Y Combinator Startup Podcast - #48 - The Future of Bike Sharing with Ofo Cofounder Yanqi Zhang and Anu Hariharan

Episode Date: November 17, 2017

Yanqi Zhang is a cofounder and COO of Ofo.Anu Hariharan is a Partner at YC.

Hey, this is Craig Cannon and you're listening to Y Combinators podcast. Today's episode is with Yanqi Zhang and Anu Hari-Hran. Jan Chi is the co-founder and COOO of OFO and Anu's a partner here at YC. So before we get going, if you haven't yet reviewed the show wherever you get your podcast, it would be awesome if you did. All right, here we go. Hi, I'm Anu Hari-Hurin, a partner at YC. And thank you, Yanki, for being here today with us for our global growth event. for doing this interview with us.

So Yankee is the co-founder and CEO of OFO. Can you describe what OFO is for our audience? Absolutely. So, first of all, Iguan, thanks very much for having us. I'm very excited to be here to share our story. So O'Fo is the first smart station-free bike sharing program in the world. So what we do is that we provide the station-free bikes and for people to use to unlock with their smartphones.

It's very easy accessible. It's very green, and it's a very smart way of getting around the city, especially for the last mile. Great. Can you talk about the scale as well? Like how many cities are you in? How many trips do you cover? Yeah, absolutely. So as of today, we are in 180 cities.

We're doing over 25 million trips a day. We have deployed over 10 million bikes already. And we're in right now globally 13 countries. The goal here is that by the end of this year, we're going to go to that. at least 20 countries. Oh, that's phenomenal. I mean, it's truly unique for a Chinese company to really go international this fast. Before I touch into how you did the international expansion, I wanted to hear your story on

year one. What was the motivation to launch OFOFO? And how did you guys go about doing that? Basically, the company was founded by a group of passionate cyclists. So it was founded in university when we were in university. So the very natural thing is that you have your bike in the university. But I think probably like this is something everybody has experience is that your bike been stolen. A lot in San Francisco, actually.

Unfortunately. Yeah. So found himself like he lost four bikes during his university period of time. So yeah, so we come up with this idea of what if we have a shared bike system? Like people don't need to worry about like bike being stolen, be like when you go out of a station, right, they're still like 1.5 miles to go, but you don't have any transportation methods. You have to walk or something like that. What if we can solve that problem?

So that's where, like, the idea came about. I think this kind of last mile commuting problem is, we feel that in China, but right now we know, like, it's actually universal. Like, it's everywhere, right? There is not a, before us, there's not a, like, very good solution for that. So, yeah, so that's where we came from. And it's very natural, like any other startup in the world, like, we came up from, like, wanting to solve, like, very simple pain point of people's regular daily life, right? So, yeah, so we started in 2014 and in university, right? Which university was it?

At Beijing University. Okay. Yeah. So basically that's where the founding team studied. So basically, like, we started from there, and we spend some time to expand to 200 universities at the beginning. And starting from 2016, end of 2016, we expanded to the cities. So in the city markets, because we have the experience of running operations and all the systems, like within the university, so in the city, it's a different scenario, but there are a lot of things we can borrow.

So we scale really quickly. We spend eight months to expand to over 170 and 80 cities. In China also include some international cities. right now mainly in China. And the trips grew from at the beginning of it was 200,000 trips last year. Right now is over 25 million. So there's a big growth there. Yeah.

So that's sort of like our story and our trajectory. Yeah. So I wanted to go back a little bit on the first year when you launched in the university. How did you launch? Did you make your own bikes because you obviously wanted to move from doctor stations to something that people could access anywhere? So how did you – what was the supply in the first year? Yeah. So it's very interesting because at the beginning, the idea was to use like, we call it 100% sharing model.

It's like we use the existing bike to sort of like as a supply source. So in China, well, in China is called the kingdom bicycles. We have a lot of bicycles. So when we started the business, we look at the statistics, the annual sales of bike and an existing stock of bike. Actually, in China we have 400 million bikes. So it's very natural thoughts. Like what if we activate that existing supply? Just like Uber or Diddy for my chat. Yes, it's like a huge amount out there. So that's what we did at the beginning.

So basically what we do is that we go out there and telling people through different channels, like we're doing this shared bike program, and everybody's welcome to donate their bike to our system. We start something like we call Exchange 1 for N, which means like you donate your bike, and then you have access to all the awful bikes in the system. So we thought that people would love it, and a lot of bike would be donated.

But what actually happened is that it's a very new idea, first of all. And second is that not actually many people are willing to give up their personal positions, like then like share, right? At the beginning, it's hard for people to quite understand the idea. So at the beginning, the supply growth is actually quite slow. Like every day, I think we collect 30, 40 bikes. That's a number. At the same time, like the demand actually,

grew very fast. Like when we put the first batch of bike on the street, in the university, people just love it. Explosive. Like people talk about it, different, like, on forums, everything, like people love it. The only thing is that people complain is they cannot find the bikes. The people, they saw people, students are riding bikes, but they cannot get one for them because there's not enough bikes. So that was the moment, really, we started thinking the supply growth is very bottom-knock for us. So how do we solve this? So I think back in 2015, we came up with this idea of going to the largest manufacturers in China, which is in Tianjin, as a city very close to Beijing.

That's where most of the largest bike manufacturers are in China and in the world. So we go talk to them. Basically, we start to expand the idea of sharing, bike sharing. So it's not just about sharing the existing stocks, but also about bringing the new bikes that the manufacturers tend to sell, plan to sell to the individual buyers to our platform to be shared by people. So on average, our bikes are 20 times more efficient than the private bike. On our platform, on average, one bike serve over 20 users compared to like one private bike

serve one people, right? So yeah, so we came up with this idea and then we approached them and then we started to have a small trial at the beginning because we do not have much funding in the beginning. So we buy some bikes and then we put in the university and while. Also, like, when you have your own suppliers, it's very easy to control the qualities and also, like, the colors and the bike itself. Right. So we have 100% of control of your supply. And that's when we call the small yellow bike was born.

So we have that. Back then, it's like we have the bikes and we paint the bike, but it's like it's not very good. So when we have our own supplies. So you painted the existing bikes. Yeah. So back then. Which you were sourcing from others. So we've been paint the bike.

Sometimes we don't paint. We just use the bike and add our lock. So when you walk in university, you see our lock, that means that's an awful bike. So it's not very easy to identify sometimes. And also, like, there are different bike types. It does not look very identical. So, yeah, so basically, like, the new model works better. So that's the solution we find to scale our supply, and it works out pretty well.

And very quickly, we expand to, like, 10 universities, 50 universities, and we'll spend a year to expand to 200 universities, which is quite a lot in China. 100% market share, profitable, and people love it. Students love it. And so, yeah, so that's how our story begins. So you expanded to 200 universities by the end of first year. Yeah, but then so basically we started, well, like, graduate pick it up, and we started in 2015 of this, like, new model of purchasing the bikes and expand.

And we spent a year from 2015 to 2016, and we focused in. expand to different universities because universities like it's a closed area. In China, most of universities have like the walls. So it's sort of like a natural closed area where like you don't need to worry about like a bike be taken out of the university because we hire people are stuff like at the gate, like people cannot go out, right? The same bike. Got it.

So it's very easy. And also like it's much more, much. simpler environment in terms of the user behavior. Like students, it's like where they live, where they go, where the libraries, refactories, like the gyms, the football fields, it's dorms. Like it's very, very easy to sort of like do the rebalancing. It's very simple, I would say, the scenario. So I think that one year actually give us a lot of insights, a lot of like a practice about

how to organize the operation. Even back then we do not have a very smart system. We use Excel sheets. We use like sometimes just write down at different universities. But that gives us like a very good experience on how to manage a ground team. So like how many people we need, how many bikes, and how to maintain the bikes and what bikes we need. Because people like, we'll see like bike been broken and like how do we do rebalancing? So how much time in advance do we need in what scenarios?

So actually we have a lot of experience and being playbooked into. that, into our operation model in that year, which where we can borrow for the city expansion. So I think that's one of the most important reason why we can expand so fast, scale so fast in the city. Because when you figure something out in a smaller scale, I think it's sort of like a natural like a business development. Yeah. Sort of like a like a like a trajectory.

It's like when you figure something out in a small environment and then you gradually take it out. And also at the beginning we do city expansion, we do not just go to a lot of cities. We go to districts in Beijing and Shanghai. And we try to sort of have a managed expansion and have some experience, okay, this is how it feels like in the city and how people use it. This is like how should we do rebalancing? Because university is very easy.

We use certain type of vehicles to do the shifting and rebalancing stuff. But in a city, different cities has different regulations. So for example, in Beijing, we can use. this three-wheel electric vehicle to sort of like to move the bikes, so it's very efficient. But in this lot of southern cities in China, like Shen Zhen and Guangzhou, the cities, these vehicles are banned. So we have to find different vehicles, sort of like sometimes we need to modify some of the small vans to do the rebalancing.

So yeah, all this kind of like challenges specifically in the city we need to tackle. It's an interesting observation here. It's even though you're very different from a social network, I think. Facebook often talks about that. Like the learnings for them was the fact that they did a university and university rollout. First, they figured out Harvard, then Stanford. And then, you know, before they opened it up to the, you know, basically to everyone, they had 80% penetration in universities in the U.S.

And they talked about how they figured how to build the products such that there was high engagement in universities. So it's a very interesting parallel. And so I wanted to ask you when you were in 200 universities, how did you decide when to switch to the city and what motivated you to switch to the city? And how did you think about, you know, because as you said, for all the elements why the university is so different from city, how did you tackle that first city that you went to? Yeah. I think there have been discussions of going to the city very early, right? So probably like from day one. I think there are a lot of,

we think there are a lot of conditions of going to the city, right? So we need a lot more bikes in the, then in the university. And the bike itself need to be upgraded because in the university, we use simple bikes, regular bikes. In the city, probably we need to use bikes with gears, bikes with much more, like, a lot of, like, anti-theft features being implemented, which will increase the cost and everything. So, yeah, so basically we spend a lot of time, first of all, researching, developing the bikes that we need for the city and do test in a very small area.

So, yeah, so basically, like, we developed the bike for the city, and we expand operationally to the university. So that's what we do in the first year. And at a certain point, we think that, first of all, we're ready on the product side, and also, like, we're ready on the operation side. And third is also on the market, right? So we think that in China, I think there are 2,000 universities, but the best ones are like the top 10%.

So we're quite, we have a quite good coverage of the good quality universities, like which has the best, I would say, the campus construction, and also, like, the university students' quality and everything. So, yeah, so basically you feel like in U.S., you have all the Ivy League or the UC system, right? So when you have all that, so we think that, okay, so the next stage probably like we can go to the city. And also, like, in terms of like the team, right? At the beginning, the company was created by, founded by like a student project, but later, like we upgrade the team, up with the team. And a certain point, like we have like the CD management, CD management team set up.

We think that's the time because when we go to the city, we want to make sure that that is a win. And we have, so this is sort of like a bet we take. So we need to make sure we have the necessary resources, the necessary team, and product ready before we go to the city. So that was the moment, I think last year, I think November, October and November, was the moment like we think, okay, this, we're ready, right?

We start expecting. Got it. Can you also talk about the playbook for each city? So when you're desiring to launch in a new city, What's the first step? Is it hiding a city GM? Is it putting up a wait list? Like, what is it? Yeah. So how do we launch cities, both domestically and internationally? Is that, first of all, we would do a research. We would do like a city list digging, right? Into like which cities are the target cities. So we look at a lot of metrics. Like, I think there are over like 20 of them, like including the population density, like the credit card penetration.

smartphone ownership, how the city's plan, and also like the price for the public transportation, for taxi, how much money, how much pens and money, the household average spend on every month on transportation and stuff, all this like metrics, et cetera, et cetera. Like we come up, there's a formula, so we come up with a city list on market size, right? And also, like, what's the history of the local 30s, like about like the new technologies, new companies and like all these, right? We quantify it and then we have like a list of cities out there. in China and internationally.

So we find the cities. So that's sort of like we call Air Force approach. Like you look at it. And then we look at the Marine approach. So we send people on the ground. Sort of like the launchers or basically the launchers. Sometimes we send operations managers as a launchers, but some people on the ground. How many people do you send?

In China, actually, because we're already in 200 universities to start with, So we have teams in 200, not 200 cities, but like, I think, 50 cities. So like we have like very good resources. Like we just have some people like to go to the city to do some tasks and stuff. So it's easy to start. But for international markets, we do. We need to hire launchers and send them. So but the hiring process start like a few months before we, we, the launch date of their national business.

So yeah, we hire the people. Like most of them like Chinese at the beginning, they have experience of living international market, international countries, or like doing business there and stuff. So we send the first batch of launchers landing team on the ground. Is it like five or ten people? Something like that, yeah. I think it's like 10 plus. 10 plus.

So, like one people to one country, something like that. And then they go there and they launch the, do the basic setups. And well, first of all, what they need to do is that actually the market research, the greenlight process. So they go there spend like a week or two. actually like to cross-check with what the Air Force has found, right? So all those metrics, what it's really like and how people think about and everything and talk to people, actually try out the local public transportation and feel about experience

and everything. So, yeah, so the green light process will take one week or so, one or two weeks, and then we have them start to establish the basic foundations of the company. So we will start to initiate contact with the local 30s. We start to the recruiting process because, well, like I said, we need a local team around the business like in the long term. So they start to recruit the local people. What's the team composition initially for a local team? Like what are you looking for?

Is a GM with few support? Like, what's the structure? Yeah. So the ideal structure would be like a GM, city GM, and with operation because we're like quite heavy on operation. So maybe one or two operations. One in terms of managing the maintenance stuff, one is managing the bike. So it's like operation. And also like on the marketing side, we also need people do the user acquisition, the PR, like the things.

But in terms of like hiring orders, so it's the CDGM and the operation team and the marketing team. But sometimes to find the city GMs is slower, right? So it's not very fast. So we'll probably start with the operation team side. So yeah, every city is different. So most of the city are run by launcher. So launcher is everybody, right? So launcher is doing everything.

So it's like that. And it's very similar to most of the expansion model by other internal companies, I would say. Only one difference is that we are operation heavy. So basically, like, we'll put more, I would say, the attention on the operation side. So that's how we do. So, yeah, so basically that's the basic structure. Can you talk a little bit about the operations? Because it's a little different from ride-sharing.

Right-sharing, you have someone who's driving their own car, and you're requesting via the app and it helps you, take point-a-point-B. At OFO, you have your supply of bikes, and people can drop it anywhere and pick it up anywhere. So can you talk about what is the operations entail? Yeah. So the operations, there is different categories. So the most important ones are the bike deployment.

the bike we're balancing, the bike maintenance. So these are the key things. So at the beginning, obviously, like, we have the bike in the city, and basically we ship bike from China or some factories in Europe and in different parts of the world. So when we have the bike, when the bike enter the city, it's the operation team's responsibility to maximize the efficiency of the bike. So basically, the deployment, right?

So where to put the bike to begin with, right? For example, like in small cities, it's easier, right? So it's just like put it in the city, like the transportation hub. But for big cities, like London, like maybe like, well, like, like, like, like, like, like, like, like, like, like, like, like, uh, like, uh, like, uh, like, uh, like, uh, San Francisco. Like, well, we're not in San Francisco yet, but if we, if we do, probably like, we start with one district, right, instead of, like, going to the whole city. So, so, so, yeah, so basically, like, we will find, like, the most back-front, either the most back-friendly or the, like, the most dense city in terms of population.

There, there are some criterions you can use. to find those locations and then you start to deploy bikes. And also, there's another criterion that other thing you can look at is like, where are the areas in the city where people open our apps most frequently, like before we're in the city. So we see that heat map. Okay, this is area actually. Maybe it's a student area, like a lot of Chinese students.

Right. So that's how you found out about Singapore, right? Yes. So in Singapore is that we see that heat map. We say like, okay, when we try to decide which is the first national city to go, We see like the global map where like we have the most usage, most user activity before we're in the city. So we see Singapore obviously is the most popular one and we go there and, yeah, so the business took off very quickly. Yeah, so that's, and there must be some areas in the city like has that factor, but that's only one thing to consider.

So there are different things. And also like in some places like we talk to the government and there's some areas like the local authorities need us. to provide assistance. For example, in Austria, like we have very good relationship with local authorities. Like, we started with District 2. And the second district we go is the District of 21, which is, I would say, in that city is kind of like far away from the center and less like mobility options for the citizens. And if we go there, it's actually add up to the, add up to the mobility like people can have there. So yeah, so this type of things are all into our considerations. So basically our principle is that we go to the city, we want to help

the city solve their mobility problems, we want to listen to the city what they need, and we provide our solution. So that's our way to work with the cities. So yeah, so I think whenever we go, we make sure like we have that conversation. We'd be transparent. We're transparent with the local people and local authorities on what we are and what we can do. to make sure like it's a win-win situation. Yeah, and you're present in the U.S. now, right? Yes, so we launched Seattle. Okay.

And I think we launched two cities in East Coast, in the greater Boston area as well, and more cities to come. Great. So what do you think, you know, many people are skeptical about the U.S. right, for bike sharing? And I know you've often told me

that even in China, you changed the behavior of the people to use bike sharing. So what's your view of the U.S.? Do you think that there's enough demand for the bikes and do you think there'll be a behavioral change? Yeah. So I think we can sort of like take one step back and ask the question of like what a city needs, right?

No matter it's a Chinese city, U.S. city, like like what city needs? What are people living the city need? Right. So I think to get around the city, right, get around the city, you know, smarter, greener, like more affordable and like more accessible way, I think it's always better. So if there is something like you can use, like match all these criteria on above, people must use it. So that's our basic assumption. So if that assumption stands, we would say like there are actually not quite big difference between city to city in the world in terms of like what city needs.

So that's that's where we start to think. So secondly, obviously like there are different characteristics of different cities. Some cities are more dense. Some cities are more sort of like, I'd say the hilly than the others. Some cities are more like the transportation, public transportation development stage is different. People like in different cities, people are willing to spend different money on transportation. For example, in Brazil, like people, the average household spending on the transportation is 30% of their income. So that's super high.

But maybe in U.S. is much lower. and also like what people's like habit, like like existing behavior. For example, in Paris, the percentage of transportation by bike is only like single digit, like three to five percent. But in Amsterdam is 30 percent. It's 10 times higher. So yeah, every city is different. Different.

But the thing is that, first of all, the last mile commuting problem is universal. Number two is that I think for most of the cities, people need, I think, like, well, most cities people don't have sort of like a very convenient, like a smarter and a cheap way of getting around the last mile. So, so that's why we are and where we kick in. So I think U.S. cities also have that, have that sort of problems or like characteristics. Obviously, I think in San Francisco, in Los Angeles, in New York, because, like, the cities are very different. But I think we're very confident, like, our solutions will be, something like people have never experienced before.

And we are here to solve the problems. Probably we need to make some product upgrades, like our adjustments. Probably we need to make some operation adjustments. But in terms of making adjustment to adapt to local markets, I wouldn't say, like, a U.S. market is very different from, let's say, Japan or France or UK, right? So we make adjustment all the time. That's what we do. Got it.

Great. Yeah. A few adjustments are obviously needed to cater to the local markets. Let's talk about electric bikes because I know you've briefly mentioned that. So what's the vision for OFA going forward? Yeah. So if you look at the trip distance right now, first of all, like in the city, right? So it's actually most of the distance are short.

It's a short trips in the city. And you can see like what we do as a bike sharing company, regular bike and mechanic bike, right? So what we do is that normally the trip distance ranges from 300 meters to three kilometers. That's where the distance bike can help. So that is like for every city. That's good. So obviously I think the natural next step is that we want to extend on both ends. Well, on the short ends, we see that happening already.

People are getting lazier. So it's like before that people are willing to walk 300 meters. But right now with bikes, even like 100 meter, I tend to ride. And on the, probably we don't need to much things on there. just improving the writing experience. On the far end, probably, like, there are a lot of things to do. Right, so with mechanic bikes, we are constantly improving our riding experience. Right now, we see a very steady, like, like, a growth of the trip distance.

Like, from 3 kilometers, right now is close to 4 kilometers. So that's like, like, people tend to ride more, like ride longer when it's easier to ride. But I think with electric bike, that far end of the distance can be extended significantly, probably to 3 kilometers to like 10 kilometers. So that can actually cover most of the trips that are done in the city. So basically, just imagine, like, in the city people ride electric bikes, and there will be a lot less cars in the city, right? The pollutions, the congestions, the accidents, like all those things, will be massively

much better for the environment. Yeah, you'll see, like, people, like, the city is getting better. The people are also like in terms of price, right? You can take cars and it's expensive, right? So owning a car is expensive. So yeah, so like if you think about how people get from point A to point B in the city, regardless of what like vehicles we use, I always think about like what is the most economic way, safe way, green way, right, easy way to get around,

to complete that distance. So we're trying to find like a more advanced vehicles. to satisfy that demand. So I think e-bike is a very natural option of that. And I saw, like, there's their companies here, like a school, like similar things to do that. So, yeah, so we're very optimistic. And actually, we have e-bikes in China been tested in different cities, and we see very promising results. On that, obviously, there are different economic metrics.

There are different, like, models, right? So how to change the batteries, how to sort of, like, charge the batteries. and how to do the operations will be different. But I think e-bikes is one of the things we are going to sort of like put efforts on in the future. Got it. Great. And how big is the team right now? So in China we have 2,000-plus employees.

It's a big team. And for international, we have a bit more than 100. So it's still very start, very early stage. And yeah. So I think the principle of the company in terms of hiring people is that while our obviously like we're very lean team. We hire the best of the best. And we also like love to have people on board who share some passion with us. So like we want to, we want to really make changes of the world, of the city, like on how people get around and provide

more options, more choices to the to the people. So yeah, so that's that's us. How do you think of of your product when, you know, self-driving cars are pretty much dominant everywhere in the world? Today, there's obviously more challenges they need to solve to go to level five automation with computer vision problems and, you know, solving the last mile. But in a world where you have, you know, pretty much self-driving cars, how do you see, you know, 04 bike sharing or electric bikes playing a role? Yeah, yeah. First of all, like, I love to see that day. I love to experience that, like, like, experience.

So, yeah, so first of all, like, I think even if, like, people have choices like autonomous driving cars, like everywhere, still there will be a very huge demand in terms of the bike sharing. I'm not sure whether we're going to roll out of, like, self-driving bikes, but even if you consider, like, the current product, right? So I think there are a lot of advantages here. Number one, I would say, is the density, right? So you look at how the city is planned. Like maybe in San Francisco, like the car can go to almost everywhere because the city is like blocks and everything. But in many cities like in South Asia, in developing countries, like in China, there are a lot of like small roads. Cart cannot get in.

Obviously there are arguments. Like you can probably like put the driverless car. When you take the driver out of the car, you can change the shape of the car like it looks like a Batman's like a motorbike, something like that. You can do that, right? So yeah. But within that being said, with that been said, still like I think density, is something like the shared bike has advantage on. Number two is cost.

So basically, like, autonomous driving car is a beautiful product. But my thinking is that to manufacturing that, to put in the skill, it's going to require a lot of investment. And to operating that system and to use that system is going to be very costly. So it's like you think about it, right? Especially at the beginning of that, you still need a lot of it to maintain the system, right? And at the same time, like, I would say how much, what is the pricing of taking a ton of driving car, right? Like, like, I don't know the pricing, but my imagination

feels like a fancy service, and it could be like, could be kind of... Well, the cost of making car is significant, but the driver cost and the right channel should go away ideally. Yeah. So basically, like for the, for the bike itself right now, like we do not have a driver, we do not have, we do not have a cost of energy, we don't have electricity, right? So it's all like manpowered. So it's like cost-wise, we're going to be much lower in terms of initial. So can you compare the cost per trip today on an offer versus a Diddy?

Yeah, so I mean, well, Diti is not self-driving yet. Yeah, so it's like a... No, just based on today, because that's an interesting observation. Yeah, so I think there's a, there's a, a few like a interesting, a few like main categories. So number one is the driver opportunity cost. So basically for a returing model, you need to pay the driver. Like when he gets online, there's always like charging. So it's like you need to pay for him like the hourly rate or whatever rate, like you need to, there's a cost there. And that is a variable cost. So basically like you do one more trip,

there's always like a driver's take on that. Right. So that's number one. Number two is on the energy cost. It's basically right now, most of the cars are running on gasoline, maybe some electricity, but still, like, most of them are on gasoline. So there is a huge cost there. Also, that one is variable cost. So, like, for every trip, more distance you travel, like the more you pay for the gasoline and for the driver. So these costs are not going to go away. But for bikes, like we, most of the costs we have is fixed cost. So basically, the cost does not go up as we scale, as we do more trips. So at the beginning, when we have less users like the bike or maybe the bike user twice a day, right?

So that's that's that's that's how we, how the structure is. But later when we have more users, the bike are used like five times in university was wrong, sometimes go up to 10 times. So when that comes, the cost does not go up when we, when we do the, we increase the number of trips. So still like we pay the same amount of salaries to the maintenance workers. And with our smart system, like actually their working load does not increase.

But still, like, well, when they manage, like, 200 bikes, they maintain, like, they spend time to repair maybe 30 bikes a day. And when we have them, like, manage 1,000 bikes a day, what we do is that still, like, maybe repair 35 bikes a day. So still, like, that doesn't go. But the efficiency, the cost, operation cost per bike, has actually goes down a lot. Because you get more leverage. Yeah.

You said, like, you get 20 rights a bike, right? In some university, we can get to 20. And our record is over 100 times a day. So that is, that is, well, that is one, I saw it once. So, but normally it's like 10 around in university. Got it. So, yeah, so it's quite, it's quite different model. So for us, it's more like a rental model.

For ride-sharing, it's more like a ride-sharing model, I would say. And while if you think about like a ton of driving, obviously you take the driver out. Yeah. So there is no cost for the driver. That's a massive increase of the efficiency. But still, like I said, like I said, like the, there's a maintenance cost of the car. I think that would be possibly be higher than a maintenance cost for a bike. Right.

So, and also like there is density issues. And there's also like the system management, right? So it's like the, I don't know too much about the autonomous driving, but my guess is that the total cost of operation and management will be, will not be cheap. cheap and maybe higher than what it takes to your bike yeah maybe I don't know like maybe the whole the total cost is even higher than the car system yeah the car system yeah only time we'll tell us that's true yeah we'll see yeah but I'm pretty I'm a big fan of a ton of driving so got it yeah well thank you so much Yankee thank you so much for taking the time to speak with us thanks for having me yeah thank you thank you all right thanks for listening so as always the

video and transcript are at blog dot ycombinator dot com and if you have a second please subscribe and review the show. All right. See you next week.