This used to be an unlisted post that I referred to people who emailed me asking for advice. I think it holds up relatively well with several years of hindsight, so I've added it to the main index of articles.
Hey, you’ve asked for startup advice! Read this first and if at the end you have more specific questions, I’ll try to answer them.
The first thing to know is that I’m really bad at evaluating startup ideas, so asking me what I think of yours is probably not going to be that useful. I might have strong opinions but they’re just as likely to be wrong as right. Two big reasons for this are that I don’t know you (your background, your strengths, your weaknesses) and I probably don’t know your space. Even if it superficially looks like I’m in a similar space, there’s lots of nuance in the fine structure of markets and I’m probably too far away to know what you’re doing well enough to give tactical advice.
If you’re in school right now and thinking about starting a company, one thing I’d encourage you to do is work somewhere first. Work in the industry you want to disrupt before trying to start your own thing. You should think of yourself as the startup, with a mission you want to accomplish, and a venture funded C-corp is just one vehicle for having that impact. In all likelihood, it’s not the right vehicle for the startup of you yet. You want to meet future coworkers, customers, partners, see the problems firsthand and understand why the field has evolved the way it has, and become bankable first. There’s a huge difference between arguing over a few hundred thousand dollars as a new college grad and first time founder, and knowing who you’ll call to lead a Series A when you’re ready to leave a company they already know.
If you’re a grad student having done a relevant PhD looking to start a company driven by that novel technology, that can count as having worked in the industry. You’ll have more to learn about how to be a company, but there may not be a “relevant industry” for you to work in, and new technology doesn’t stay new for very long so it can make sense to skip the industry experience part. You’ll probably already have your coworkers and connections to investors, anyway.
My long term mission is to develop sophisticated brain-machine interfaces that obviate the need the solve a wide range of hard biology problems as well as improve the human experience. I believe Transcriptic is the right vehicle to advance that mission most rapidly now. If there was a better way to do that, or if I was waiting on some enabling technology trigger to come along, I’d be doing something else. (Edit: am now doing Neuralink, to build sophisticated brain-machine interfaces.) If you don’t know what you’re trying to do yet, a startup is a really bad idea. If you just want to make a lot of money, go into finance. It’s not all that hard to just make a ton of cash there.
Before starting a company, there are some questions to ask yourself. These are what I’d want to know if I were an investor after my experience as an operator:
It can obviously be hard to answer these about yourself but the point is that the human elements of building a company are just as important, if not more important, than the technical and financial aspects. Like a political campaign, a startup is an MRI for the soul, and if you’re not psychologically prepared for that it’s not going to turn out well. Even if you are prepared for it, you’re still going to have a hard time.
Another reason not to ask me for advice is that a lot of questions centers on cofounders, and if I don’t know much about you or your market, I definitely don’t know anything about your existing or prospective cofounders. Cofounders are a tricky topic and only someone who you respect who knows you all well can really help there.
Cofounder disputes are most common cause of startup failure. If you can avoid having a cofounder, it’s extremely powerful for a company to have one person who can quickly and easily end the inevitable stupid arguments that will arise.
On the other hand, being a solo founder is a powerful amplifier of the stresses you’ll experience. This might not make any sense to you now, but you’ll eventually be all alone and only you will be responsible for the impending failure of the project on which you’ve spent years of your life and millions of investors’ dollars. If you can’t compartmentalize that and survive and succeed anyway, cofounders are a great idea. Make sure you know them very well before starting a company together. If you don’t, there’s a good chance they’ll make the problems worse rather than better.
Prospective founders often talk about some big potential partnership they think will make them successful. It rarely does. If you want to know what I think of your partnership, I think it’s probably a waste of your time. Big companies will drag you along and in the end will be too risk averse in ways that won’t even make sense to you to do anything. Small companies don’t have anything of value to give you. Startups in general shouldn’t partner together. Neither of you have anything yet, and trying to rub together a couple zeroes in search of a one is usually a waste of time. Always ask where the customers will come from. The only acceptable time to partner with someone else is when they’re directly bringing in customers, at which point you can just treat them as a big customer rather than a partner.
Big customers are great as long as they actually close. You should get used to not having “good” meetings anymore. Whenever you walk out of a meeting thinking it went really well, stop for a second and make sure you’re being real. Come to understand that the only meetings that go well are meetings that end in an offer being signed or a wire transfer being sent. Everything else is “not bad.” You can walk out of a positive meeting and know that it could have gone badly, and whatever happened wasn’t that. But it hasn’t gone well yet. Make sure things close. Get skeptical when they don’t.
I have this theory that nontechnical founders actually outperform technical founders. (Edit from 2018: I still think this is true for SaaS and some other areas, but not for startups that center on solving hard technology problems. SpaceX would never have worked if Elon didn’t really understand engineering as well as he does.) It’s very hard to get good data on this so I don’t know if it’s true, but if it is the basic mechanism works like this: technical founders understand how hard the technology is to get to work really well, so they can engage with all of the problems that come up. Non-technical founders can’t, so they have to select for people that just make problems go away or die. Make sure you’re evaluating people, deals, meetings, and everything else on “all-cause” success or failure. If a VC isn’t closing or a project isn’t getting shipped, it doesn’t really matter why. You probably don’t know the true reason anyway, even if you think you do. It doesn’t matter why you fail at something, it matters that you succeed. This is a big part of why people are so important. In the beginning you’ll have to do everything yourself and it’ll be up to you to succeed. After that, get people that make problems go away. If a big customer is taking forever to close, find a different value proposition or find a different customer. Going deeper than that is recipe for trouble.
Related to this, one final reason not to ask me for advice is that I often can’t easily summarize how or why something works. It’s amazing how much startup advice comes down to, “I don’t know, you just kind of do it.” There are interesting tactics to know for everything but they often aren’t enough. You have to just kind of do it.
There are many different ways to build a company. There are no right answers; the only wrong answers are ones that will make you fail. One of the first things you have to get used to if you want to start a company is that no one can tell you how to succeed. Every now and then I wish we could just hire a CEO who would come in and make us successful, but the reality is that I don’t get to be that lucky. You have to figure it out on your own, but at least it’s a lot less complex than people make it out to be. Most things are distractions. Most things people tell you you can’t do or how things don’t work some way are wrong. Stay focused, be persistent far beyond what you might initially think is smart, and realize that no one else can tell you how to succeed, and you might be ok.