My Notes on How to Start a Startup by YC
docs.google.com
docs.google.com
My point is that a lot of the advice I’ve heard about startups in general is framed (intentionally or unintentionally) as a series of directives that will optimize your chances of success. Moreover, no other additional context is provided to help you frame the advice, such that you can tell whether or not it’s applicable or appropriate advice to follow. The lack of context coupled with the existence of so many blatant counterexamples and contradictions prevents the advice from being useful.
(Note: I edited the last paragraph for clarity)
Some people smoke cigarettes their whole life and live until old age without ever getting cancer. Should we conclude smoking does not cause cancer just because there are counter examples?
As you said, the advice is meant to maximize the chance of success. It is still not sufficient or without exceptions.
In the end, advice is only valuable if you put it through your own filter and use your own judgement to make the best decision.
Because there's no set of instructions that will work for sure.
What one needs to success with a startup is several of these: work hard, work smart, connections, piles of money, idea, marketing, good timing and luck.
You can have most of them and not triumf, you can have only a pair and succeed. With none it's not possible. The ones that seem to matter the most are the ones you come with (money/connections) and the ones that you can't affect much (good timing/luck).
I’m not sure marketing belongs in the list though. It’s critical but it’s a part of execution and overlaps with several other items on the list. Just stood out to me.
I’d also challenge the last paragraph. I don’t think anything on this list matters as much as working hard. Literally every single successful business owner I’ve ever known has worked their ass off, usually for many years. For me the correlation is 100%. That’s not to say it’s sufficient but it seems necessary and the most equalizing force on your list [1]
1. Although I sometimes think in our fetishization of hard work, we ignore the fact that not everyone can grind away for years on a speculative venture even if they wanted to. I know many wonderful people who just don’t have that personality. Telling them to “outwork their competition” makes as much sense as telling someone to change their favorite color or decide to love a food they hate. It’s just not how they’re wired and that’s OK.
There are plenty of people working as hard as they can, that doesn't mean their starup will work out. They might be working on the wrong problem, they might have bad sales/virality, they might be doing a great product but too soon/late for the market, someone bigger might be fighting for the same market, they might just be unlucky.
Without the hard work there is no startup, for sure. But fetishizing it leads to the whole "you were not working enough/you didn't want it enough" toxic concept of people only getting what they deserve. Life is not fair, good people have bad things happen to them from time to time, bad people sometimes have good things happen to them, hard working people don't always succeed.
- Peter Thiel
[edit - disclosure: I was one of the presenters in this series.]
What’s the point of having access to a ton of great advisors if they’re constantly all telling you conflicting things? Is the answer just to go with your gut? If so, why do you need the advisors? Is it more to prime your decision making with possibilities? Or is it more for tactical execution advice once you decide on a direction? Or something else?
Here’s my conflicting bit of advice. ;)
The point is to get the best head start you can, and to get it by knowing many ways other people succeeded. They will share some common themes on how to succeed, and maybe more importantly, common themes on how to accidentally fail.
You’ll always get conflicting advice no matter what the context is. The same thing happened to me before and after getting my startup funded- advice was all over the map. And it has also happened in academics and at jobs and among my friends and family. Ask 5 people anything, you’ll get 5 answers. People simply have different points of view, and there are usually multiple right answers.
It’s a feature because it’s important to understand that there isn’t a right answer but a continuum of possibilities that you get to influence. It’s important to understand how people construct and use narratives about their successes. It’s important to see clearly that there’s some luck involved, even if nobody admits it.
Customers, by the way, will give even more wildly conflicting information, so being able to sift through conflicting information is a good skill to have.
Going with your gut will only work if your gut knows how to make good business decisions, which a few people have but not everyone. Software engineers, for example, have guts that tell them that writing awesome software is the one thing that will do the trick, the product will sell itself. Advisors will all tell them that marketing is more important, but they will all disagree on how & where to do the marketing.
There are other side benefits to advisors, including: having influential people rooting for you publicly, having help meeting new investors, getting tactical advice on business and management problems, having a support group of people who’ve gone through some of tough things you’re going through because friends and family don’t understand. I think there’s more but you get where I’m going...
It's how investors do technical diligence, it's how executives of large companies make decisions, it's a part of the peer review process in science.
Advisors don't exist to tell you how to do your job. If that was the case they'd be your boss, not your advisor. They are there to help you see the full picture, and assess different perspectives, and challenge your thinking. But ultimately you make the call
This is a quote from Tren Griffin about the stock market but it holds true for startups as well. Successful people who help out entrepreneurs often take their survivorship bias'd situations and assign single factor causation to them. Thus you end up with conflicting advice because "sometimes it works and sometimes it doesn't".
Generic advice is not hurtful, it's just not contextual. It's your job (the founder) to decide whether it can fit your context or not.
The existence of counterexamples and conflicting advice doesn't mean that advice isn't useful, conflicting advice is just a symptom of startups being uncertain. With a large enough sample size of advice, you can start to make sense of it, understand what's good and bad advice, understand how people's life experience and incentives may color their advice etc. this makes it much easier for you to turn that advice into decisions and actionable steps
If you think about it this way then more advice is better. It's just up to you to synthesize that advice and decide what to do with it
I guess what I’m saying is embrace ambiguity in the advice more clearly. Don’t say “Do X” and “Do the opposite of X”, which together give you no ground to “synthesize”.
Another approach is to describe the effect of “Doing X”. Maybe building fast means you can get to market faster. Maybe building something nontrivial means your competition won’t gobble up your market. But you, the founder, are the arbiter of that decision/tradeoff.
ISTM that many startup-advice-givers like to create these short memorable platitudes instead of actually describing tradeoffs and decisions that the founder must make.
To be clear, I'm not saying that there's not good advice floating around. There is. The trick is knowing how to separate the bad from the good, and as someone who's new to the startup world, that's pretty hard (impossible?) to do.
Youtube was a hard pivot from a video dating site.
Furthermore, they are extremely biased towards start-ups, which is understandable. You could not easily make a worse financial decision, statistically speaking. But they need most of us to do it for them to stay in business.
I hope some of you find this useful (:
Many suggestions are practical with case studies.
I'm not sure if I can agree with that. In general that may be true, but I would have to say most people do something they don't like 'just for the money' at some point in their life. Something like working as a kid for minimum wage in a store. It's most likely that you are just working there for money.
Also, startups, where startup means creating a new unproven solution intended to achieve exponential growth, are unfathomably hard. It's hard (in my experience) to stay motivated in the face of startup hardships when one's primary objective is money.
And it all depends on your financial goals.
Want low 6-figures a year? Get a career.
Want medium 6-figures a year? Bootstrap a business.
Want 7-figures a year? Take VC money.
And yes, the success rates get progressively worse, but if your goal is to average $500k-$1M+/year over the next 5 years, your chances of making that as an employee are probably far less than your chances of making that owning a startup.
Exactly. Startup is the only way to make money, while being reliably preserved from management errors which you can't prevent or compensate for.
You can join an early Facebook, but that's a luck game, not a systematic approach where you'll actually influence chances.
It's a term without a strict definition.
- This pressure to have a co-founder. I see that quoted as a major reasons for why startups go bust but also necessary for doing anything meaningful. I can imagine there being truth to both statements. I've seen startups go bust in my own network on both sides of the divide. But I feel there isn't enough advice out there for folks who are doing it Solo - that's still a valid design pattern for many startups in the past or present.
- The whole idea of go big or go home - lot of the advice seems to assume you want to be really big one day. There doesn't seem to be enough advice for someone wanting to be a 10M-100M dollar business as their main milestone, not as a path to hit a billion dollar valuation. That kind of assumption results in very different kinds of design patterns, often borrowed from other billion dollar startups that may not make sense to your own little venture in the beginning.
I wish startup advice wasn't universal as every author or speaker would like you to believe - there could be a system where it was doled out based on the current phase and situation, or at least recognizing the fact that there isn't one size fits all for something as varied as starting your own venture.
I don't know why people think otherwise. Do you think venture capitalists are your friends? Pretty sure they're not!
They want two cofounders to test for sociability like they say, but also because it lowers their risk (if one develops problems or is a loser, the other can step up), and actually increases their leverage over the founders. I mean, one founder can always be some stubborn type who says "no." It's easier to get to "yes" when you have two people making choices -this is basically an algorithm. The sociability thing is reasonable and a good test in many ways, but I know about as many solo founders who made it as I know founder-teams, so the advice is obviously wrong for some. Distribution is about what you see in big firms: plenty of Zucks and Bezoses out there to balance out the famous dyads.
The "go for unicorn" thing should be familiar to you as a trend follower. I believe the old Turtle systems would have a majority of initiated trades as expected to lose money (aka start up failures -most of them fail!). You make up for the losing trades by doubling up and riding the winners. It's also vastly lower risk to do it this way than try to get a lot of medium sized deals/trades: the longer the trend rolls, the more you are certain the trend is real. The only way to make money on lots of little/medium moves is stat arb; a trading strategy where you can hedge, and VCs can't hedge.
The founder's utility function is vastly different from that of a VC. VCs optimize their utility; not yours. You're just one of many dice they roll.
If you want founder advice, get it from non-VC founders who succeeded at something, or people who have tried to do what you're doing. Otherwise: you are the product.
Personally I don't mind the idea of going lean in the beginning, which I think applies fairly well to early founders. The current Silicon Valley notion of throwing money at it to generate enough revenue + growth, without potentially having profit for a long time is a bit uncomfortable to me. I have a hard time distinguishing between businesses that are re-investing their income for growth and no profits (like Amazon) vs. those subsidizing their growth from VC money and having little chance of being self-sustainable in the long run (e.g. I don't know where Uber will land).
Ideally, I would like to do a venture where profits do come in even at smaller unit sizes and you can test that before you decide to go big. I am not sure if there is a term for that or if anyone thinks like that.
PS: like your blog and background.
I'll say it a different way: VCs don't give a shit about the company _ever_ making a profit. They give a shit about ther VC making a profit. It's not the same thing at all! VC makes a profit if they invest at good valuations and sell at much higher valuations when the company goes public. The company doesn't have to be profitable! For all they care it will never be profitable! Pets dot com made some VCs a bunch of money!
(thx for kind words -good luck with your startup!)
VCs are optimizing for the unicorn, and there is nothing wrong going for the non unicorn either. It's just you shouldn't go for VC financing. If you make a business that hires 50 people and throws off $10 million a year to you, you'll probably make more than most founders ever will anyway after 3 years of that business.
Overwhelmingly the greatest outcomes are produced by companies with founders at the helm for long periods of time. The best VCs in the industry know this well, with few exceptions. If you're dealing with VCs that routinely like to replace or marginalize founders, you're dealing with either a subpar firm or one of the couple old bureaucratic dinosaur firms that occasionally are prone to 'IBM thinking.'
Examples: Amazon, Microsoft, Apple, Google, Facebook, Alibaba, Tencent, Baidu, Netflix, Salesforce, Qualcomm, HP, Dell, nVidia, Sony, Intel, Oracle, Airbnb, Twitter. Even IBM, at its most successful it was run by the Watsons.
It's in their interest to be both nice to you, and strike a good deal with you, if they want to do business with you at all.
I wouldn't want to work with a VC who couldn't be nice to me. To be firm or offer pointed criticism is fine.
In business people can be perfectly nice and at the same time completely ruin your life. In fact, that's standard practice.
I'm telling you to be wary of VC advice on how to start a startup, basically because VC advice is entirely self serving; it's not designed to help you. It's designed to help them.
This type of comment isn't really productive to make. Yes, I read what you said. Because I replied to it.
My point is that I wouldn't work with somebody who wouldn't treat me reasonably-- I wouldn't accept their VC money.
I don't understand how VC advice can be self-serving, though, since in order for them to get a return on their money, your business has to take off. And you don't give up 100% of the company. So you win too. I view it as mutual, in that regard.
If you lose, they lose money. If they invest in you, they don't want to lose.
> I don't understand how VC ...
And you also tell me you don't like being told that you don't understand. Which one is it? Go read what I said again. Now read the following.
You are one of N (N being large) investments by VC. Your utility function is to increase your probability of getting rich. Their utility function is to increase their probability of making bonus. How on earth can you assume you are optimizing the same thing? You are not. You're just one of 100 or 200 dice they're rolling. Your success is completely meaningless to them; they want the loaded die, and they want to be able to identify it and bet the house on it. You might have a great idea for a business which makes YOU rich, but if it doesn't fit their model, you're SOL with VC. You may even have a great idea which makes the VC rich. They're not optimizing for your wealth here either: they are optimizing for THEIR wealth, and they will happily take your company, your idea, and all your profit and leave you with jack shit. As such, taking advice from a venture capitalist is like taking advice from a financial advisor that works in a boiler room. Guess what? They don't have your best interests in mind either! They just want to make bonus. Just like the VC. You are product to them.
If you've been around the block a few times, you'd know this. I'm trying to tell you how it works, but you keep telling me VCs are your friend and everything they tell you is the truth. They're not! They're almost all scumbags! You should use them to achieve your goals, and that's it! And only a fool takes what they have to say about founding a startup as some kind of revealed truth. It's just propaganda to make their job easier.
Thank you for finally highlighting something pivotal in our discussion: the shared interest (or lack thereof) in your success. Like I said, it doesn't totally add up to me. If you remain majority shareholder in your business, it seems hard for them to walk off with more than you would at the liquidity event (e.g, your company is acquired, an IPO, etc).
The truth is probably somewhere in the middle. They're not 100% against your interests, but not necessarily 100% for them, either.
I've seen super terrible things happen to people. Sometimes you have to jump up and down and make a lot of noise that "notes on how to start a startup by YC" should be taken with much salt.
Anyway :allies:
But so far this one has done a good job at that. The first thing they do is even try to convince you that entrepreneurship isn't the most profitable path. The statistically best path is to work at an early stage company. But if you have a certain kind of insane resolve, then they'll guide you down it.
I think 'partial' success does get celebrated though, sort of like Reddit. Again, this is something seed investors do far better than Series C investor types.
The patterns are still pretty similar, as a good deal of this covers the path from idea to "ramen profitable".
That's just them, being self-interested since they need employees for their founders' startups more than they need new founders.
Realistically, the most profitable path statistically is to work at a FLNG or decacorn.
I didn't know that users could change the title of their submission. How is this possible or is this actually a post by YC?
If latter, a hint would be great that this might be YC promoted content and not actual user content.
Folks probably downvoted you because they deemed your tone unwelcome here.
Still strange. Why should an admin change the submission of a random user if it doesn't breach any of HN's rules?
Maybe because it's not a random submission by a random user?
Guess I must be wrong but would love to hear what that admin who did the change has to say.
I skimmed the first pages and most of the advice is neither wrong or bad. Still, reading a lenghty 57-note in order to learn is wrong.
You learn by doing. If you want to start something stop procrastinating on HN and execute the first step: found the legal entity for your endeavor. This will keep you busy for the next days, and you learn.
Btw, you don't need to have a good or any idea or co-founder now. This will all come. Just start, make mistakes, stop reading random advice.
Edit: Don't downvote if you disagree, downvote if a comment doesn't add anything to the discussion.
Find the nugget of your idea. Do something to get market feedback.
Eg your idea is Uber for pet food - go interview people leaving a big pet food store and ask them if they’re interested. Old school market research.
Or say your idea is a sports news site for minor teams. Set up an MVP that serves for your local team and promote it, see how many many people go to the site. MVP could be a stream of article links from other sites, or perhaps better a newsletter via email. On site have sign up box for other teams.
Market feedback is key. Once you have a good signal, continue building and gather more feedback.
I predict this posts hits 100 points and then it crashes. (Currently at 31.)
Welp, I'm over 22, guess I'm never starting a company.
Appreciate the share.