Everything We Teach at YCombinator in 10 Minutes, Startup Istanbul, Michael Seibel

01.01.1970

Everything We Teach at YCombinator in 10 Minutes

How to start a technical startup?

Two to four co-founders, at least 50% engineering. They all have to have somewhat around a year's worth of money in the bank, but that doesn't mean a year's worth of money living in a one-bedroom apartment, like in a high-rise. That means a year's worth of money eating ramen noodles and being very poor. And everyone has to have quit their job. This is what you need to start a startup. Notice no idea is required. You just need this. That's it.

Idea #

I always think that it's better to start brainstorming with your teammates. Typically, one member of the team has the kernel of an idea that'll become the company, but it's always best before that gets too solidified to discuss it with your teammates so that everyone buys in and gets ownership. We tend to like companies who are trying to solve personal problems. Every investor says that.

The real thing is we like companies where the founders know what the hell the companies are doing and know what the problem is. So if it's not solving your problem, it should be solving a problem that you're very, very aware of.

The other thing I tell founders is try to focus on daily and weekly problems as opposed to monthly or yearly problems. It's a lot easier to do something like Uber. The typical person in America has to figure out how to get somewhere three times a day. It's a lot harder to do a car sales website. The typical person in America only buys a car once every seven years. Try to focus on daily and weekly problems. It tends to be a lot more successful.

Market #

This is an area that people talk so much about. Do an hour of research. Figure out whether there are billions of dollars being made in your market and use your competitor's product. After that, I don't care.

Legal #

If you guys are interested in raising money in the US, you're going to have to incorporate in the US. That is a very, very, very simple process. For $250 and clerk.com, you can have an American company and you can raise money from American investors. Yeah, it's, that's easy. Let's demystify that.

MVP #

Where most companies will screw up is on team. Minimum viable product is the second place most companies will screw up. How can you get something into people's hands? I can't tell you — I've done this many times myself — the number one piece of advice we give to YC companies before they launch is launch. I can't tell you how many people will just sit there and iterate and iterate before users ever see anything. So when I ask a company how long is going to take to launch, I always just ask them, why does it take longer than two months? Like, I don't care what it is. I don't care if they haven't started yet. You should be able to build something in two months and get it in front of users. This is just saying it one more time: you're nothing until you launch.

Growth #

How do you achieve growth? I would say that for the typical Silicon Valley investor, this is the number one metric for determining whether you're going to raise money. It's not team, it's not past experience, it's not fancy investors — it's growth.

Typically you're going to go one of three places:

You're going to experiment with ads. I tend to like that strategy the least.

If you're B2B, you really want to focus on reference customers — the customers who you can provide amazing service to, who's going to talk to other people in their industry and spread the word about you.

If you're doing anything related to consumer, whether it's consumer services, social media, etc., usage should equal sharing. I can't tell you how many founders I've talked to said, "Oh, I'm going to make my thing go viral". Well, if you want to make your thing go viral, it doesn't mean just adding a share button. The very fact that people are using your product has to create some type of sharing. That has to be built in from day one. And so don't think of sharing as an extra activity or a side activity. Think of it as 100% part of using your product.

Press #

A lot of people think that they need to hire someone to do their PR. At Justin.TV, me and my co-founder had a little competition on how much money we'd wasted, and we'd write it up on a whiteboard. Every time we wasted money, we'd write it up there and add it up. I've wasted over $150,000 on PR firms, so I'm giving this advice because I don't want you to do the same thing.

99% of PR in the early stage you can do yourself. The best piece of advice I've been given about PR, the thing that completely changed my whole perception, is that PR is exactly like business development. When you're trying to do a BD deal, you get a warm introduction, you follow up, you build a relationship, and you provide something of value. When you're dealing with press, it's the exact same thing. You need to get an introduction from someone, hopefully who that reporter has always written about. You need to be able to structure your pitch so that it's real news — something launching, money being raised, a significant hire, a significant new BD deal. You can't just expect the profile piece. And what you need to do is make sure you treat that as a relationship. Once you have a reporter who's written about you, you've got to follow up with them. That becomes a very, very important relationship that can continue to provide value if you continue to follow up. PR: very simple, and don't spend money.

Fundraising #

Typically, companies will come to YC and ask, how do we start fundraising? And the number one thing we say to them is that if you don't need money, people love to give it to you. So how do you put yourself in the position of not needing money?

Let's not start the company with a $1 million marketing and advertising plan, right? You need money for that. Think about how you can structure your company so all you have to do is pay for the living expenses of your co-founders. That's it. Hopefully your MVP requires so little money to get up and running that you can produce some growth without needing it.

People don't quite understand that speed when it comes to fundraising is extremely important. When you have meetings with investors, you want them to be scheduled as tightly as possible, like one week, every single introduction meeting you have with an investor. The purpose of this is that, one, it creates buzz around your fundraise. Two, it lines all the investors up, so if one investor wants to take a step forward, you can contact everyone else you just met in the previous week and say, "Hey, look, we've got someone on hook here. Do you want to come along or not?" The biggest mistake we see people make is doing these investment meetings serially: I'll take one this week, I'll take one next week, I'll take one the week after that. Investors move because they have a fear of missing out — FOMO. You're not creating a fear of missing out if you're only talking to them one at a time.

Have growth. Having growth is like the solution to every single problem. The more you're growing, the more investors are going to invest time finding you as opposed to the other way around.

Four things to think about if you're having fundraising problems:

Are you growing? I should have asked, have you launched? If you have a launch plan that requires fundraising, change it.

Are you talking to the press? It's investors' jobs when they wake up every morning to find a company to invest in. That's their job. They try to trick you into thinking that their job is to sit at their desk and wait for you to walk in the front door, but that's not their job. So you need to be getting the word out about your company so that they hear about your company from someone else, not just you.

Build momentum. Make sure you're talking to as many investors at the same time as possible. Sometimes this means cheating. Oftentimes this means doing stuff like telling the investor you're really busy and you can only meet a month from now, so you can spend a whole month lining up other investors for that week. That's totally fine.

Try to focus, especially for your first money, on people who understand your problem because they've had it or they have it right now. Oftentimes customers and potential customers can be great initial investors early stage.

Operations #

The number one problem that companies have is they spend too much money. That's the number one problem they have. So spend less money, pay yourself less, get a crappier office, just suck it up and use less money. This is the number one way you can extend your runway, and it's 100% dependent on you, no one else.

The second thing, and I can't tell you how many people do this: look at how much you spend every month. It's very simple. Just go to your bank account, download your spreadsheet that says exactly every line item what you spent, and read it every month. If you're the CEO and you don't know what your expenses are, you're not doing your job. Then figure out how to spend less money. That seems pretty simple.

Hiring #

One of the goals I like to kind of instill in people in hiring is figure out how you can increase the average talent with each hire. Oftentimes founders think that they are like the smartest people in the universe. Founders tend to be pretty smart but very, very, very willing to take on risk, and so your first couple employees should be a lot smarter and just a little bit more risk averse than you are.

If when you're hiring someone you don't think you're increasing the average intelligence of the company, you made a mistake, because those people are basically your hiring advertisement. If those people are smart, other smart people are going to want to come to your company. If those people are not smart, everyone knows it. One of the things I always think about is, like, if you can't hire someone who's smarter than you, just do it yourself. Hiring a lot of people is not required.

Be fair and transparent. When you give someone an offer, you need to tell them how much stock they're getting — they should get stock. You need to tell them how much stock is outstanding. You need to tell them whether their salary is typical for what you give, and if not, be honest. A lot of the times your first employees are your most viable, and if you show them loyalty, they'll show you loyalty back.

And then the last thing is: hire slow. My second company, Social Cam, had three founders. When we sold, we had three founders, no employees. We didn't necessarily want it to be that way, but we were able to grow a product to over 20 million downloads with three people, so don't tell me that you can't do a lot with just your founding team. You can do a whole lot. I think Instagram sold for a billion and they were under 20, so hire slow. You can do it.