How to start a startup. Sam Altman tries to teach 30% of the advice given to Y Combinator companies. 720 companies. Every guest speaker has been involved in the creation of a $billion+ company. All the advice is geared towards building a company whose goal is hypergrowth. The four areas: you need a great idea, a great product, a great team, and great execution. You may still fail. The outcome is idea x product x team x execution x luck, where luck is a random number between 0 and 10,000. You can do it if youre young and inexperienced, or old and experienced. You should never start a startup just for the sake of doing so. You should only do it if youre compelled to solve a particular problem, and a startup is the best way to do it. The passion should come first, and the startup second. All of the big successes at YC followed this. Its become popular in recent years to say that the idea doesnt matter. Youre just supposed to start and pivot. Things do evolve in ways that are hard to predict. But the pendulum has swung way out of whack. A bad idea is still bad and great execution towards a terrible idea will get you nowhere. Your idea is something you will be working on for 10 years. The idea needs to be difficult to replicate. Wait to start a startup until you have an idea you feel compelled to explore. Another way of looking at this is that the best companies are almost always mission-oriented. Missions compel teams to the fanatical focus that is necessary. Theres no way to get through the pain of a startup without feeling the mission is important. People outside the company are more willing to help you. Its easier to start a hard startup than an easy startup. The best ideas often look terrible in the beginning. A way to stay on strangers couches? That just sounds terrible. You want an idea that turns into a monopoly. You cant get a monopoly in a big market. You have to start with a small, specific market and expand from there. If you come up with a great idea, most people will think its bad. This is good: it means they wont compete with you, and its ok to tell people about your idea. The truly good ideas dont sound like theyre worth stealing. You want to sound crazy but be right. You need a market thats going to big in 10 years. Never pay attention to the size of the market today. Its best to go after a small, but rapidly growing market. Customers will put up with an imperfect, but rapidly improving product. You cannot create a market that doesnt want to exist. Software is eating the world and theres more of these small but rapidly growing markets than ever before. Why now? Why couldnt this be done 2 years ago? And why will 2 years in the future be too late? Its best if youre building something you yourself need. If not, get very close to your customer -- try to work in their office. The best ideas are a clearly articulated vision with a small number of words. Any company thats a clone of anything that already exists, with a small differentiator, usually fails. Until you build a great product, almost nothing else matters. PR, conferences, building partnerships, ignore all of that. Build something peolpe *love*, not that people merely *like*. Merely like means youll probably still fail. Its better to build something that a small number of people love than a large number of people like. If you make something people love, youll get growth by word of mouth. If youre not growing organically, your product isnt good enough, and its a sign of real trouble. Dont worry about your competition. Very few startups die from competition.They die because they didnt make something users love. They spend their time on something else. Think about the really successful companies and how comically simple the first versions of their products were. The first version of Google was a webpage with a textbox and two buttons. Simple is good because it forces you to do one thing extremely well. Go recruit the first users manually. Dont buy Google Ads or anything like that. You dont need very many, you just need ones who will give you feedback every day. Do things that dont scale. Dont hire sales and customer support people. Talk to customers yourself. Make the feedback loop hours. If your product gets 10% better every week, that compounds really fast. Its a lot of work, but theres no magic. You will eventually get to a great product. Dustin Moskovitz explains why you should start a startup. Common reasons are bad: Its glamorous, youll be the boss, youll have flexibility, such as over your schedule, youll make more money and have more impact. Startups arent glamorous like the movie The Social Network. They are stressful: fear of failure is worse than a regular job, because your followers depend on you, if not for their livelihoods, then for their opportunity cost. Youre always on call, fundraising is stressful, media attention can be negative instead of positive, and if youre an employee and its too stressful, you can leave, but if youre a startup founder, youre in for 10 years if its going well, and probably still at least 5 years even if its not going well. You think if you work at a big company that everyone is stupid, but youre probably wrong. They have difficult decisions in front of them that in your fantasies of being CEO you dont have. You have flexibility: you can work any 24 hours you want! If you take your foot off the gas, your followers will, too. As for impact, you have to be a founder of a $100 million company to have as much impact as employee number 100 at Dropbox and a $2 billion company to have as much impact as employee number 100 at Facebook, and the risk is *much* lower. Brett Taylor, employee #1,500 at Google, invented Google Maps, a product you probably use every day -- great impact. Didnt need to start a company to do that. So whats the best reason? You cant *not* do it. Youre super-passionate about the idea, youre the right person to do it, you gotta make it happen, the world needs it, and the world needs you to do it. Dustin Moskovitz and Justin Rosenstein had to start Asana, the teamwork management app company, because at night they would work on this internal task manager at Facebook, and it was so clearly valuable they couldnt do anything else and had to start a company. What does it mean if we dont actually start this company? Nobody else is going to build it. The idea was beating itself out of our chests and forcing itself into the world.
Posted on: Sun, 05 Oct 2014 21:20:11 +0000