Lecture 8: Doing Things That Don’t Scale, PR, and How to Get Started 第8课:做一些不符合规模的事情,公关,以及如何开始

Lecture 8: Doing Things That Don’t Scale, PR, and How to Get Started



图片[1]_Lecture 8: Doing Things That Don’t Scale, PR, and How to Get Started 第8课:做一些不符合规模的事情,公关,以及如何开始_繁木网


Stanley Tang
Thanks for having me! I’m Stanley, the founder of DoorDash. It’s really amazing to be here, because it wasn’t actually that long ago when I sat in your seats. I was class of 2014, graduated in CS, as well as my cofounder Andy. For those of you who don’t know what DoorDash is, we’re building an on-demand delivery network for local cities. I want to start off with this photo that I took a few months ago. This was the night when we just raised our series A. I took this photo as I was walking back to where I lived; I actually lived in Roble at the time on campus. I took this photo because I realized just how ridiculous the combination of things I was holding in my hand was at the time. I was holding my CS247 homework, my tax forms (it was April – so I had to fill out my tax forms), that yellow speeding ticket, and right below that was a $15 million piece of paper I had just signed from Sequoia. And that kind of summarizes just how ridiculous our journey has been, doing this while I was at Stanford, and then transitioning this to an actual startup. I want to share with you that story today.
It all began two years ago in a macaroon store. It was my junior year at Stanford during the fall quarter. At the time, I was really passionate about building technology for small business owners. I sat down with Chloe, the owner of Chantal Guillon, a macaroon store in Palo Alto at the time, just interviewing her, trying to get feedback on this prototype we’d been working on, and also just learning about what her problems were in general. It was during this meeting when Chloe first brought up this problem of delivery. I remember she brought down this really really thick booklet. She showed me pages and pages of delivery orders, and a lot of these orders she had to turn down because there was no way she could have fulfilled them. She had no drivers, and she ended up having to personally deliver all these orders. That was a very interesting moment for us.
这一切都始于两年前的一家杏仁饼店。那是我在斯坦福大学三年级的秋季学期。当时,我非常热衷于为小企业主构建技术。我和Chloe坐了下来,他是当时帕洛阿尔托一家叫Chantal Guillon的杏仁饼店的老板,只是采访了她,试图得到我们一直在研究的这个原型的反馈,同时也了解了她的一般问题。克洛伊是在这次会议上第一次提出分娩问题的。我记得她带来了这本厚厚的小册子。她给我看了一页又一页的送货单,很多订单她都不得不拒绝,因为她不可能完成。她没有司机,最后不得不亲自交付所有这些订单。对我们来说那是一个非常有趣的时刻。
Over the course of the next few weeks, we talked to around 150 to 200 small business owners, and when we brought up this idea of delivery, they kept agreeing with us; they would say, “You know, we don’t have delivery infrastructure. It’s such a huge pain for us. There aren’t any good solutions out there.” This led us to wonder, delivery is such a common thing, such an obvious thing; why hasn’t anyone solved this yet? Like, we must be missing something here right? We thought it was maybe because people had already tried this in the past, but they failed because there wasn’t consumer demand for this. We asked ourselves, “How can we test this hypothesis?” We were just a bunch of college kids at the time. We didn’t own trucks or delivery infrastructure or anything like that; we couldn’t just build a delivery company overnight right? So how could we test this assumption we had?
We decided to create a simple experiment with restaurant delivery. We spent about an afternoon just putting together a quick landing page. When I went on the Internet, I found some PDF menus of restaurants in Palo Alto. We stuck it up there and added a phone number at the bottom, which was actually our personal cell phone number. And that was it. We put up the landing page and called it PaloAltoDelivery.com. This is actually what it looked liked (PowerPoint slide): It was super simple, ugly, and honestly we weren’t really expecting anything – we just launched it. What we wanted to see was just would we receive phone calls, and if we got enough phone calls, then maybe this delivery idea was worth pursuing.
我们决定做一个餐厅送货的简单实验。我们花了大约一个下午的时间整理了一个快速登录页。当我上网时,我发现了一些帕洛阿尔托餐馆的PDF菜单。我们把它贴在上面,在底部加了一个电话号码,实际上是我们的个人手机号码。就这样。我们把登陆页挂起来,然后叫它PaloAltoDelivery.com网站. 这就是它看起来的样子(PowerPoint幻灯片):它超级简单,丑陋,说实话,我们并没有真的期待什么-我们只是推出了它。我们想看到的只是我们是否会接到电话,如果我们接到足够多的电话,那么这个送货的想法也许值得一试。
So we put it up there; we weren’t really expecting anything, and all of a sudden we got a phone call. Someone called! They wanted to order Thai food. And we’re like, “This is a real order; we’re going to have to do something about it.” So we’re in our cars and we’re like, “We’re not doing anything right now, might as well swing by, pick up some Pad Thai, and let’s try to see how this whole delivery thing works.” And we did. We delivered it to some guy up on Alpine Road I remember. We asked him, “How did you hear about us, what do you do?” He told us he was a scholar, and then he handed me his business card and told me he was the author of a book called Weed the People. That was our first ever delivery. It was like the best delivery/worst delivery you could ever ask for.
And then yeah, the next day we got two more phone calls. The day after that we got five, then it became seven, and then it became ten. And then soon we began to gain traction on campus through PaloAltoDelivery.com which is pretty crazy, because think about it: This was just a landing page. You had to look up PDF menus to place your orders and then call in. This isn’t exactly the most professional-looking site, yet we kept getting phone calls; we kept getting orders. And that’s kind of when we knew that we were onto something. We knew we found a need people wanted when people were willing to put up with all of this.
I think another key point to remember is we launched this in about an hour. We didn’t have any drivers; we didn’t have any algorithms; we didn’t have a backend; we didn’t spend six months building a fancy dispatch system – we didn’t have any of that. We just launched because at the beginning it’s all about testing the idea, trying to get this thing off the ground, and figuring out if this was something people even wanted. And it’s okay to hack things together at the beginning.
At YC there’s a mantra we like to talk about that is doing things that don’t scale. So at the beginning we were the delivery drivers. We would go to class, and then after we would go deliver food. We were the customer support; you know I sometimes had to take phone calls during lectures. We spent afternoons just going down University Avenue just passing out flyers trying to promote DoorDash. We didn’t have any dispatch system so what we had to do was use Square to charge all of our customers. We used a Google Doc to keep track of our orders. We used Apple’s Find My Friends to keep track of where all of our drivers were. You know, just stuff like that, just hacking together solutions to try to get this thing off the ground. In fact at one point we were growing so fast that Square actually shut us down because we were under suspicion for money laundering. I mean think about it, we were getting small chunks of $15-$20 orders coming in at a rapid pace. Luckily, my cofounder Tony worked at Square so he just emailed some buddies there and everything was solved.
Another thing about doing things that don’t scale is it also allows you to become an expert in your business, like driving helped us understand how the whole delivery process worked. We used that as an opportunity to talk to our customers, talk to restaurants. We did dispatching which helped us figure out – you know, we manually dispatched our drivers and that helped us figure out what our driver assignment algorithms should look like. We did customer support ourselves, getting real-time feedback from customers. I remember for the first few months when we got started, we would manually email every single new customer at the end of every night asking how their first delivery went, and how they heard about us. We would personalize all these emails: If I saw someone order chicken skewers from Oren’s Hummus, we would say “Oh I love Oren’s Hummus. How are your chicken skewers? How did you hear about us?” Feedback like that was really valuable, and customers really appreciated that.
I remember this one time – this was during YC – we had just come out of a meeting with one of our restaurant partners, and we heard about this ice cream place that had just opened up on University Avenue called Cream, and we wanted to go try it out. Then all of a sudden, our cofounder back at our office/house texted us saying “Oh we need drivers on the road; we got a huge spike in demand.” So we debated for maybe about 10 seconds if we should go get ice cream or should we go deliver. We obviously went to deliver, but that kind of became our motivation on scaling, like you know, if we would scale, then we could go get ice cream next time.
Now of course we scale across different cities. Now we have to worry about building automated solutions, building dispatch systems, and figuring out how to match demand and supply – all that fancy technology stuff. But none of that mattered at the beginning because at the beginning it’s all about getting the thing off the ground, and trying to find product-market fit.
Just to summarize, there are three things I would say I learned from doing DoorDash. First, test your hypothesis. You want to treat your startup ideas like experiments. The second thing is, launch fast. We launched in less than an hour with a really simple landing page. And finally, it’s okay to do things that don’t scale. Doing things that don’t scale is one of your biggest competitive advantages when you’re starting out, and you can figure out how to scale once you have your demand. And then maybe once you’ve scaled, then you go get that ice cream. Thank you
Q: How did your first customer hear about you?
Q: 你的第一个客户是怎么知道你的?
A: Our very first one, I have no idea. We just launched in Palo Alto; we didn’t do any marketing, so I assume he just must have typed in “Palo Alto delivery” into the web browser. And then after that, we did barely any marketing. I think I sent out one email to my dorm, and that was about it. It was all through word-of-mouth. And that kind of just validates how strong the need we found was when people are just talking about you, and willing to put up with a terrible user experience, terrible design, and stuff like that.
A: 我们的第一个,我不知道。我们刚在帕洛阿尔托上市;我们没有做任何营销,所以我想他一定是在网络浏览器中输入了“帕洛阿尔托送货”。之后,我们几乎没有做任何营销。我想我发了一封邮件到我的宿舍,就这样。都是通过口耳相传。当人们谈论你,愿意忍受糟糕的用户体验,糟糕的设计,诸如此类的事情时,我们发现这种需求是多么强烈。
Q: When you started, it seemed so obvious to you, you were wondering why, what the reason was nobody had done this before. What’s your answer now looking back?
Q: 当你开始的时候,你觉得很明显,你想知道为什么,为什么以前没有人这么做。现在回想起来,你的答案是什么?
A: Looking back I think the biggest thing is mobile. Now everyone has one of those in their pocket, and we saw that trend and thought what if you could design a delivery system that was entirely based off mobile, where you didn’t have to have any infrastructure, or delivery fleets. Instead of hiring drivers full-time or purchasing vehicles, what if you could tap into more of an on-demand pool of independent contractors, and only send orders to them when they have time. So that’s kind of the insight we had; everything was done through mobile.
A: 回顾过去,我认为最大的事情是移动。现在每个人的口袋里都有一个这样的系统,我们看到了这种趋势,并认为如果你能设计一个完全基于移动的交付系统,在那里你不必有任何基础设施或交付车队。与其雇佣全职司机或购买车辆,不如利用更多的按需独立承包商,只在他们有时间时才向他们发送订单。所以这就是我们的见解,一切都是通过手机完成的。
Q: Did you know you were going to be a startup, or were you just making some money at first?
Q: 你知道你将要成为一家初创公司吗,还是你刚开始只是在赚钱?
A: At the time we were all just really passionate about building technology for small business owners, and obviously this delivery thing came out of an experiment with the landing page. It was literally an experiment. We weren’t expecting anything, and it just took off, and we just went with it. And logistics was always something we were really passionate about as well, like logistics of transportation – the perfect fusion of how you can help small business owners through delivery.
A: 当时,我们都非常热衷于为小企业主构建技术,很明显,这种交付方式来自于对登录页的实验。这简直是个实验。我们什么都没想到,它就这么起飞了,我们就跟着它走了。物流也一直是我们非常热衷的事情,比如运输的物流——你如何通过送货帮助小企业主的完美融合。
Q: Did you launch the mobile site first or the website?
Q: 你是先开通手机网站还是先开通网站?
A: We started with this landing page right here which took us about an hour to launch.
A: 我们从这个登陆页开始,花了一个小时才发布。
Q: How does DoorDash stand out amongst a very competitive space?
Q: DoorDash如何在竞争激烈的空间中脱颖而出?
A: At the beginning consumer demand was never a problem, even up until now. So for us it’s just about finding a need and just focusing on serving that demand. At the beginning competition doesn’t really matter.
A: 起初,消费需求从来都不是一个问题,甚至直到现在。所以对我们来说,这只是寻找一种需要,并专注于满足这种需求。一开始竞争并不重要。
Q: How long did it take you to get incorporated into a company?
Q: 你花了多长时间才加入一家公司?
A: We launched in January 2013, and then we did YC that very summer. When we decided to take this idea through YC, we incorporated.
A: 我们在2013年1月推出,然后在那年夏天推出了YC。当我们决定通过YC采纳这个想法时,我们合并了。
Q: Where do you plan to go beyond food delivery?
Q: 除了送菜,你还打算去哪里?
A: For us when we started DoorDash, it was always about helping small business owners and figuring out how you served this for any local merchant whether you were a macaroon store, restaurant, or furniture shop. That’s still our focus; that’s our long-term vision. For now we are just focused on restaurant delivery as a way to scale, but ultimately that’s where we want to end up in.
A: 对我们来说,当我们开始DoorDash的时候,总是要帮助小企业主,弄清楚你是如何为当地的商人服务的,不管你是一家杏仁饼店、餐厅还是家具店。这仍然是我们的重点;这是我们的长期愿景。目前,我们只专注于餐厅送货作为一种扩大规模的方式,但最终这是我们想要结束的地方。
Sam Altman: Next is Walker Williams, founder of Teespring. He’s been working with YC for about a year and a half, something like that. I almost rejected him, which sounds like a dumb idea, but now they’re making hundreds of millions of dollars in revenue, so very luckily I did not. Walker is also going to talk about doing things that don’t scale.
Walker: Thank you guys for having me! My name is Walker; I am the CEO/Founder of Teespring. For those of you who don’t know what Teespring is, we are an e-commerce platform that allows entrepreneurs to launch products and apparel brands without risk, cost, or compromise. Today the company is about 180 folks and we ship tens of thousands of products each day. I want to talk to you about one of the most fundamental advantages you have as a start up, and that’s that you are able to do things that don’t scale.
I define things that don’t scale as things that are sort of fundamentally unsustainable; they will not last; they will not bring in the millionth user. Where they break, it’s usually time but it could be a number of other things. But it’s really growth strategies that won’t take you to a million users. There are three places I want to focus on today. First one is finding your first users. The second one is turning those users into champions, and the third one is finding your product/market fit.
So finding your first users: The first thing you have to understand is that there’s no silver bullet for user acquisition. You know, everybody, and this includes me when we got started, looks for that dream solution, that paperclip campaign that has tremendous ROI, some accelerating partnership that’s going to springboard you into the stratosphere, and affiliate agreement; something that solves it for you. But the reality is for the vast majority of companies and in fact for every company that I’ve had the chance to speak to the CEO of, that’s just not possible – those are unicorns. And most of the companies that from the outside look like they’ve had this dream growth curve, the reality is that those first users were impossibly hard to get. Let me tell you about the story about this ridiculously unsustainable business.
So this is Teespring in 2012 (PowerPoint). When we first launched, the business couldn’t have looked worse. It took days of meetings; we had to offer free designs, and days of revisions back-and-forth, we’d have to launch the product ourselves, we’d have to do the social media, all to sell about 50 shirts to a local nonprofit and generate about $1000 of revenue. Anybody looking in would’ve said, “You guys have to give up, this is a terrible idea.” But as time went on, those users started to add up, and I think something you have to understand is when you first launch a company, just by virtue of the fact that it’s a new product, you’re going to be bad at selling it right? You’ve got no idea what the pain points of customers really are. You’ve never sold that before. You don’t have any success stories to point to, or testimonials. Those first users are always going to be the hardest.
And so it’s your responsibility as a founder to do whatever it takes to bring in your first users. It’s going to be different for every company. The common thread that I hear is, founders need to spend personal time and effort, a lot of their personal time and effort to bring those users in themselves. It could mean a number of things – anything from sending 100 emails a day, getting on the phone and just calling as many people as you can, going through a network like Stanford or Y Combinator. Anything you can do to just get that first user. I really equate it to pushing a boulder uphill. And if you think of a smooth hill when you get started, the incline is the steepest and those first inches are the hardest. But over time as you get farther and farther, the incline steadies out, it gets easier, and eventually you reach a point where you’re at the top of the hill and the boulder starts to roll on its own.
因此,作为一个创始人,你有责任尽一切努力引进你的第一批用户。每个公司的情况都不一样。我听到的共同点是,创始人需要花费个人的时间和精力,大量的个人时间和精力来吸引这些用户。这可能意味着很多事情-从每天发送100封电子邮件,打电话,打电话给尽可能多的人,通过斯坦福大学或Y Combinator这样的网络。你能做的任何事都能得到第一个用户。我真的把它等同于推石头上山。如果你一开始就想到一座平坦的小山,那么斜坡是最陡的,最前面的几英寸是最硬的。但随着时间的推移,你越走越远,斜坡越稳定,就越容易,最终你到达山顶,巨石开始自己滚动。
And so for those first users, you cannot just focus on ROI in the sense of time. Do not expect to spend an hour and return thousands of dollars. Maybe Stanley was one of those unicorns – really incredible story. But for most of us, those first two users are going to take a lot of handholding, a lot of personal love, and that’s okay – that’s essential for building a company. The one caveat of that is, I don’t recommend giving your product for free. And there are plenty of exceptions to this rule, but in general, cutting costs or giving the product away is an unsustainable strategy I wouldn’t recommend. You need to make sure that users value your product. And you know, people treat products that are free in a much different way than a paid product, and often times it can give you a false sense of security like, “Oh we’re getting all these users; surely we can convert them to paid.”
The second aspect is what happens when you get those users? How do you turn those users into champions? A champion is a user who talks about and advocates for your product. Every company with a great growth strategy has users who are champions. The easiest way to turn a user into a champion is to the delight them with an experience they are going to remember, so something that’s unusual or out of the ordinary – an exceptional experience.
The easiest way to do this early, and again something that is completely unsustainable – it’s not going to scale forever – is to just talk to those users. People will say this all the time, it’s sort of the core tenant of Y Combinator, is talk to users. I cannot stress how important it is that you spend a large chunk of your time talking to users. You should do it constantly, every single day, and as long as possible. Today at Teespring, I’m still the catchall email address, so anytime anybody misspells “support” or writes an email address that doesn’t exist, I get that email. And so I still do about 10 to 20 customer service tickets every single day; I spend hours each night reading every single tweet, probably a little bit OCD, but that’s okay; I read through all the Teespring communities. You’re never going to get a better sense for your products than actually listening to real users. Especially in the early days, the product and the feature set you launch with is almost certainly not going to be the feature set that you scale with. So the quicker you talk to users and learn what they actually need, the faster you can get to that point.
尽早做到这一点最简单的方法,也是完全不可持续的——它不会永远扩展——就是和那些用户交谈。人们会一直这样说,它是Y Combinator的核心租户,是与用户交谈。我不能强调花大量时间与用户交谈是多么重要。你应该经常这样做,每一天,并尽可能长。今天在Teespring,我仍然是最受欢迎的电子邮件地址,所以任何人拼错了“支持”或写了一个不存在的电子邮件地址,我都会收到那封电子邮件。因此,我每天仍然要做10到20张客户服务票;我每晚花几个小时阅读每一条推文,可能有点强迫症,但没关系;我通读所有的Teespring社区。你对产品的感觉永远不会比听真正的用户说话更好。特别是在早期,产品和你发布的特性集几乎肯定不是你扩展的特性集。因此,你越快地与用户交谈,了解他们真正需要的东西,你就越快达到这一点。
There are three ways to talk to your customers. You can run customer service yourself. Up until Teespring was doing about $130-$140,000 a month, my cofounder Evan and I did everything in customer service. This is one where there’s going to be an instinct to quickly pass off, and that’s because it’s painful. Even today when I open our customer service portal, I have an emotional reaction where my stomach sinks because it sucks talking to so many users who have had a terrible experience, and it’s painful that something that you love and put so much effort into, to know you got it wrong or somebody didn’t treat them right. But it’s so important that you go through that and learn what you need to build, and what you need to fix.
The second step is to proactively reach out to current and churn customers. Churn customers are customers who have left. This is one that often falls by the wayside in the pursuit of new customers, but you want to make sure that your customers are having consistent good experiences; you don’t want to take your current users for granted. When a user actually leaves your service, you want to reach out and find out why, both because that personal outreach can make the difference between leaving and staying; sometimes people just need to know that you care and it’s going to get better. And even if you can’t bring them back, there’s a chance that you can learn from the mistakes you made that caused them leave, and fix it so you don’t churn users out in the future in the same way.
Finally, the one I’m probably most OCD about is social media and communities. You need to know how people are talking about your brand. You need to try to make sure that when somebody does have a bad experience, and they’re talking about it, that you make it right. Problems are inevitable: You’re not going to have the perfect product; things are going to break; things are going to go wrong. That’s not important. What’s important is to always make it right, to always go the extra mile and make that customer happy. One detractor who’s had a terrible experience in your platform is enough to reverse the progress of 10 champions. That’s all it takes, is one to say, “No you shouldn’t use those guys for X reasons,” to ruin a ton of momentum.
There are examples in the early days where we would mess up massive orders. We’d print out colors slightly wrong; it would be the wrong size, and it would be half of our GMV for that month. We would know we got it wrong, and the customer would be unhappy, and the instinct was to say that it was only a little bit off, not completely wrong, or that it would be fine. But the reality is you just have to bite the bullet and make sure it’s right. And the customers who are originally the most frustrated tend to turn into the biggest champions and the longest term users.
The last one I want to talk about is finding product/market fit. What I mean by that is the product you launch with will almost certainly not be the product that takes you to scale. So your job in those early days of a startup is to progress and iterate as fast as possible to reach that product that does have market fit. And as engineers your instinct is building a platform that’s beautiful, clean-code, and that scales. You don’t want to write a duct tape code that’s going to pile on technical debt. But you need to optimize for speed over scalability and clean code. An example of this is in the early days, we had a couple enterprise customers come in, sort of bigger nonprofits, and say “Hey, we really like your service, but you’re missing these fundamental things, so we’re not going to use it.” And we looked at what it would take to build out those features, and we weren’t sure if they were going to work out long-term, but we wanted to try it.
My cofounder Evan, who is our CTO and a million times better developer than I am, ran the math and figured out that if we did it the right way, it was going to take about a month to build out these features. A month for a startup – you live in dog years – a month is a year, and that just wasn’t going to do. So he actually went out and duplicated the code-base, duplicated the database, and was able to basically build a completely different product so that he didn’t have to worry about the existing users to serve these enterprise customers. We gave them the tool, they on-boarded, and generated a lot of revenue. Eventually we learned what features were core, and we integrated them into the core product. But what would’ve taken a month, we were able to do in three to four days.
A great rule of thumb is to only worry about the next order of magnitude, so when you have your tenth user, you shouldn’t be wondering how you are going to serve one million users. You should be worried about how you’re going to get to 100. When you’re at 100, you should think about 1,000. It’s one of those things where necessity is the mother of invention, so when you hit the breaking point (the Twitter Fail Whale is a great example), and in Teespring there were month-stretches where every single night the site would crash – every night. Every single person on the team would go to sleep with their phone on loud, under their pillows, so inevitably when their phones went off, we could quickly restart the server and go back to sleep; this would happen daily. But the reality is that it was worth it, and you’ll end up with these huge pain points and all this technical debt and regret, but it’s worth it just to get to that end goal and that product fit faster. You will make it work; you will survive. Those bumps are just speed-bumps, and speed is so so important early.
一个很好的经验法则是只关心下一个数量级,所以当你有第十个用户时,你不应该想知道你将如何为一百万个用户服务。你应该担心你将如何达到100。当你100岁的时候,你应该考虑1000岁。这是其中一个需要是发明之母的地方,所以当你达到临界点时(Twitter Fail Whale就是一个很好的例子),在Teespring有几个月的时间,网站每一夜都会崩溃——每晚。团队中的每个人睡觉时都会把手机放在枕头下,声音很大,所以不可避免地,当他们的手机关机时,我们可以快速重启服务器,然后再继续睡觉;这种情况每天都会发生。但事实是,这是值得的,你最终会有这些巨大的痛点和所有这些技术债务和遗憾,但这是值得的,只是为了达到最终目标和产品适合更快。你会成功的,你会活下来的。那些颠簸只是减速,而速度在早期是如此重要。
The lesson that I’ve been learning lately is that you want to do things that don’t scale as long as possible. There’s not some magical moment; it’s not Series A, or it’s not when you hit a certain revenue milestone that you stop doing things that don’t scale. This is one of your biggest advantages as a company, and the moment you give it up, you’re giving your competitors that are smaller and can still do these things, that advantage over you. So as long as humanly possible, as long as it is a net positive, you need to spend time talking to your users, you need to move as fast as possible in development, but don’t give it up willingly; it should be ripped from you.
The last thing is, we’ve created an official “How to Start a Startup” T-shirt with Sam. All proceeds are going to Watsi.org. I couldn’t miss this opportunity to sell, so if you guys want to grab one of the official tees, just go to teespring.com/startup; it’s supporting a great cause.
最后一件事是,我们已经和山姆一起制作了一件官方的“如何创业”T恤。所有收益都将用于Watsi.org网站. 我不能错过这个销售的机会,所以如果你们想买一个正式的球座,就去teespring.com/启动它支持一项伟大的事业。
Thank you.
Q: What convinced you to get into the market of T-shirt printing when there is so much competition in this business?
Q: 在这个行业竞争如此激烈的情况下,是什么说服你进入T恤印刷市场的?
A: I think there are two factors to it. First, I completely agree. From the outside, people have been telling us that this is a silly idea since day one in every order magnitude we reach. People will come and say, “This is a terrible idea. Why are you doing that?” But the reason why we launched Teespring is because we ran into a personal pain point. We identified a need and found that there were no great current solutions. I was a student at Brown trying to create a “Remember the Bar” shirt for a dive bar that got shutdown, and I realized that nothing matched my needs. And so because I knew that I had that pain point, and I knew there was market fit, and I had seen people adopt the product, I knew there was something there. And it was also one of those things where you could sort of feel the wind on your back where people were adopting the product quickly. The pain point was clearly there; it’s not a met need. So I would say that most times, great ideas start out by looking like the silly ideas, and then you can feel out whether or not there is a scalable business by how people are adopting it and whether it is possible to bring customers onboard.
A: 我认为有两个因素。首先,我完全同意。从外面看,人们一直告诉我们,从第一天起,这是一个愚蠢的想法,在我们达到的每一个数量级。人们会说,“这是个可怕的主意。你为什么这么做?”但我们推出Teespring的原因是因为我们遇到了个人的痛点。我们确定了一个需求,发现目前没有什么好的解决方案。我是布朗大学的一名学生,当时我正试图为一家被关闭的潜水酒吧设计一件“记住酒吧”的衬衫,但我意识到没有什么能满足我的需要。所以,因为我知道我有那个痛点,我知道有市场适合,我看到人们采用这个产品,我知道那里有一些东西。这也是其中一个你可以感觉到风在你的背上,那里的人们正在迅速采用产品。痛点显然就在那里;这不是一个满足的需要。所以我想说,大多数时候,伟大的想法都是从看起来像愚蠢的想法开始的,然后你可以通过人们如何采用它,以及是否有可能将客户带到公司,来判断是否有一个可扩展的业务。
Q: Are nonprofits your biggest customer base?
Q: 非营利组织是你最大的客户群吗?
A: No, today our biggest customer base is entrepreneurs who are trying to build brands and businesses. We have a little over 1000 people who make their full-time living on Teespring today via brands they’ve launched. And the other side is influencers, so YouTube stars, Reddit communities, bloggers who want to add product merchandise as a way to create a brand and monetize that affinity. Those are our two biggest markets. We still work with a lot of nonprofits, and love working with them. They are still part of our business, but just not the majority.
A: 不,今天我们最大的客户群是那些试图建立品牌和业务的企业家。我们现在有1000多人通过他们推出的品牌在Teespring上全职生活。另一方面是有影响力的人,比如YouTube明星、Reddit社区、博客作者,他们希望通过添加产品商品来创建品牌并将这种亲和力货币化。这是我们最大的两个市场。我们仍然与许多非营利组织合作,并且喜欢与他们合作。他们仍然是我们业务的一部分,但不是大多数。
Thank you.
Sam Altman: Now we have Justin Kan. Justin was the founder of Kiko, and the Justin.tv which became Twitch. He is going to talk about Public Relations.
Justin: I’ve started a lot of start-ups, but I think you’ve heard a lot of awesome “how I got started” stories, so I’m going to talk about something very specific that people always have questions about, which is press: how do you get it and how does it work? This is kind of like an abridged version of what we talk about at Y Combinator. Hopefully you guys will find it helpful.
贾斯汀:我开过很多初创公司,但我想你听过很多很棒的“我是怎么开始的”故事,所以我要谈一些人们总是有疑问的非常具体的事情,那就是媒体:你是如何得到它的,它是如何工作的?这有点像是我们在Y Combinator讨论的一个简化版本。希望你们会发现这很有帮助。
When most people get started with entrepreneurship, they think about press and being in the press as something that happens magically. They think about journalists out there, trying to get the best stories – it’s like a meritocracy – which is absolutely not the case.
Before you think about press, one of the things you really want to consider is who you want to reach, as well as your actual goal. I know when I got started I wanted to be in the news because I thought that’s what you did as an important company. It turns out that if you don’t have any goals, you’re not going to achieve them – that’s true of pretty much everything. And with press, if you aimlessly just want to be covered, it’s not going to do anything for your startup. If you don’t have an actual business goal, then it’s not a good use of time.
So there are many different goals. With Socialcam, which is a spinoff of Justin.tv, our goal was to be known as like video Instagram and be thought of in that context. When it was time to pitch to our Silicon Valley investors and influencers, we really wanted to get covered in tech press and be positioned as this new, hot social app.
With Exec one of my goals was to get customers. Exec was like a local cleaning service, and our goal was to get people in San Francisco to use it. It wasn’t useful to get national press because 99% of those people couldn’t use it. So we targeted initially local press like SF Chronicle, who would directly talk to people who could potentially use our app. TwitchTV, which is probably what you guys mostly know, is like ESPN for gamers, kind of like a live stream community for gamers. Our goal was to reach the gaming industry. Now it’s like 55 million uniques and people in the gaming industry know about it, but when we started nobody really knew that; it wasn’t a place to advertise, we were a very small gaming community. Our goal was to get people in the gaming industry, whether they were developers or advertisers, to think about us as an important place where influencers were. So we really targeted industry trades and game dev blogs. Stuff that the industry was reading.
So what’s an actual story? I think there’s a bunch of different types of stories, but these are the ones you usually see in startups. Product launches are when you launch a different version of your app.
There’s fundraising; for whatever reason, the press loves to write about fundraising even though it’s not very interesting. So like if you raise a million dollar seed round you can pretty much get that covered.
Milestones are metrics, like if you’ve achieved one million dollars a week in revenue. The company that bought Exec just announced that they achieved one million dollars a week in revenue, and it was covered pretty widely. Business stories, which happen when you’re already a successful company, The New York Times, The New Yorker, or Business Magazine, will want to cover your startup story. You don’t have to worry about that in the beginning.
What I like to call stunts – I don’t know if you guys remember, but a couples years ago, this YC company called WePay dropped a block of ice with money frozen in it outside of a PayPal developer’s conference because PayPal was in the news for freezing various developers’ account. It was widely covered because it was such an interesting thing, you know, it got them in the story, they wouldn’t have been talked about in the context of PayPal at all.
Hiring announcements: If you’re a big enough company and you hire someone really important, people will want to cover that.
And finally contributor articles which would be you writing some sort of industry overview or some opinion piece, maybe a tech blog, stuff like that.
Basically any of those things can be a story. Something that people usually don’t think about when you’re trying to start a startup is, when you start a startup, you think that everything you’re doing is interesting, but that may not be true for other people. What you really need to think about objectively is, if I wasn’t the founder of this company, would I want to read a story about what I’m pitching? So your incremental feature release or your 2.01 feature release might not be interesting just because you added “Find Your Contacts” on Facebook. You really want to take a step back before you invest the time in actually trying to pitch a story, and think, “Will anyone actually really want to read this?” What journalists and bloggers are really looking for are things that people actually want to read.
The other thing is you don’t actually have to be very original – your press doesn’t have to be original. It just has to be what I like to call “original enough.” You don’t want to be the second-coolest company to raise $5 million on Kickstarter – the first guy gets all the news. The first video game console to raise $10 million on Kickstarter was huge news because they were the first in that category, even though a lot of people had raised a lot of money for Kickstarter before. Think about your stories in the context of other things that have been written and if they’re like novel enough or aren’t something that was just written about in the news.
So one of the actual mechanics of getting the story (this is pretty tactical), if you want to get your news in the press, basically there are some really easy steps to follow: Think of getting press like a sales funnel. You’re going to talk to a lot of people but not all of them are going to convert – so you shouldn’t be upset when one individual person or reporter doesn’t write your story. The first thing is you have to think of it like a story. The second thing is you want to get introduced to any reporter or multiple reporters who are going to write about your thing. It’s much much easier, just like any business development, to get in touch with them through someone, rather than cold emailing them.
The best thing to do is go to entrepreneurs who were just written about or friends who were just covered on TechCrunch, get them to introduce you to that reporter who wrote about them. The reason that’s good is because from the entrepreneur’s perspective, the easiest thing in the world to do is to introduce you to the reporter who already wrote about them. They don’t need anything else from the reporter, and they’re actually doing that person a favor if the story is interesting; it’s not like you’re asking for intros to investors or people they would want to hire as employees. And then from the reporter’s perspective, they’re getting introduced to someone who they already vetted as interesting; they’re getting an intro from someone who they believed was interesting enough to write about, so by the transitive property, they’re going to think you’re probably interesting.
So you get an email from this guy who introduces you to the reporter, and you want to get in contact with them with enough time to get them to write a story – let them know probably a week in advance or more, because they’re not going to drop everything they’re doing to just write about your news. A lot of people, especially first time entrepreneurs, will come and say “Justin I’m launching this product tomorrow. Can you get me in TechCrunch?” That’s probably not going to happen unless you already have a relationship. The best thing to do is give yourself some lead time; get that intro in advance.
So once you’ve set a date for your news to go out – you’re going to launch a product in two weeks – you have this intro, you’ve set up some sort of meeting, and you really want to get the reporter to invest time and effort into you because there’s a sunk cost fallacy at play. Basically, the more time they spend with you, the more likely they are to write about you. The best thing to do is get a face-to-face meeting. Some bloggers don’t actually want to meet face-to-face, but like if not that, then get a phone call. The worst thing to do is just have an email exchange because it’s very easy for them to forget about it or ignore it.
The next step is actually pitch them. What I do is actually write out the ideal story that I want to see published in bullet points, and memorize it. And when I have a conversation with them if it’s in person, the conversation is structured like my outline, and they’ll be taking notes. Then they’ll go transcribe those notes into a story. So basically what I wrote will be translated into an actual story. By preparing, you can actually control more of the conversation and not forget critical things like mentioning your cofounder’s name or the awesome features in your app are.
If I’m doing this over the phone, I’ll make sure to have the bullet point list in front of me and I will make sure to walk through a conversation that includes all those things. So you have a pitch, they take notes, and they’re going to write the story at this time. The next thing to do is follow-up like a couple days or day before your actual news goes out. You want to send them an email that says “This is the time we’re launching the app; thanks for meeting; here’s collateral if there are any videos, photos, or screenshot you want to include; how to spell your cofounder’s name or your name. I just include all the information that I really care about and bold it. And then that’s it. Hopefully the day comes when you press submit on the release to the App Store and at the same time, they released their article on TechCrunch and you are famous.
So a lot of people ask us about PR firms. So I think in the beginning it’s kind of like, everything you do in a startup, you want to do yourself. And it’s actually pretty easy especially with tech press and bloggers who constantly need new things to write about. I strongly encourage people to try it themselves, and kind of get started by learning the process themselves before they hire anyone. One thing I’ll say is that firms can only help you with the contacts and the logistics, but they can’t help you know what’s interesting about your company, or – I’m never been able to have any one that’s told me what the stories that I’m producing on; they’ve only been able to give me a list of reporters I might want to contact. So you really have to be responsible about thinking about what’s interesting about your company and what you’re doing, you know, the roadmap of interesting things that you’re working on.
They’re also really expensive. I think we were spending about $5,000 and $20,000 a month, and for various firms, it’s a lot – for a startup right? It’s generally not a good use of money especially in the early days. Getting press is a lot of work, so you should really make sure it’s worth it. Getting press is like a vanity metric: It feels like you’re being successful because many successful companies like Facebook are covered in the press all the time, but it doesn’t actually mean you’re successful; it doesn’t actually mean you’re making money, getting users, or making those users happy.
Sometimes it’s a really good strategy for getting your first hundred or two hundred or a thousand customers, but it’s really not a user-scalable acquisition strategy, so it’s really something that’s just a bootstrap; you can’t just get like infinity articles written about you. Eventually people are going to get tired of hearing about your company, and usually that happens really quickly right? The pull point about news is that it’s new, so it’s pretty hard, unless you’re like Google, to get covered in the press every week. If you decide it’s worth it though, that you do want to have a regular heartbeat of news, you’re thinking about what you’re doing that matches those seven story types in the future. When I was working primarily on marketing and PR, I would make a schedule on the calendar of when we’re going to launch things and make sure to space them out, but have them appear at regular intervals so people didn’t want to forget about us and we could maximize our coverage.
You also want to keep your contacts fresh; it’s really a relationship business. Once someone writes about you, you should keep up with them because they could potentially write about you in the future. You’re more likely to do something for someone that you’ve already done something for in the past. I would try to establish good relationships with a couple reporters that you could go to for breaking news; it could come in handy later if you’re fortunate enough to have people writing negative things about you; your relationships will help you get out your side of the story.
The last thing is kind of like a golden rule or more of like a “pay it forward:” You should help your fellow entrepreneurs get coverage because they’ll help you get coverage. The best way to get coverage is really through these introductions. Whenever I’m meeting with reporters, I throw out names of other things I think would be interesting stories for them, and usually that comes back. The reporters like it because you’re helping them find interesting stories, and you’re more likely to get leads back from those entrepreneurs that you helped out.
If you’re interested in learning more about press, here are two resources that I really like: Jason Kincaid, who is a former TechCrunch reporter, goes through a great overview that covers a lot of things I talked about in more depth. And then kind of an evil resource is this book called Trust Me, I’m Lying which was written by a former marketer at American Apparel. He talked about ways that he evilly manipulated the press, but I think it’s a pretty good look into the psychology of how stories spread on the Internet; it’s valuable to take a look at. Cool, that’s basically it.
如果你有兴趣了解更多关于新闻界的知识,这里有两个我非常喜欢的资源:前TechCrunch记者杰森·金凯(Jason Kincaid),他对我所谈论的很多事情进行了深入的概述。还有一种邪恶的资源是这本书叫《相信我,我在撒谎》,是由美国服装公司的前营销人员写的。他谈到了他邪恶地操纵新闻界的方式,但我认为这是一个很好的观察新闻如何在互联网上传播的心理,值得一看。酷,基本上就是这样。
Q: When is the right time to start worrying about press altogether?
Q: 什么时候才是开始担心媒体的时候?
A: The first time I launched my first startup, you know my first products, for a lot of them we got zero attention and we didn’t really know how we could get 100 users. I think it’s a really fine way to get 100 users, and a lot of companies at YC when they first launch their product, we’ll encourage them to get out and do one TechCrunch story to get a few people to see it. It’s good to get in the practice. I wouldn’t obsess getting coverage in multiple outlets or anything like that in the very beginning.
A: 我第一次创业的时候,你知道我的第一个产品,很多产品我们都没有关注,我们真的不知道怎么才能获得100个用户。我认为这是获得100个用户的一个非常好的方法,YC的很多公司在第一次推出产品时,我们会鼓励他们走出去,做一个TechCrunch故事,让几个人看到它。参加练习很好。一开始我不会沉迷于在多个渠道或类似的地方获得报道。
Q: How much of a role did you guys play in getting the Pokemon thing out?
Q: 你们在把口袋妖怪弄出来的过程中扮演了多大的角色?
A: Twitch had this thing called “Twitch Plays Pokemon” where developers set up a Pokémon Gameboy game that was controlled by chat, so millions of people would be typing in A or B, and the character would wander around aimlessly. That was a huge news story. I think what we did was set the stage by having other news stories that someone from the BBC would Google “Twitch” and be like, “What is this crazy thing that everyone on Reddit is talking about?” They would have some context. The other thing is we didn’t come up with the idea for Twitch. That was like fortuitous, but we helped give it legs by making the company available to talk to reporters and suggesting follow-up stories, not only about “Twitch Plays Pokémon” with 100,000 people watching, but also stories about when they beat the game or when Twitch played Crystal, the new Pokémon version. We didn’t originate the story; it was the community who I think really originated it.
A: Twitch有一个叫做“Twitch Plays Pokemon”的东西,开发者在里面建立了一个由聊天控制的Pokémon Gameboy游戏,所以数以百万计的人会输入A或B,角色会漫无目的地四处游荡。那是个大新闻。我认为我们所做的是通过其他的新闻故事来做铺垫,BBC的人会在谷歌上“抽搐”,然后说,“Reddit上的每个人都在谈论什么疯狂的事情?”他们会有一些背景。另一件事是我们没有想到抽搐。这就像是偶然的,但我们帮助它的腿,使该公司可以与记者交谈,并建议后续的故事,不仅是关于“Twitch玩神奇宝贝”有10万人观看,而且故事,当他们击败游戏或Twitch玩水晶,新的神奇宝贝版本。我们并不是故事的始作俑者,我认为真正的始作俑者是社区。
Okay I think that’s it, thank you very much!

© 版权声明
点赞12 分享
评论 抢沙发