Office Hours with Adora Cheung

by Y Combinator6/13/2018

Adora Cheung is a partner at YC and she also cofounded Homejoy, which went through YC in the Summer 2010 batch.

For this episode we took questions from the internet. If you have questions for a future office hours episode, just tweet them our way.


0:18 – topherPedersen asks – How many users did you have when you applied to YC? Also, how much revenue had you generated?

2:32 – Hatlii asks – How do you keep going if you can’t raise any money?

4:31 – Alejandro Ruperti asks – When/how do you decide to walk away from something you started?

10:18 – Chris Melnick-MacDonald asks – What advice and lessons did you learn in entering the #Canadian market?

13:16 – Yahya Elamrani asks – Do I have to work on culture in the very early stage?

15:17 – Adam Sanders asks – What was the best decision you made for Homejoy?

16:30 – Ujjawal Chauhan asks – Would love to know what’s the one thing she’d do differently in hindsight if she were to start over again?

20:12 – Nikita Butakov asks – What are some unique data science / machine learning challenges faced by Homejoy?

21:44 – Alejandro Ruperti asks – From @tferriss : how has a failure or apparent failure set you up for a later success?

32:14 – Siamak Freydoonnejad asks – What are the best practices when doing a cold reach out to an investor?

33:51 – 😎rliesaurus @ ✈️ 🇪🇺 asks – Is Uber for X still a thing people would invest in, in the US?

36:58 – Fedor Paretsky asks – How have YC’s views on cryptocurrency/blockchain-related startups changed since Coinbase?

42:33 – Manav asks – What type of companies is YC seeing more of this time around?

46:23 – Yahya Elamrani – What are the best marketing strategies for year one for a b2c startup?

53:32 – Yahya Elamrani asks – What would you say to a founder from a third world country where there is a big lack in tech talents (and you can’t compete with big corps due to lack of resources) in term of hiring?

55:27 – Aspiring Angel asks – What’s the best way for investors interested in startup seed/angel funding get started? How does location affect the process?


Google Play


Craig Cannon [00:00:00] – Hey, how’s it going? This is Craig Cannon and you’re listening to Y Combinator’s Podcast. Today’s episode is with Adora Cheung. Adora’s a partner here at YC and she also cofounded Homejoy, which went through YC in the summer 2010 batch. For this episode, we took a bunch of questions from the internet and, if you have questions for a future Office Hours episode, you can just tweet them our way. All right, here we go. All right. Hey, everyone. Today we have Adora Cheung. She’s a YC partner and cofounder of Homejoy.

Adora Cheung [00:00:28] – Thank you.

Craig Cannon [00:00:28] – How’s it going?

Adora Cheung [00:00:29] – Good. How are you?

Craig Cannon [00:00:30] – Good. All right. We’re going to do some Office Hours questions from the internet. Let’s go.

Adora Cheung [00:00:36] – Cool.

Craig Cannon [00:00:37] – First question is from @topherPedersen and he asks, “How many users did you have when you applied to YC?” And, “Also, how much revenue had you generated?”

Adora Cheung [00:00:46] – When we applied to YC back in 2010, we were a company called, at the time, Pathjoy, not Homejoy, and we were a marketplace for online services. We were trying to bring on, typically offline services, like tutoring, life coaching, therapy and all these legal services online and our thesis was this was the time to put it to video and we would be the one to do it. When we applied, if I remember correctly, we had onboarded a lot of people for the supply side of stuff, so service providers, but had onboarded absolutely no one yet on the demand side. We had, obviously, generated no revenue yet. That’s what we had when we applied.

Craig Cannon [00:01:30] – This was 2010?

Adora Cheung [00:01:32] – 2010.

Craig Cannon [00:01:32] – Okay. Had you started a company before?

Adora Cheung [00:01:36] – No, I had previously worked at a company called Slide. I ran product there and we built Facebook apps. I started in 2007 when the Facebook platform just launched.

Craig Cannon [00:01:48] – Okay. What made you want to start a startup?

Adora Cheung [00:01:51] – I’ve always knew that I wanted to create stuff, make stuff, for the world. It just was a matter of finding the right idea. After a couple years at Slide I decided… It was the right timing also, because my brother, who would be my cofounder, had just graduated college and I said, “Hey, you should just come out here. Let’s work on stuff.” We started working on stuff.

Craig Cannon [00:02:13] – Did you have any apprehension about starting a company with your brother?

Adora Cheung [00:02:20] – Not really. It’s funny because my brother and I grew up kind of in separate worlds. It’s not like we were together all the time or knew each other’s, if we were compatible working together, so it was a little of a trial and error in the beginning. But, having a sibling, working with a sibling, you know you can trust them and they’re not going to do anything horrible to you. There’s just that initial layer of trust that was pretty good for us.

Craig Cannon [00:02:45] – I guess you were out here for only a couple years so you didn’t totally have a network either.

Adora Cheung [00:02:49] – Exactly.

Craig Cannon [00:02:49] – Right. Makes sense. Cool. All right. Next question. @Global16Citizen asks, “How do you keep going if you can’t raise any money?”

Adora Cheung [00:03:01] – You just have to keep your personal burn really low. I would say, just from experience where my brother and I weren’t able to raise much money for 2 1/2 almost three years, and we were just able to just keep going in terms of trying to find users and stuff like that. You just got to be incredibly scrappy because, obviously, you can’t afford to acquire any users. Which is actually a good thing because it forces you to get out of your chair and you have to go talk to users because it’s the only way that you can get them to use your product. Going through a period of not having enough cash to work on your startup helps you develop a lean mentality, which is good to keep moving forward if and when you raise money.

Craig Cannon [00:03:45] – Had you raised before YC?

Adora Cheung [00:03:47] – No, we hadn’t.

Craig Cannon [00:03:48] – Okay, and then you raised a hundred after that?

Adora Cheung [00:03:50] – Yeah, we raised a hundred K around demo day, and then used that for the next 2 1/2 years.

Craig Cannon [00:03:59] – That’s amazing.

Adora Cheung [00:04:03] – Yes.

Craig Cannon [00:04:04] – Was your employees just the two of you?

Adora Cheung [00:04:06] – It was just the two of us. Yeah, it was just the two of us.

Craig Cannon [00:04:10] – Wow. Okay, cool.

Adora Cheung [00:04:12] – It was cheap to live in Mountain View at that point, so it wasn’t–

Craig Cannon [00:04:15] – Really?

Adora Cheung [00:04:17] – Well, cheap relative, around that time it was about 900 bucks a month for like a one-bedroom and bath apartment.

Craig Cannon [00:04:23] – You took the Y… How much was the YC money at the time?

Adora Cheung [00:04:26] – It was, I believe we got 15K, so–

Craig Cannon [00:04:29] – Each or total?

Adora Cheung [00:04:30] – Total.

Craig Cannon [00:04:30] – Total. Okay.

Adora Cheung [00:04:32] – Then we raised a hundred K from someone else.

Craig Cannon [00:04:34] – You made that last for two years?

Adora Cheung [00:04:36] – 2 1/2 years, yeah.

Craig Cannon [00:04:36] – 2 1/2 years. Wow. Impressive.

Adora Cheung [00:04:38] – Just didn’t do much besides work on our startup.

Craig Cannon [00:04:40] – Yeah, yeah, yeah. There’s another question that’s like further down the list but somewhat related to keeping going or quitting. Alejandro Ruperti asks, “How or when do you decide to walk away from something that you started?”

Adora Cheung [00:04:59] – That’s a tough question. I’m assuming he means the startup or an idea that you’re– working on. This is one of the things you actually have to think about before you start, which is, okay, great, this is an idea, maybe you got some users, so on and so forth. Before you even think about raising money, or especially hiring people, you should really think about, is this something you want to do for the next five, six, seven, 10 years? Because success is way down the road and, I think, if you can’t imagine yourself doing that, then that’s not something you should jump into. But if it is, then, certainly, you should go for it. During your journey, your startup journey, if you ever start questioning yourself of, am I in the”right game,” or not, then, yeah, I think it’s important to always zoom out and re-evaluate. But, I wouldn’t be doing it in the corner by yourself. I would always find a set of friends in the startup world, but also outside of the startup world, to give you some of that perspective. If that doesn’t help, you can always

Adora Cheung [00:06:06] – find a coach to help you talk through that. Then, if you have raised money, or you have employees, you do have a responsibility to them, or to your investors as well as to your employees, which is not to say keep going because of that, but it’s just to say, you need to find a transitionary role or period in which you take care of all those duties.

Craig Cannon [00:06:32] – Did you struggle with that at Homejoy?

Adora Cheung [00:06:36] – No, not in particular. I just kept on going. There were times where we were like, are we working on the right idea or not? Aaron and I, my brother, who’s my cofounder, we pivoted a dozen times before we got to Homejoy, which was helpful. The fact that we pivoted that much was helpful in the sense that we learned how to push product out very, very fast, which was helpful when we finally got to the right idea. Certainly, each time you pivot, you sort of just think, maybe two more weeks, maybe there’s one more feature. Maybe if I just got one more user and things will take off. You just have to step back and always re-evaluate that.

Craig Cannon [00:07:25] – How do you pivot 12, because where you ended up from the initial application isn’t that far. Did you like circumnav–

Adora Cheung [00:07:30] – Isn’t that crazy?

Craig Cannon [00:07:32] – Yeah yeah yeah.

Adora Cheung [00:07:33] – We started with a marketplace for online services and then, it was just, marketplace for an offline service, starting with cleaning and then, eventually, some other home services. I don’t know if, I’m not sure if that pathway could have been shortened or not. I do think we learned a lot along the way in terms of working with each other that really helped us make that take off at the time that we came to it.

Craig Cannon [00:08:02] – Do you remember what you did in between? Was there anything that was just objectively, crazy, or terrible?

Adora Cheung [00:08:08] – There were a lot of terrible things. We also built, At least I built, over-engineered a lot of things because I thought it was cool to build some stuff. After, I’m not going to remember all of them, but, Pathjoy was trying to connect service providers with clients. When that didn’t start working out, when we couldn’t find enough clients for the service providers, we had all these service providers, so we started, we’re like, what can we do with all these service providers? Well, they still needed clients, and we thought, okay, what if they just, we just made it into a Q&A site. We had a Q&A site for a while. After a while, a lot of the service providers got a little tired of just creating free, essentially free, content.

Craig Cannon [00:08:54] – Oh, they weren’t getting paid, okay.

Adora Cheung [00:08:56] – Yeah, they weren’t getting paid for it. Sort of, kind of, the pitch was if you write all this content, then it’ll help you develop your brand. Then here’s a profile.

Craig Cannon [00:09:03] – Right, it’s like marketing.

Adora Cheung [00:09:04] – Yeah, marketing. Then it rolled into a bunch of content sites and then it turned into, basically, like a demand media type play and we weren’t just, I think, by the time it rolled into that, we were just like, this is not going to be huge and we’re not really into this, so, let’s just rethink the whole thing.

Craig Cannon [00:09:22] – Oh, to go back to your other ideas, you said you applied with a Twitch type idea as well, right?

Adora Cheung [00:09:29] – On my application, I just reread it, I applied with the marketplace for online services, but there’s a question on there which, says, your other ideas. My other idea, literally, is Twitch, which is not to say that I made a mistake to not build it, because I’m not quite sure I would have made that a huge thing. It’s just that it was kind of funny for me to see that.

Craig Cannon [00:09:52] – You literally just wrote like video game streaming.

Adora Cheung [00:09:54] – It was just watching people play games and checking in and commenting and doing all these things. Kind of like ESPN for games, essentially. How I came up with the idea was because I was always watching my brother play games and he would also then go on YouTube and watch other people play games.

Craig Cannon [00:10:12] – For sure.

Adora Cheung [00:10:12] – I’m just like, that’s so weird to me. But, he was doing it, and a bunch of other people were doing it and I was like, “Okay, there’s something here.” But, the thing is, I’m not really into games that much and so, I’m not quite sure that would have, I’m pretty confident it wouldn’t have worked out.

Craig Cannon [00:10:31] – Right. Yeah. It may have been a little early as well.

Adora Cheung [00:10:34] – At that point, yeah. Live streaming was, yeah, not there yet.

Craig Cannon [00:10:38] – Cool, all right, let’s go to the next question. Chris Melnick-MacDonald asks, “What advice and lessons did you learn in entering the Canadian market?”

Adora Cheung [00:10:48] – With the Canadian market, just like any other international market, from the customer perspective there wasn’t that much of a difference. You have a dirty home and you need to get it cleaned, so on and so forth. From the cleaning professional side, the service provider side, there were some differences there in terms of where do people find jobs? In the US, it’s Craigslist. In other places, in other markets, it’s something else. There are those kind of differences that we had to learn about, but I wouldn’t say they were super major. There was, how do you acquire service providers? There are maybe unique channels to acquire customers in each one of the local markets. There was things just like weather, like where are the traffic jams? Public transit. How are people actually going to go from house to house? That had to be all baked into the scheduling stuff. Anyway, going back to the Canadian point, there was nothing super special about Canada itself. It’s just the fact that it’s another market that we went into. We also launched in the Europe as well. Berlin, Paris and London and, similar to that, there were some locale differences and, obviously, language differences as well, but, nothing super major.

Craig Cannon [00:12:12] – What are your pro tips then? If I’m expanding to, I think Canada’s very close to the US, but if I’m expanding to Europe, what are your pro tips?

Adora Cheung [00:12:22] – For that is, I’d go to the market and identify the opportunity first. Go talk to potential users. For us, we’re a marketplace, so it had to be both sides of the users. There were some markets where we looked, and I don’t, can’t recall exactly the right cities, but, that we decided just not to enter because we didn’t think there was enough demand or supply there for us to make an impact. Once you do that, then the other thing is, if you’re launching a ton of these cities, is to find somebody who’s, hopefully, done this before and we call them, what we called them was the city launcher. This person would go into the market, find the first few service providers, first few cleaning professionals, and then we would match them with the customers. There was kind of like a manual back and forth going on to get the supply and demand running and then, if it worked out, then we would find a city manager. We would hire a full-time city manager to run the operations and be in charge of the P&L there. And I think sort of that playbook, that’s been played out now, and so a lot of startups have executed on this well. That’s a pretty good framework to apply.

Craig Cannon [00:13:32] – Okay. Cool. Another question related to kind of like getting started when you’re small. Yahya Elamrani asks, “How do I work on culture or do I even work on culture in the very early stage when I’m just building a team in a project?”

Adora Cheung [00:13:47] – The number one thing about working in a startup is finding the right cofounders. Finding your cofounder is pretty much setting the culture because how you work with that person is going to parlay into, probably, how you work with other people. I don’t think it’s like you need a handbook and write out your core values and all these things right away. It’s much more important to know if your startup is going to survive or not. How do you get your users? How am I going to make my first buck? Once you start hiring your first few employees, then that is an important time to try to figure out, try to articulate that thing so that, one, you’re hiring the right people, but two, the people coming in have some expectation of how things are going to be run.

Craig Cannon [00:14:34] – How did you guys figure it out? Because there’s so much unspoken with siblings.

Adora Cheung [00:14:39] – Yes.

Craig Cannon [00:14:39] – How do you determine what’s part of the company culture and what’s just like, you guys?

Adora Cheung [00:14:44] – That’s a good question. We wrote out our core values at some point. We had five or six of them. That’s the framework we chose and which a lot of companies choose. But, in the beginning, both of us would interview everybody and then we would actively talk about it. Is this the right person? Would they work with us well and so on and so forth? And then go from there.

Craig Cannon [00:15:11] – Okay. Seeing if the person’s a good fit rather than explicitly writing all the values down.

Adora Cheung [00:15:16] – Yes.

Craig Cannon [00:15:16] – In the early days.

Adora Cheung [00:15:17] – In the early days, yes. We eventually wrote all the values down, but, in the beginning, we were just looking for smart people who were going to work hard, who wanted to help cleaning professionals get work and were into the mission of the company.

Craig Cannon [00:15:35] – Okay. Cool. Another Homejoy question. Adam Sanders asks, “What was the best decision you made for Homejoy?”

Adora Cheung [00:15:46] – Starting the company, obviously, I don’t regret it at all. We were able to bring on a great crew of people to work together. What I am most happy about and was most passionate about was the fact that we created a platform that created work for people who needed the work, as well as, did a service for people who were busy and having a clean home is one of those things that, it’s hard to say, it’s hard to articulate why, but, when you have a clean home, you’re just happier and you’re more relaxed. After a busy day at work it’s nice to be able to go home and just not have to clean it. Those two things, Starting that company, I’m proud of that and, and like I said, our team was a great group of people who really cared and I would take nothing back to for that, yeah.

Craig Cannon [00:16:48] – Cool. Next question. Ujjawal Chauhan asks, “Would love to know what’s the one thing she, [you] – would do differently in hindsight if you were to start over again.” And I think that’s like Homejoy and life context.

Adora Cheung [00:17:03] – Yes. With Homejoy, there are so many things that I could have done a lot better. But if I had to aggregate a lot of them into kind of like one concept,

Craig Cannon [00:17:19] – Yeah, like a bucket, yeah.

Adora Cheung [00:17:19] – Yeah, a bucket. it would be, I would have prioritized, unit economics much earlier in our in our company. By that I mean, when you’re building a marketplace you’re always in this constant battle of trying to figure out how to allocate resources between growth, user experience and unit economics. What we did was, we obviously, we heavily worked on all three of those things, but if you had to make me stack-rank those three, user experience and growth came before unit economics. This makes sense if, and only really if, there is one ample funding available and, what I would think, three other things. One is, there are strong network effects, there are really strong network effects, or two, there are strong economies of scale, or three, there is a clear first-mover advantage. One of those and ample funding, or a combination of those last three that I said, strong network effects, economies of scale and first-mover advantage. But in any case, in the cleaning space, it turns out that there are some of those there but not to the extent of Uber or Airbnb, which meant that we did have an opportunity, in hindsight, to have a much slower growth model and still, maybe, came out at top, winning in the long-term. Like I said, in hindsight, had I known about, had I foreseen what the dynamics, the funding dynamics would have been when we needed to raise money, I would have certainly made the decision to, at least, equalize and put equal effort into those three things,

Adora Cheung [00:19:36] – growth, user experience and unit economics, if not had prioritized that much even higher.

Craig Cannon [00:19:43] – When you’re advising YC companies, do you have a rule of thumb? Every company’s different, but, say, there was a company similar to Homejoy,–

Adora Cheung [00:19:49] – Yes.

Craig Cannon [00:19:50] – You should be operating at like a, whatever, like X percent margin to be in a good spot.

Adora Cheung [00:19:57] – Funding a particular, there’s no constant margin you should be at. It’s really dependent on the local market and the prices you can actually bear. But I do emphasize unit economics and user experience above growth. I think, unless, in some very odd scenario, which there are very few that, like I said, there are really strong network effects and that I think you can, potentially, raise a lot of money, to just drive through some not-so-great unit economics. But, in general, I don’t think that’s really the case.

Craig Cannon [00:20:29] – Yeah, those are usually the catastrophic failures.

Adora Cheung [00:20:31] – Yes.

Craig Cannon [00:20:33] – All right. Next question. Nikita Butakov asks, “What are some unique data signs and machine learning challenges you faced at Homejoy?”

Adora Cheung [00:20:43] – Some of the cool things we worked on. Well, there are a lot of things, but, maybe I’ll list three. The top three, the first three that come to my mind. One is, we had to create a lot of prediction models for demand so that we knew how much supply to onboard so that supply and demand weren’t out of whack. They’re always, almost always, out of equilibrium, out of whack, but you don’t want it to be too, too much imbalanced. Another one we worked on is, we did a lot of analysis on customers and finding out which kind of customers were the best for us in terms of lifetime value. We figured out things like, we should be targeting pet owners, for example. They were like two to three X more valuable than regular, all things equal, than regular customers. Then finally, one of the more difficult challenges was scheduling when you have thousands and thousands of cleaning a day and you have preferences on both sides of the marketplace as well as things like transit, transit issues, and you don’t want people traveling too far from one house to another and so on and so forth. That was a difficult challenge that we had to solve.

Craig Cannon [00:22:04] – Cool. Next question. Alejandro Ruperti asks, from Tim Ferriss, “How has a failure or apparent failure set you up for a later success?”

Adora Cheung [00:22:15] – I hope people mostly take failure as a learning experience and not to let it get you down. But you should be honest with yourself and know what you’ve learned from it. Today I advise, I help a lot of startups out, and so I think the previous things that didn’t work out for me, it provides a really strong foundation and a basis of how I can help people. But I always caveat that and every time I help somebody I always think, “Okay, what is the situation they’re in?” It never is going to be a copy and paste from my experiences. I always think about that first and foremost. That’s what I tell most people. When they get advice from people, especially if they’re going around to way too many people, is that, you are your own situation, and people will give you, potentially, directionally, correct advice, but that’s just directionally in the right direction. Not like…

Craig Cannon [00:23:14] – Right.

Adora Cheung [00:23:15] – Like I said, it’s not a complete one-to-one mapping. You always should be taking that into account.

Craig Cannon [00:23:20] – Yeah, it’s difficult. And it messes with your confidence if you take advice from too many people, especially when they’re from all over the market. Like, if I was starting a Homejoy-type company now, and I talked to you, I think I would take your advice. But just some random other startup founder, it’s like, why?

Adora Cheung [00:23:35] – Yeah, yeah. I know, you must talk to so many founders that–

Craig Cannon [00:23:37] – Yeah yeah yeah!

Adora Cheung [00:23:37] – Sort of coming with patterns in your head of–

Craig Cannon [00:23:41] – I mean, that’s like kind of the way I see things too. It’s like just pattern matching. Unfortunately, so many of the pieces of advice just sound trite after a while Because it’s tweetable. But, in reality, it’s kind of like all about the nuance behind the scenes. So, yeah, you can tell like, culture’s important or whatever,

Adora Cheung [00:24:03] – Go talk to users. There are so many ways to say that. Stop burning money so fast. Yeah, there’s–

Craig Cannon [00:24:10] – Yeah, and so all of these things are just… At the end of the day, you have to build the thing yourself and you have to have some confidence because taste matters a lot in your company, in your product. If you don’t have your own opinion, I don’t know how you’re going to get through it.

Adora Cheung [00:24:25] – Yes, I agree.

Craig Cannon [00:24:27] – Which is, yeah, I mean, obviously, you develop that. But over time, you’re trying out 12, 13 different products before you get to, like, Homejoy. Code around into it. Actually I did want to go back because this was about like, so something a failure for a later success, I wanted to talk about you doing your PhD, as well because I think it’s also interesting to people. Would you regard that as a failure?

Adora Cheung [00:24:54] – I would regard that, not as a failure, but as time, sometimes I wish I had that time to get back and I could have been working on creating things instead of creating, I don’t know, economic models and being a data monkey and trying to do simulations and stuff like that, on my, all my fancy, very academic models. I spent, right after I graduated from undergrad, I actually got a degree in computer science and didn’t, I come from South Carolina and there, for no reason, your choices are a little limited in terms of what you’re exposed to. I was never exposed to Silicon Valley. I was never exposed to tech startups or any of these things. By the time I was like 20, I just didn’t even know that existed yet. The thing I decided to do, because I didn’t want to work at a big company as a code monkey, was to go get a PhD in economics, which sounds weird, but, I was doing some engineering work for an econ professor and so I fell deeply in love with the idea of becoming an econ professor.

Craig Cannon [00:26:12] – Could be fun.

Adora Cheung [00:26:13] – Yeah. I think it could be fun, but, I generated a lot more meaning and value doing what I’m doing today.

Craig Cannon [00:26:22] – Yeah. Value, definitely. Meaning, I don’t know. People make their own meaning.

Adora Cheung [00:26:27] – For me, personally. I have no qualms with people who decide to become econ professors.

Craig Cannon [00:26:32] – Right, yeah, I know. Totally. Question not on the list. Side projects. Are you building side projects right now to keep that product-building, I don’t know, gene muscle, working?

Adora Cheung [00:26:44] – Yeah, I started working. This is one of the things because I’m not working on a startup, I start a lot of things. I see something cool and then I just start working on it. One of those things, I’m trying to build a DApp right now.

Craig Cannon [00:26:57] – Really?

Adora Cheung [00:26:58] – Yeah, on Ethereum. Just playing around with solidity and all these things. We’re starting to invest in a lot of the crypto blockchain startups now and so I think it’s really important that I truly understand how this stuff is built. I’m doing that and nothing else major besides that. But I’m always tinkering around with something.

Craig Cannon [00:27:21] – I have the same itch and I, desperately, only want to complete things.

Adora Cheung [00:27:26] – Yes.

Craig Cannon [00:27:27] – I hate letting things die on the vine. But it’s hard because, yeah, I’m equally curious and just want to like, oh, hack away for a weekend and see what happens.

Adora Cheung [00:27:35] – Well, what’s the last thing you’ve hacked on?

Craig Cannon [00:27:37] – I put something up this week, this like auto-transcribing podcast.

Adora Cheung [00:27:40] – Oh, that’s right. You put this on Hacker News.

Craig Cannon [00:27:42] – Yeah, yeah, yeah. Which, speaking of unit economics, I might be committing the mortal sin of selling a dollar for 80 cents.

Adora Cheung [00:27:51] – Oh, because you still have a human doing it. Fixing the little stuff. Got it.

Craig Cannon [00:27:56] – I have to figure out either full-time employees or Mechanical Turk which I’ve never gotten good results from. At scale, actually at a scale a little bit larger than we’re at right now, I could have someone full-time, fixing podcasts. But right now it’s a lot of like Craig waking up at five in the morning and being like dupe dupe dupe dupe dupe.

Adora Cheung [00:28:20] – This doesn’t cost me anything because I’m doing it.

Craig Cannon [00:28:21] – Yeah, yeah, that type of thing. I’ve like create my own $5-an-hour job. It’s like, so dumb.

Adora Cheung [00:28:29] – Raise prices. You can always raise prices.

Craig Cannon [00:28:32] – Yeah, I think that’s a fear that I personally have, but should be okay with breaking, because I didn’t, I knew that, if I started too high, I wouldn’t get enough customers. Fortunately, I didn’t give away any free demos. If I gave away free demos, because we hit one on HN, and, if we were doing free demos, it would have been a disaster.

Adora Cheung [00:28:55] – Yeah, this is a very actually classical error, not to pounce on you or anything.

Craig Cannon [00:29:01] – No, that’s fine.

Adora Cheung [00:29:02] – I did it myself at Homejoy too, which is, to get the first few users in the door, people tend to want to discount and then, which you haven’t done yet, but, is that you continue discounting and then all of a sudden you have a stream of just the worst kind of users. ’cause the people that come in just because of the discount are super value-driven, so they want the cheapest thing and they want the 10-star service for what you can only provide at maximum as a five-star service.

Craig Cannon [00:29:27] – Yeah, yeah, yeah.

Adora Cheung [00:29:29] – And they always drive the most customer service costs and it’s just a big headache. It’s, maybe okay, to get the first few in because, what they’re, hopefully, also giving you in return, is a lot of feedback. You’re improving on the experience and stuff like that. But at some point, you definitely don’t want to be scaling that.

Craig Cannon [00:29:47] – Uh, definitely not. No, it’s only a couple, a handful right now. It’s totally fine. But, I just, honestly, didn’t know if people wanted it and that’s why I was like, this would be a easy way to find out. Turns out they do and it’s a lot of work.

Adora Cheung [00:30:04] – Yes. Podcasting is getting really popular–

Craig Cannon [00:30:06] – It’s crazy.

Adora Cheung [00:30:07] – And the need to edit a lot of them is pretty high too.

Craig Cannon [00:30:11] – Yeah, I know. Well, we were doing the, not to make this a Craig podcast, but we were doing the text-based audio editor for a while, and I found that I was using it a little bit and I was using the other ones, like Descript and there are some competitors, but I’m still so much more proficient in Premiere. But, at the end of the day I still wanted transcripts, and so I was like, this is a product I want. And, yeah, it turns out–

Adora Cheung [00:30:35] – So the idea is that it do transcripts and then you can also edit it, based on the transcripts.

Craig Cannon [00:30:38] – We’ve killed

Adora Cheung [00:30:38] – Oh you moved that one, okay.

Craig Cannon [00:30:38] – We’ve killed the editing because it’s really interesting on the technical side. But, yeah, like I said, it wasn’t the product that I needed anymore.

Adora Cheung [00:30:49] – Got it.

Craig Cannon [00:30:50] – Which is, I’m a little rare, because I do video as well.

Adora Cheung [00:30:55] – It’s a lot more helpful when you put it on the blog that there’s a transcript there, as well, to follow and look back on, and probably to search through as well, I imagine.

Craig Cannon [00:31:03] – Yeah, so it’s kind of, it’s multifaceted. It’s like SEO for sure. Sharing, because we can generate little quotes and stuff. Some people only read podcasts, which is personally, is like an affront, but, that’s fine.

Adora Cheung [00:31:19] – Are you a 2x listener?

Craig Cannon [00:31:21] – No.

Adora Cheung [00:31:21] – Really?

Craig Cannon [00:31:22] – I’ve done it before. I used to do it and then I realized I was talking faster.

Adora Cheung [00:31:28] – Yes.

Craig Cannon [00:31:28] – Because, I mean, we’ve talked about this before, but you and I would like, be at home all day and not talk to someone. I had a moment where I was like, I’ve only listen to podcasts today and I’m at a coffee shop talking to the host at like 2x speed.

Adora Cheung [00:31:44] – There’s, if all you do all day is just listen to things at 2x and then you go talk to somebody, you always wonder, it’s like, why are you talking so slow? They’re talking normal speed but it’s like you are talking unbearably slow right now, speed up.

Craig Cannon [00:31:57] – I’ve actually met a couple podcasters in person after I was listening to them at 2x speed. It was like, “Oh my God, you’re so slow.” Then you get really used to listening to podcasters so you know what, like, uh, Joe Rogan, I’ve heard so many of his episodes that I, basically, know the anecdote that’s coming up and so I’m like, skip, skip, skip, skip. But, anyway, podcasting is awesome. TBD how much money there is in it.

Adora Cheung [00:32:24] – Mm, yes.

Craig Cannon [00:32:24] – Because it’s all like content marketing, or, a lot of content marketing right now.

Adora Cheung [00:32:27] – Yes, yeah. Agree.

Craig Cannon [00:32:29] – Yeah. Cool. All right. Next question. Siamak F., I’m not going to pronounce that last name, asks, “What are the best practices when reaching out cold to an investor?”

Adora Cheung [00:32:42] – Really easy. Keep it short. Say what you want to say. If you can do it in two to three sentences and then ask a specific question, which is, can we get, which is not can we get coffee? Which is, actually, something that they can answer and actually help you with. That’s not to say that that investor would never meet with you. It’s just, you have to remember these investors are getting lots of these emails and they’re also, their job is to take meetings from people that are interviewing them, so all these warm intros. You just need to figure out a way to develop a relationship with that person. One way is just to have a back and forth over email over time. And then when it becomes interesting enough, basically, when I get to the point where I’m writing three or four paragraphs back, I’m like, okay, nevermind. Let’s just either get on the phone or meet in person because this has now gotten to the point where I’d like to actually talk to you for a long bit.

Craig Cannon [00:33:35] – Also investors hang out on the same sites that hackers hang out on and product people hang out on. If you’re out there making stuff, you can get attention.

Adora Cheung [00:33:45] – Oh, yeah. Reddit, Hacker News, all these places.

Craig Cannon [00:33:48] – They’re all there. They hang out on Product Hunt. It’s like you don’t have to just pretend that only email is the way to get their attention. Because this is one of those things that’s like you make up these little excuses for yourself. Where like, Adora didn’t reply to my email. It’s never going to work out. That’s just cheating.

Adora Cheung [00:34:07] – Right. I agree, totally.

Craig Cannon [00:34:09] – All right. Next question. From a name with a bunch of emoji in it, I think it says rliesaurus. They ask, “Is Uber for X still a thing people would invest in in the US?”

Adora Cheung [00:34:22] – I definitely think so. We’ve seen the scooter craze going on now. When I think about Uber for X, I think it’s push a button and a bunch of logistically-hard operational things happen in the back and then you get a product or service like within a few seconds, few minutes, within a day or something like that. That, to me, shouts convenience and a lot of people want convenience. That’s something people want. There’s always products and services to be built in the Uber for X realm. I think that, yes, there’s definitely less, overall less, money being put into it than, say, four to five years ago. But, and that’s a result of investors understanding and learning about what these business models look like, the fact that most of these businesses grow a little bit slower so that you can build up a great user experience. They understand the uneconomics much better as well. There’s, I would say, not as much hype, but there’s certainly, if you’re growing and you have good retention and good engagement, you certainly can probably get funded.

Craig Cannon [00:35:31] – Has there been a large success that was kind of like a follow-on company in this way? A copycat or just like a derivative? I’m trying to think of one.

Adora Cheung [00:35:42] – What’s an example?

Craig Cannon [00:35:44] – Using the Uber for X example, then you’re just like, “Oh, we’re going to make this thing because it’s just like Uber.”

Adora Cheung [00:35:54] – When Uber came out there were a lot of these Uber for cookies and, these like little things like that.

Craig Cannon [00:36:00] – I know there are a million failures. I’ve been to a hackathon before.

Adora Cheung [00:36:02] – Oh, you’re looking for a success. I would group all this into kind of the sharing economy stuff as well. If you look at mobility, there’s cars, bikes, scooters now, ebikes. Outside of that, let’s see, there’s It’s a good question. For some reason, nothing jumps in my mind.

Craig Cannon [00:36:30] – I’m sure it exists, but I think it’s just not like a dominant player, like something massive.

Adora Cheung [00:36:35] – Yeah, I can’t think of anything massive except for, let’s see, if we go to Asia and China, all the food delivery stuff is rocking and rolling there. Same here, actually. There’s Uber Eats and DoorDash. I think that’s working well. There’s some storage companies that may do well. If we try to compare it to Uber, I think it’s going to be very hard.

Craig Cannon [00:37:06] – Right, but actually… Google wasn’t the first search engine, so, it’s definitely not about being first.

Adora Cheung [00:37:13] – Yes, yes. Exactly, yes.

Craig Cannon [00:37:15] – Yeah. Nevermind. All right. Next question is Fedor Paretsky. They ask, “Have YC’s views changed on cryptocurrency/blockchain-related startups since Coinbase?” Coinbase is a YC company.

Adora Cheung [00:37:31] – Right, so, has it changed? We’ve learnt a lot more just because there’s more activity in the space. Every partner probably has a different viewpoint of it. For me, personally, I’m super excited about what’s going on. In general, people should be excited when a lot of smart people are working on something diligently. I think there’s a lot of tools and stuff like that still to be built for this to take off. But, if I try to game out how far this could go I just think the possibilities are limitless. The stuff that excites me are when you look at countries, cities or countries, in which there are bad currencies, or just bad financial infrastructure, I think these are obvious places where blockchain and crypto can help a lot and move them forward, if not even leapfrog, maybe, other countries. Then, similarly, in those countries, usually there’s also issues with property rights and stuff like that. I’ve talked to the folks in Greece and places in South America where the concept of who owns land, like each piece of land is almost non-existent. It’s somewhere in a drawer, maybe, in pencil and paper, but, even then, who knows if that’s real. If any of these countries or cities are serious about it, they could leapfrog centralized services and just put it onto the blockchain. Wouldn’t it be cool if you do that, not only… Well one is, just identifying who owns property. That would drive, I think, economic growth in itself because it would incentivize people to actually do stuff. But on top of that is

Adora Cheung [00:39:25] – increasing the number of transactions that can happen. Even in America, trying to buy property or exchange property is extremely hard. It’s extremely expensive. It seems to be unnecessarily so. Anyway, I think, like land registries is an example of something I would be super excited about to see happen.

Craig Cannon [00:39:44] – Is that the DApp you’re building?

Adora Cheung [00:39:47] – No. What I’m building is just something very simple. I’m not…

Craig Cannon [00:39:52] – Oh, is it just like you’re building a blog?

Adora Cheung [00:39:54] – Yes. Almost.

Craig Cannon [00:39:54] – It’s like the online tutorial?

Adora Cheung [00:39:56] – Almost exactly that. Sending myself fake tokens. The thing that comes to mind for me with crypto and blockchain is, there is a lot of skepticism because of what’s being built right now. But, I actually think that, what’s being built now, are the right things being built, in the sense that one, it’s such a new technology that there are still so many issues, potential security issues, and putting the wrong types of smart contracts on there and so now you see a slew of consultants and also companies, doing the security audits and we’ve just invested in a couple companies that are building AI to help ensure better security and ensure that the economies don’t go out of whack and stuff like that. I think doing playful things and building playful apps is actually the right way to help develop the ecosystem because you don’t want, I think, super serious things on there without all of this stuff being built already.

Craig Cannon [00:41:06] – Right.

Adora Cheung [00:41:08] – One of the things that we talked about earlier was, for this to go anywhere as well, we need UX people and product people. Right now it is just way too hard for somebody to use these DApps. Do you know what the DAUs for DApps are? Try to guess.

Craig Cannon [00:41:29] – I’m guessing it’s really small. 10,000?

Adora Cheung [00:41:33] – Wow, okay, that was pretty good. I don’t exact number now, but the last time I read it, it was around 10,000.

Craig Cannon [00:41:38] – Really? All right.

Adora Cheung [00:41:38] – Yeah, 10,000.

Craig Cannon [00:41:39] – That’s very, Well, I just know how much of a pain in the ass MetaMask is.

Adora Cheung [00:41:43] – Yes, and you got to And you’re like, come on.

Craig Cannon [00:42:19] – For sure.

Adora Cheung [00:42:20] – So, there’s a bit of an education. I sort of think that, when it goes mainstream, we don’t need to use the word blockchain per se. Just like we don’t use the word database to explain–

Craig Cannon [00:42:31] – Facebook, yeah. No. Like my parents don’t know what AJAX is. It’s not important.

Adora Cheung [00:42:35] – Yes. Exactly. You just need to tell them what it does.

Craig Cannon [00:42:39] – Yeah, it’s, ultimately, it’s the product. But I think with so many smart people working on it, it seems highly unlikely that something won’t come out of it.

Adora Cheung [00:42:49] – Yeah. It’s a hope.

Craig Cannon [00:42:52] – Cool. Probably related to that question, Manav asks, “What type of companies is YC seeing more of for this batch?”

Adora Cheung [00:43:02] – We’ve seen a lot more applications on the crypto and blockchain front. Ssuper excited about that. We continue to see more and more applications on AI, particularly machine learning and deep learning and the applications for it across so many fields now and so that’s exciting. I don’t think we’ve seen anything where we’ve seen complete automation yet of anything really. But, it’s kind of cool that we’re building tools and stuff like that to help doctors, people in the field doing work, and stuff like that. What else are we seeing? Biotech, we’re seeing lots of those, which is exciting, particularly the intersection of software and biology. Also we just launched our YC Bio program which all the companies go through the three-month program that all the other startups go through, but it’s targeted towards people, or companies that are still in the lab, research phase. We’re focused on one sub area right now which is health span and longevity. We’ve seen a lot of those and we’re really excited about that.

Craig Cannon [00:44:16] – Do you have a strong opinion on how long you think you’re going to live?

Adora Cheung [00:44:21] – I don’t know. I hope for a long time. I would like to live over a hundred I guess. Or whatever the current life expectancy is, which is lower than a hundred.

Craig Cannon [00:44:31] – I think it’s 78 or something.

Adora Cheung [00:44:32] – Yeah, something like that. I try to do all the things to make sure I’m healthy and sane, but, I don’t do any drugs or anything to make myself go.

Craig Cannon [00:44:48] – No caloric restriction?

Adora Cheung [00:44:51] – Oh, so, I’ve tried doing that and I didn’t notice any specific change other than a little bit of weight loss. But, I actually go back and forth on that a little bit, trying it out.

Craig Cannon [00:45:04] – Oh, really?

Adora Cheung [00:45:05] – I do the whole, I’ve done in three six-month spans the whole 18, like don’t eat for 18 hours, and, actually, the first week is kind of hard. It’s just like probably any diet you go on. It’s like the first week is kind of hard, but it’s actually pretty simple. I haven’t done the whole fast for 48 hours thing though. Have you?

Craig Cannon [00:45:30] – Yeah, yeah. I mean, it’s fine. Like, you–

Adora Cheung [00:45:32] – You feel better?

Craig Cannon [00:45:33] – You feel fine. You get used to it. The first, by the end of the first day, you really are starting to get hungry. And then you push through it and then you could go like another day or two and be totally fine. But, also, my body fat isn’t 1%, so, maybe if I was that skinny it would be different? But, I’m similar to you. I’ve tried everything out. I know when I do the intermittent fasting or whatever I can be leaner, but, the selling point is that you’re going to live longer, right? So it’s just hard to feel a difference. Ultimately with all this stuff, I’m just kind of like, man you got to live. I want to hang out with my friends and have a beer. So, I also want to make it to a hundred, over, but, I don’t know, we’ll see. We got some time.

Adora Cheung [00:46:23] – Yeah, I want to make it pretty far and still have an active mind. I think that’s probably the most important for most people is just having an active mind still.

Craig Cannon [00:46:31] – Yeah, it seems like the further we can push the quality of life, the happier we’re going to be overall. ‘Cause it’s not really about, like, yeah, a lot of people lose it well before 78. Cool. All right. Next question. Yahya Elamrani asks, “What are the best marketing strategies for a year one B2C startup?”

Adora Cheung [00:46:55] – The best ones are always the unpaid ones. This is almost a cop out, but, if you build a product that people love, they’re going to talk about it and that’s the best way to grow. Tactically speaking, there are other things you can do. Content marketing, which you’re an expert in, so maybe you can talk more about that. There are all the tactical things. If people are coming to your website and no one’s signing up or anything, one of the things that, this seems like counter-intuitive advice, but, is to just put a phone number on there and people still want to call. People still like to talk to people. Surprise!

Craig Cannon [00:47:39] – I never installed the Drift or Intercom on our product until last year. People use it all the time. It should be obvious because it’s a successful company, but, they use it all the time.

Adora Cheung [00:47:49] – Yup. Yeah. Having a chat on there. Putting a phone number on there. It’s actually a good thing because if no one’s signing up, at least you can talk to these people and figure out what it is that they want or what’s going wrong. Finding just easy ways for people to contact you is also another way.

Craig Cannon [00:48:07] – I would ditto the content marketing. I would say, don’t waste your time if you’re not going to make something good because it’s really easy to get on the content-marketing treadmill and be like, “Ah man, we just got to bang out this stuff, put something out every week.” If you look at the analytics on YC content, it’s like a power law with like the stuff that gets the most attention. It is definitely not linear. Even just the things, I’m sure with you as well, like the things you work on, what people talk to you about, you’re like, “Dude, I did that article in like two days. Three years later you’re talking to me about this thing.”

Adora Cheung [00:48:49] – The ones with the least amount of effort you get the most views.

Craig Cannon [00:48:51] – It’s crazy. I don’t know. Maybe you should must write silly, pithy tweets. Be a thought later.

Adora Cheung [00:49:00] – Do lists still work well? Because one of the top posts is that “Essential Advice,” listing you think?

Craig Cannon [00:49:05] – Yeah, that works well. We couldn’t replicate that essential YC advice post ’cause that was an aggregation of posts over time. People like lists. People like book lists. But, you also want to think, why are the people even reading it? Like, we can, with a lot of our content, we could get more attention by interviewing like YouTube celebrities, but, to what end? I was talking to someone who watches the channel and they were like, “Oh, that’s cool. I like all your like science and tech videos. Like, what’s Y Combinator?” And it’s like, “Oh shit.” If you’re not getting them all the way through the door, it’s, yeah, especially when you’re tiny and you don’t have a full-time person making stuff. I would just figure out what actually works for you. Focus on doing a good job of it rather than just mailing it in.

Adora Cheung [00:50:08] – Sure.

Craig Cannon [00:50:08] – Because that stuff is, I’ve never had luck with.

Adora Cheung [00:50:11] – Do you think that it’s articles that work best? I guess maybe it’s depending on the product. Or, do you think podcasts, videos, like what form, what medium works well these days?

Craig Cannon [00:50:22] – People still undervalue the ability to write clearly. There aren’t that many people that can write well. If you can get someone who can write well, specifically in a niche, and then get a channel to engage with at like HN, you can get a ton of traffic because people can consume the entirety, for the most part, consume the entirety of the content, whereas with a podcast, you have to be like hardcore podcaster or YouTuber. YouTube works pretty well when we cut it up into clips. But, if you look at the retention, it’s like anything. There’s a huge amount of drop off. Whereas with the articles, it’s easier to do. But, I find I don’t always have a great thing to write about and the podcast is a much easier way to keep consistent.

Adora Cheung [00:51:17] – Are there, not to make this into a podcast about content marketing, but, this is, to me, intriguing. Are there like blogs or what companies, do you feel like, do content marketing really well?

Craig Cannon [00:51:31] – I would break it apart into different categories. Actually, I think, not to blow YC’s own trumpet, but I think PG did a great job. I think YC’s actually one of the best content marketing companies ever, except they would never say that. But HN and PG essays are content marketing. On the companies, I think 37signals, Basecamp, does a really good job. Those guys have established themselves as thought leaders but also make content that’s pretty relevant. And they’re gutsy enough to have an opinion, whereas a lot of people aren’t.

Adora Cheung [00:52:11] – Sure. Strong opinions matter and probably get more views for sure.

Craig Cannon [00:52:16] – Especially when it’s contrarian. They exist in the software space, but they’ve pushed pretty hard to be kind of independent. In a similar way, Indie Hackers did quite well for that. I think that that Stripe acquisition made a lot of sense because people connect to Courtland. Intercom has done a really good job. Honestly, I pay attention to a lot of people on YouTube because I think, for the most part, Silicon Valley still doesn’t understand how big it is. Just the amount of pure traffic and consumption.

Adora Cheung [00:52:51] – It’s a rabbit hole.

Craig Cannon [00:52:52] – It’s insane.

Adora Cheung [00:52:54] – I found myself watching some weird videos after two or three hours.

Craig Cannon [00:52:57] – Yeah yeah. Totally. Usually one after the next, after the next, after the next and you’re like, so we did a podcast with Casey Neistat and he gets more views a week than almost any TV show. And you’re just like, what? This guy rides a Boosted Board around. And then you find out that he’s representing a significant percentage of Boosted’s sales. Just this one dude in New York. He’s super talented and that’s like a lifetime of work and a ton of creative energy, but it works really well like if you can find someone like that. But then, on the SEO side, I think NerdWallet has done a great job. They have all those best credit card type things and it’s not spamy. We did a whole content marketing podcast actually.

Adora Cheung [00:53:43] – Oh, we should link to it. It’s in the link, below.

Craig Cannon [00:53:45] – Yeah, First Round and Andreessen Horowitz.

Adora Cheung [00:53:47] – Oh, cool.

Craig Cannon [00:53:47] – Yeah, so we’re like–

Adora Cheung [00:53:49] – Oh, that’s right, I listened to that.

Craig Cannon [00:53:49] – They’re competitors.

Adora Cheung [00:53:49] – That was good, yes.

Craig Cannon [00:53:52] – All right. Next question. Yahya Elamrani asks, “What would you say to a founder from a third world country where there is a big lack in tech talents and you can’t compete with big corporations due to a lack of resources in terms of hiring?”

Adora Cheung [00:54:07] – The way to always get around this is to seem more exciting than these big corporations. One way is, if you’re technical, so you just build the product yourself and start getting users and, when you start growing, and you start getting revenue you can probably go poach these people. I always think that, in a big company, there’s always a fellow entrepreneur in there somewhere, stuck somewhere. But if you’re not technical it’s a little bit harder. But there are enough SaaS tools out there these days where like Weebly and all these things where you can just patch together something and get something out there and iterate a bit to the point where you can also get users and revenue. I even see today, engineers, like real engineers, they’re sort of embarrassed to do it, but they have to do it and they’ll just put up a Shopify site first. They totally hate it because they’re like, “I can do this much better if I can customize it myself,” but And I don’t have to pay the fees.

Adora Cheung [00:55:13] – But, in reality, that’s how you’re going to get your first few users quickly–

Craig Cannon [00:55:17] – Totally

Adora Cheung [00:55:17] – is doing that. Once you do that, then, again, you can go through the same cycle of trying to go find people to join you with something that looks exciting.

Craig Cannon [00:55:29] – I would encourage people with sales aptitudes to get people to pay them. Because if you’re good at sales, you’re probably good at convincing people to think that your product is good before it exists. And only once they give you money do you know if they really want it. I’ve made that mistake before. All right. Aspiring Angel asks, “What’s the best way for investors interested in startup seed or angel funding to get started and how does location affect that process?”

Adora Cheung [00:56:02] – These days it’s really easy to get started. I think one place you can go is, we have a Startup Investor School that we held a few weeks back that one of the partners here, Geoff Ralston, did an amazing program with it.

Craig Cannon [00:56:15] – It’s all on YouTube.

Adora Cheung [00:56:16] – And it’s on YouTube, yup. I would watch all of those a couple times. Another thing to do is, there are, like we said earlier, there’s a lot of online communities where people are making things. It’s just start hanging out there and talking to founders and seeing how they think. More importantly, if not most importantly, is using the products and figuring out, developing hypotheses on, is this going to get big? How could this get big? Where could this go? Finally, just contacting the founders. I think founders, especially when they’re starting off, no one’s really talking to them that much.

Craig Cannon [00:56:54] – No.

Adora Cheung [00:56:55] – If you reach out to them, I think some of them, some percentage of them, will be more than happy to start talking to you.

Craig Cannon [00:57:04] – What about when you start talking to a founder, and say you’re even talking to a YC founder, and you feel like the price is high?

Adora Cheung [00:57:12] – I would say is, if you’re doing early-stage investing, price should not be really an issue because. In the venture business, it’s the exit that counts. You shouldn’t be investing, in a company that you don’t think is actually going to do well, to whatever standard it is. It doesn’t have to be a billion dollars, but, to whatever standard you want it to be. Honestly, at the end of the day, valuation is supply and demand and that is what drives some of those valuations. Sometimes, you just have to pay up to get involved.

Craig Cannon [00:57:51] – Yeah. All right. Last question. Adora, you’ve been at YC now for 2 1/2 years, roughly. How has your view of startups and the world, in general, changed since then?

Adora Cheung [00:58:07] – Despite all the horrible stories in the news, I have actually become much more optimistic about life and the world and startups in general. We who are at YC, we’re sitting in a pretty privileged position of being able to spend a lot of our time thinking about problems, talking with founders who are trying to solve these problems and then seeing all these new, cool technologies that they’re creating, that I certainly couldn’t do myself. When I think about if just some of these people do well, the world can change, for the better. I’m just optimistic, I guess. More optimistic. What about yourself?

Craig Cannon [00:58:57] – I’m in complete agreement. I think that it’s too easy, if you read the news, to believe that everything is black and white and there are, I didn’t quite realize it until I was at YC, that people want to treat Silicon Valley, broadly, and YC, specifically, as kind of like the Yankees. It sort of doesn’t matter what you do. There are always going to be haters out there. But, seeing people come in… Specifically, when they’re just like, they’re building like artificial wombs and all this crazy stuff that’s going to be the future and it’s super exciting. My personal strategy is to just not pay attention to stuff and talk to people who are building things.

Adora Cheung [00:59:46] – You know what is really, is somewhat indicative of this, is the scooters. Because the scooters, When I got on one I fell in love with it and I just

Craig Cannon [00:59:55] – Really?

Adora Cheung [00:59:56] – Yeah, I just like, this could change mobility within a city. Not just scooters, but bikes and Uber and all these other… and stuff like that. But then, when it started taking off people just started making fun of it. They’re like, what is this San Francisco thing, and like ruining everything, and everyone looks like dorks riding them. But then, the critics, I think, started, some of them, anyway, I think, started riding these scooters and they’re like, oh, it’s actually, like, I can see, I can see why this could be a thing and so just by using it and being part of it. There are some of these barriers, may be weak, that are dropping.

Craig Cannon [01:00:38] – A lot of companies and products have done a bad job at making people feel included.

Adora Cheung [01:00:43] – Yes, I agree.

Craig Cannon [01:00:43] – Because I think that people react strongly when they feel like they don’t have any agency. Ironically, people will be tweeting something about how terrible something in Silicon Valley is and how people are terrible, and then you meet the people here and they’re just like, I don’t know, I just want to make something cool. There’s this very weird divide that’s kind of just in people’s heads.

Adora Cheung [01:01:06] – Yes, and in this day and age, it’s everyone’s responsibility to know to at least try to predict the reaction of anything you put out there because the world is, the world is at where it is now, for better or worse, and so it’s just being very thoughtful of what you put out there is actually important.

Craig Cannon [01:01:30] – Yeah. Totally. Overwhelmingly positive and I think technology is much bigger than a little app on your phone.

Adora Cheung [01:01:38] – Yes.

Craig Cannon [01:01:40] – Keeping that in mind, I think it keeps you excited.

Adora Cheung [01:01:38] – Yes.

Craig Cannon [01:01:42] – Cool. All right. Well, thanks so much for making time.

Adora Cheung [01:01:45] – Thank you so much. I had fun.

Craig Cannon [01:01:47] – All right. Thanks for listening.As always, you can find the transcript and video at If you have a second, it would be awesome to give us a rating and review wherever you find your podcasts. See you next time.


  • Y Combinator

    Y Combinator created a new model for funding early stage startups. Twice a year we invest a small amount of money ($150k) in a large number of startups (recently 200). The startups move to Silicon