triggered by our actions or behavior. When some unpermitted behavior happens, our alarm system goes off. The service team can immediately see what has happened inside BingoBox and they can take over.

Hans Tung: Lock the door? (laughter)

Zilin Chen: Well, they can lock the door, but we don’t recommend that. When they push a button, they can immediately talk to the person inside. So you can see the person on the screen.

Hans Tung: Oh, that’s not scary at all. (laughter)

Zilin Chen: These are some extreme situations, and we don’t suggest these kind of measures. But we have to be equipped for these measures. So you know, the shrinkage rate is very low.

Hans Tung: In almost two years of operations, how many of these incidences have occurred?

Zilin Chen: In almost two years operation, we have encountered three times.

Hans Tung: Only three times?

Zilin Chen: Only three times, and of those three times, they were theft by kids. And because they use their parents’ phones, we send an SMS to them. Most of their parents will pay online. And so it’s very clear. You know, people steal things. They initiate it because they have no money to buy all these things. I think their initial feeling is they can get away with it. But when they know it is all tracked– people generally behave themselves. People who walk in BingoBox, they use good behavior. We don’t even have to clean for a week. BingoBox is very clean.

People tend to think–especially in Tier 3 or Tier 4 cities, we have all these BingoBoxes in Tier 4 cities. In rural areas, people don’t damage our products. I remember an example of a mother took her 5-year-old boy inside our BingoBox. He kept throwing our products on the floor. And the mother, every time, would pick up all the products, put back, put back, put back, put back. After she could not bear it any more, she was trying to get out.

Hans Tung: She was trying to get out. This would never happen in a manned store. they wouldn’t care. Somebody will take care of it.

Zilin Chen: Yes, they know some people will take care of it. I think that’s very interesting.

Hans Tung: Very interesting behavior. So you guys have a seven-person team on the back end?

Zilin Chen: Seven people take care of 45 stores.

Hans Tung: Why that specific ratio?

Zilin Chen: This ratio keeps going up. Because now, our capability is greater. It’s a team and they have different functions. You have one person in the warehouse to do the preparations for each Box to do refills. Like 外卖 (waimai), you need to prepare. So one person to do the preparation, then one person to deliver the products to the stores and to the fulfillment. And you need a team manager for every area. One manager takes over for 100 Boxes, so you need that position. You also need the back end IT side.

Zilin Chen: So seven people can take care of 45 stores today, but in the future, they can handle 80 Boxes.

Zara Zhang: Can you talk about the unit economics? How much does it cost to set up a Box? How much does it make on a typical day? What’s the payback period? If you’re open to sharing.

Zilin Chen: Actually, the income is different in different cities. When the box is located in some high population density area, their sales will be relatively higher. Now, in our over 300 stores, our average is around RMB 800 RMB per day. In Beijing, we have around 15,000 RMB per day. So you can see, Beijing is higher. And Tier 1 cities are higher than Tier 2 cities. And for those Boxes located in Beijing, we can expect now the average payback time is around 10 months.

Zara Zhang: And you chose residential communities for your entry points. A lot of BingoBoxes are in these residential communities. Why did you choose that particular setting?

Zilin Chen: Because, you know, in the future I think when you locate a BingoBox outside a residential area, you can very easily get a location, because the price is transparent. It’s a market behavior. The price is open to everybody. But inside the residential areas, right now we are the only choice. We are doing the pricing. That’s why our rental is so low. The property management team doesn’t have a second choice. And I think now when we come into an area, we lock this business down. We are the only choice inside this area.

Hans Tung: Do you sign an exclusive?

Zilin Chen: We don’t sign exclusives. Partially we will sign exclusive, but now we don’t tend to sign them. Because it’s natural you will be the only one. Property managers don’t have the incentive to introduce new competitors inside the same area, inside the same community. When they introduce a second one, the property owners will not allow because we are doing a service for them. The property owners need one 24-hour convenience store. They don’t need many. And because we share profits with the property management, when they introduce a new one, their income will not go up.

Hans Tung: So what’s the point?

Zilin Chen: In the opposite way, they need to pay more costs to manage two systems, two suppliers. So they will not do that. The last thing is, our competitors will not compete with us inside one community. They’ll go to a new one. In China, you have over 400,000 communities. That’s a lot. This market is really big, it’s extremely big. So I think in the next three to five years, the competition will not be very strong.

Zara Zhang: Are the prices of the goods in BingoBox comparable to those in 7-Eleven? Are they the same or a little premium?

Zilin Chen: Of course, we are compatible. Our pricing strategy is 5% lower than the normal convenience stores.

Zara Zhang: Okay, so it’s cheaper and more convenient?

Zilin Chen: Yes.

Hans Tung: And you have a more efficient business model than they do. Obviously, speed is extremely important in China, you see that across the board in every internet sector. BingoBox already has around 300 stores in less than two years of operations. So how do you turn BingoBox into a fast moving company? What values do you instill in your employees, and do you see any other copycats come into this space and do something similar to you, at a fast speed, as well?

Zilin Chen: Well now actually I don’t think we are fast enough, even though we have 300 stores now. This year we have a more aggressive spending plan. So in the first two years, we were doing the foundation. We have developed our own technology or own AI systems and we trained the team.

I have to say that the BingoBox team is now the most experienced unmanned convenience store operators in the world. We have a lot of practice in this industry and now we are ready for a very fast expansion plan.

And then when it comes to the copycats, I don’t worry about any copycats from day one. Because I know that definitely you’ll have copycats in China when you are a good model. A good business in China will have copycats. I don’t try to stop anyone entering this field because the market is so big. There are 6.8 million mom-and-pop stores in China. The market is so big and this category is so new, and you need enough competitors competing in this field that you accelerate the market.

So, I embrace the competitors. And also, all the copycats, I notice now they don’t know the essence of this business. Some of them just want fast money. They just copy the shape of BingoBox but they don’t put any effort to build up the whole system or to do the backend job. So, I think competition is a good thing. It’s good for all the participants in this field.

Hans Tung: As you know, the top two internet giants in China, Alibaba and Tencent, are very aggressive and expanding to new retail. Alibaba tends to do it by themselves. Tencent tends to do it through strategic investments. What is your thought on working with them or competing against them?

Zilin Chen: My philosophy is to cooperate with all the good resources. To me, Alibaba and Tencent are also good resources. But at this moment, I don’t want to sell BingoBox to anyone. But I will welcome all the good resources of cooperation, like if Tencent wants to invest in me, I’m open. Also, if Alibaba want to invest me, I’m open too. But it’s just a matter of market price.

Hans Tung: So if anyone from Tencent or Alibaba investment group is listening to this podcast, you know Zilin is open for discussion. (laughter)

Zara Zhang: So right now for BingoBox, shoppers still need to place everything on a counter to checkout. Even though it doesn’t involve human cashiers, it still takes a bit of time, whereas for Amazon Go you just take everything and leave. So are you aiming for that day when you can realize that vision and how far are we from there?

Zilin Chen: So that’s an old topic about “take and go”. So back to the point of view I just mentioned before. When we’re talking about making a payment, what is more important? The reliability is more important than the time. In fact, now you use BingoBox cashier to make the payment, and it’s already very efficient. You just need 10 seconds to make a payment. I don’t think that’s a barrier. That’s acceptable to our consumers. But you “pay” 10 seconds more and you can get reliability or a feeling of security, I think that’s worth it. It’s worth it for the extra 10 seconds.

So by now, I think the consumers feel BingoBox experience is better than take-and-go experience. I think take-and-go may be more acceptable for the low price, for the light user, or one-time user with a small payment amount ($5-10). But when you’re making a bigger payment, say 100 RMB, you need to check the bill before you make the payment. But actually I know take-and-go may, in the future, will be in the mainstream. My policy is when consumers need it, I will offer it.

Zara Zhang: Have you seen a consumer demand for that yet?

Zilin Chen: Yeah, no I don’t think consumers have any demand for that. Most consumers hope to experience this, but when you provide it, they may choose not to use it. So when I launch this kind of take-and-go, I have to say, when the cost is low enough that it’s reasonable, then I will think about it. But by this time, by now, even Amazon Go, that system that cost $1 million cannot even do 100% accuracy. So, I think this technology is not mature enough now.

Zara Zhang: And there’s only one Amazon Go store, and BingoBox has 300.

Zilin Chen: So I think take-and-go technology is not realistic now.

Zara Zhang: So I think that also a huge difference between the China and the U.S. is the mobile payment is has such high penetration rate. So everyone has WeChat Pay or AliPay on their phones and BingoBox leverages those forms of payment. So you don’t need to do a lot of user education for them to know how to use the Box. Whereas in the U.S., there is no mass-adopted mobile payment methods. So it might take a lot more time for users to actually become used to paying with their mobile phones.

Zilin Chen: Actually, now I have no plan to locate BingoBox in North America. So I really don’t care. (laughter)

Zara Zhang: Yes, I was just commenting on the factthat this idea is probably more viable in China because of how widely adopted mobile payment is.

Zilin Chen: Yeah. That’s true. It’s obvious that it’s more viable in China. But you know, we are launching a new Box in Korea and also in Malaysia. Mobile payment is not as popular as China, so we also provide another way of payment like traditional credit card debt or some debit card. Also we also support this kind of payment. I have to say that’s not an essential barrier. It’s not a barrier. The payment is not the barrier, because that’s the need of consumers. It’s a very big shakeup for them to use BingoBox. When there’s a big snowstorm, you take a lift downstairs, you can buy all the things you need. That’s the big trick. It’s not about payment.

Zara Zhang: Also in the U.S., I notice all the newspaper articles about Amazon Go mention the loss of cashier jobs. Whereas the news in China, it is rarely mentioned. I think people in China generally have a more buoyant attitude towards AI and new technology. They focus on the positive impacts rather than negative ones. So why do you think that’s the case?

Zilin Chen: I think because Chinese people are now educated by Internet companies and we are now in a changing area, a changing era in China. Every day is different. But in the United States, the pace is not as fast as in China. Also, people and the society is very developed and they care about stability.

Zara Zhang: The status quo. They want to maintain the status quo.

Zilin Chen: They don’t like change.

Zara Zhang: Also, there are a lot of existing constituencies you need to persuade for any new changes to happen and that takes time. So you recently launched a new product called BingoBox Mini. Can you tell us more about what that is and how it works?

Zilin Chen: BingoBox Mini is an interesting product and I like it very much. It’s a really low cost visual recognition solution for traditional convenience stores. We just need 7,000 RMB for each BingoBox Mini. It’s plug-and-play. You put it in the store, and give us your SKU list. After three days, you can start to sell these products. The BingoBox Mini will recognize all your products in your stores. So you can save a cashier.

Zara Zhang: So it looks like an oven, right? The size is about the size of an oven. And you can basically re-outfit a traditional mom-and-pop store into a semi-BingoBox using that machine, and that might save a lot of labor costs. So that can be very powerful because as you mentioned, China has hundreds of thousands of these traditional mom-and-pop stores. So by doing that you extend your impact beyond your own BingoBoxes and into these more traditional existing stores.

Zilin Chen: Yes.

Zara Zhang: Yeah. So how many stores are using them?

Zilin Chen: We are now just doing a pilot. We just have some dozens of stores now doing pilots.

Zara Zhang: Ok. Cool. And then the 小范FAN AI is your initiative for enhancing the accuracy of computer vision recognition of your products in the stores. How is that going?

Zilin Chen: 小范FAN AI is not only computer vision. It’s a total solution including the computer vision and also the BI (business intelligence). We have empowered it as the visual capability and in the future, “she” will be equipped with communication skills like AI, so in the future, 小范FAN can talk to you. And she will know you better than yourself. Because it will be all your records. She will know your preference. In the future, I think she’ll make some recommendations for you and give you exclusive promotions. So小范FAN AI is not just a baby. We have a big plan for her.

Zara Zhang: Yep. So is she like Alexa, she can talk to you when you’re in the store?

Zilin Chen: Kind of.

Zara Zhang: I’m really excited for the future of BingoBox. It’s down to the final part of the interview which is a round of quick-fire questions. The first one is who is an entrepreneur that you admire the most and why?

Zilin Chen: I have to say it’s pretty cliché, but I like Steve Jobs. He is the entrepreneur I admired most. That’s because I think he was very real. He didn’t care about other’s judgment and he insisted on pursuing perfection. I think it’s very difficult when you run a company that you can insist on your pursuit. I think it’s very difficult.

Zara Zhang: Yeah. What motivates you to wake up each morning and what keeps you up at night?

Zilin Chen: Waking up, my motivation is the dream. One day, you can see BingoBoxes everywhere. I have a lot of imagination and when it becomes real, it makes me very happy.

Zara Zhang: So what is your dream and what’s her vision for retail in China? How many BingoBoxes is enough for you?

Zilin Chen: My dream is that–I don’t want to set a number, but I think at least that hundreds of thousands of BingoBoxes in China, and also in the world. Because I think it’s a very high efficiency channel. It’s good for society.

The dream wakes me up every morning and the work keeps me awake at night. It keeps me awake at night because there’s a lot of meetings and things.

Zara Zhang: So what do you do for fun?

Zilin Chen: I like sports and especially American billiards. Pool. And also, I like driving.

Zara Zhang: Driving cars?

Zilin Chen: I like driving. I like things that make me feel excited, especially off-road driving. I use off-road driving as a way to release pressure.

Hans Tung: Ok. As you expand across China you have deal with the local governments and how does that work? Do you need a permit or any kind of permission from them in order to do this business?

Zilin Chen: Yeah, we definitely need to deal with the government. And now working with government is one of the most important things for us. Because BingoBox is a new thing, the government doesn’t know how to measure it. We are now working closely with the government to have a permit. Now our practice is to sign a cooperation agreement with the government that will officially allow us to run these kind of community stores inside these communities. So we are now also working with the government to issue the standards.

Zara Zhang: So you are participating in writing the rules.

Zilin Chen: Yes, we are now actually participating in this.

Hans Tung: Are these rules local government, one by one? Or is it at the provincial level, or the central level?

Zilin Chen: Now we can only do it one by one.

Hans Tung: City by city.

Zilin Chen: Yes, city by city. We have a permit for over 14 cities in China.

Hans Tung: Yes. And then how many VCs did you meet for Series A, the round that GGV led?

Zilin Chen: A lot of VCs.

Hans Tung: What made you pick Eric (Xu)?

Zilin Chen: Because he knows this business. And I think just chemistry. Day one, he met me and he was very excited. And when we got downstairs, he texted me on WeChat, he told me–

Zara Zhang: –that was the most exciting business he’d seen in a long time.

Zilin Chen: Yes. That’s what he said to me. So founders sometimes need affirmation from VCs and encouragement. Eric gave me a lot of it. (laughter)

Zara Zhang: Thank you so much for being here. And we look forward to working with BingoBox going forward.

Zilin Chen: Ok. Thank you. Thank you.

Zara Zhang: Now you will hear a short interview with Eric Xu, managing partner GGV Capital in the China office, who led our investment in BingoBox.

Hans Tung: So Eric you have been at GGV since what, early 2017?

Eric Xu: Yes.

Hans Tung: And you have done a lot of very interesting new retail investments. BingoBox is one of them. What’s the rationale for you to spend time on new retail? And what particularly interested you about BingoBox?

Eric Xu: Yes. So before I talk about BingoBox, I want you to know the pain points for the traditional e-commerce market and also the pain points for offline retail. So the pain point for the traditional e-commerce is that first, you have to acquire a lot of users for your e-commerce business. And user acquisition costs have surged these days.

Hans Tung: This is online, right?

Eric Xu: Yes online. So five years ago, the cost of acquiring a user was like 10 to 20 RMB, and now it’s 200 to 300, even 2,000 to 3,000.

Hans Tung: Wow. Was that because the smartphone penetration has gone up and is not growing as much?

Eric Xu: Yes. So actually in short, there’s no traffic dividends for today for smartphone. And the user is just bored with installing every application with the iOS store. So you may want to install one or two monthly, but you don’t have to install all of them. In other words, the consumer has already gotten used to several applications.

Hans Tung: They have their habits already.

Eric Xu: The second pain point is that the fulfillment cost of e-commerce. So even if the user needs to purchase one bottle of water, you still have to deliver to the home, so it will cost you almost the same as the price of the goods.

Hans Tung: This is 101. Every transaction, no matter how small, has its own cost.

Eric Xu: Under the very competitive situation in China, you have to do this regardless of the price or cost. Yes. OK so two major pain points with the traditional retail–

Hans Tung: So, back to the new retail, if you ship the goods to the retail point, that’s consolidation so you have more economy of scale per order, because each order is sizeable?

Eric Xu: Right. You just shipped out the total number of goods for the day and the user just takes away for himself.

Yeah that’s different from the traditional. It’s also the same. So like number one pain point is that the rental cost is very high. So as we all know, if we want to open one retail store in a first tier city in China like Shanghai, Beijing and Guangzhou, the average daily breakeven point is as high as 8,000 RMB, and this 8,000 is the lowest. So normally, we see breakeven point as high as 10,000 or 13,000 RMB per day. So with the new retail, you can virtually pay very low cost. Sometimes you put the Boxes into the middle of nowhere and you are charged nothing.

So the second one for the traditional retail is the staffing costs. So even if this is the minor portion of the total cost structure, you still have to pay attention and you have to employ like three or five people, with at least two persons at a time in the store.

Hans Tung: So there’s training costs, training time and also leakage.

Eric Xu: So there’s some catch-up costs for the staffing and you have to pay more attention to the staffing. And if they have to pay attention and stick to the general principle you set for the retail. So these are those major pain points that are true for e-commerce and true for traditional retail. But new retail solves all of that.

So for the first one, you don’t have to pay to acquire users, because the Boxes are just nearby. When the consumers just walk around, they find out if there’s a need, they just walk in and purchase it. Second one, there’s no fulfillment cost, because like I said, the consumer picks it, pays and takes it away. And maybe you can consume it immediately, like a Coca-Cola. And there is that very low rental cost, and that is the biggest part of cost inside the cost structure.

Hans Tung: There’s no rental cost because locations inside those big communities have no rental costs involved right now?

Eric Xu: Right, that’s true, and of course there’s no staffing cost. You’ll have to check inventory regularly, once or twice a day. But you don’t have to have some staff standing there to provide immediate service to the consumer.

Zara Zhang: So there are a lot of companies in the new retail space. It’s been a buzzword in China in the past two years. A lot of companies in the unmanned space especially, so what about BingoBox stood out to you?

Eric Xu: Yes. First of all is the team. When we invest early, we consider the founder and the team as the most valuable asset we’re looking for. So I spent a lot of time with the founder before we made the final decision. Zilin actually is very good. He can prove everything he says to you and get it get them finalized. This is very important.

The second one is that they have a mature solution. Why I call it a “mature solution” is when we come to the business world, we have to provide a solution to be stable. So a lot of guys come to me saying “Eric, I can provide an image processing, image recognition solution to you. This is the final version for new retail.”

But where we tested them, we found the accuracy changed from maybe 50% to 97%. Even with a 97% accuracy rate, it’s still far from being stable with a normal operation in the real world. So RFID solution actually is not so advanced compared with image recognition one. But as the first generation, we want to roll out the business quickly and we want the business to be stable.

So when we talked about the future, Zilin also replied that he is doing it with the second generation, which is also the most advanced image processing solutions. So this is the difference, the major difference between BingoBox and the other players in this sector.

Zara Zhang: What is your definition of “new retail”? We’ve talked a lot about this term but I think different people have attached different meanings to it. I think it was first proposed by Jack Ma a few years back.

Eric Xu: Yes, I think the first important definition for new retail is it has to be smart. So a lot players or investors, they’re investing in some new deals calling it new retail, but they do not use a smart terminal. So without smart, you don’t know who are your users and what they are buying, and what are they fond of. So they are not a trend of the new retail.

Second one is that with the new retail, you have to solve the full set of pain points, while some are for the traditional and some are for the e-commerce. So when you can solve all the pain points of all the problems and the using the smart solution and know the users and have a database to be used use in the future with a better possible solution, business model, then that will be ideal.

Hans Tung: Makes sense.

Zara Zhang: How has BingoBox and Zilin and the team grown since you first met them?

Eric Xu: Two levels: the team level has changed dramatically. So I think they have a lot of good people added to the team, not only on the operational level, but also like high level HR and also strategic, and most importantly is R&D. They recently recruited the head of the R&D to Intel in China to do R&D within image processing solution. The second one is business wise.

Hans Tung: So that’s interesting. Now Chinese new retail investments can take the best talent from IT world who are not doing enough smart things or Internet things. But the technology know-how is actually there. So it’s a great marriage of resources and opportunity.

Eric Xu: Right. So this is very important. And the second way is business operations. So as you guys know, they are the first runner in this space and that they have certain dividends not only for the business but also for the government. So they have developed exclusive collaboration with more than ten city governments, hopefully to run the business gradually. And when we first invested in the company, the company only operated five Boxes over two cities. And now the company runs more than 300 Boxes over 10 major cities and possibly they will expand this to 2,000 by the end of this year.

Zara Zhang: Zilin told us that you texted him right after a meeting that BingoBox is the most exciting deal you’ve seen in a long time. So we look forward to seeing the company grow and our partnership.

Eric Xu: Yes thank you.

Hans Tung: Thank you Eric. That was very informative.

Eric Xu: Thank you very much.

Hans Tung: Thanks for listening to this episode.

Zara Zhang: GGV Capital is a multi-stage venture capital firm based in Silicon Valley, Shanghai and Beijing. We have been partnering with leading technology entrepreneurs for the past 18 years, from seed to pre-IPO. With $3.8 billion in capital under management across eight funds, GGV invests in globally minded entrepreneurs in consumer new retail, social Internet, enterprise cloud and frontier tech. GGV has invested in over 290 companies with more than 45 companies valued at over $1 billion.

Zara Zhang: Portfolio companies include Airbnb, Alibaba, Ctrip, Didi Chuxing, DOMO, HashiCorp, Hellobike, Houzz, Keep, Slack, Square, Toutiao, Wish, Xiaohongshu, YY and others. Find out more at ggvc.com.

We also highly recommend joining our listeners WeChat group and Slack channel where we regularly share insights, events, and job opportunities related to tech in China. Join these groups at https://nextbn.ggvc.com/engage/

Hans Tung: If you have any feedback on this podcast or would like to recommend a guest, please reach out at https://nextbn.ggvc.com/engage/