Startups in 13 Sentences
paulgraham.com
paulgraham.com
Rule #1 -- Pick good cofounders. I picked no cofounders, based largely on the fact that that I couldn't think of anyone who would be interested in joining me who wouldn't dramatically lower the average founder quality. (Ok, that's being a bit facetious -- but I'm working in an area where domain knowledge is extremely important, so there are lots of wonderfully talented people who would be utterly useless to me as cofounders.)
Rule #2 -- Launch fast; and Rule #3 -- Let your idea evolve. I started working on tarsnap in September 2006, and I didn't open tarsnap to the public until November 2008. Over that time I ironed out some technical details behind the scenes, but the largest change in my idea was going from planning to charge $0.25/GB to actually charging $0.30/GB. Since then, I've listened to my users and occasionally added an unplanned-for feature because of a request (or more often, several requests for the same thing from different people); but most of the ways that tarsnap has deviated from my original plan have simply been in the ordering of when I implemented which features. The vision behind tarsnap, from 2.5 years ago -- an online snapshotted backup service with a tar front-end, a heavy emphasis on security, and a linear pricing model -- hasn't changed at all.
Rule #10 -- Avoid distractions. I've remained as FreeBSD Security Officer while working on tarsnap, and I don't regret it in the slightest. Yes, it has taken time away from tarsnap on occasion; however, it has also allowed me to further enhance my domain knowledge, and has brought many people to tarsnap (one person recently described tarsnap as being a backup system with "a good pedigree", which I'm presuming refers to my open source security background).
Now, for all that I've broken these rules without regrets, I still think that they're good rules -- in most cases. But I think there should be a Rule #0: Realize that there are always exceptions, and understand why these rules are usually correct instead of applying them blindly.
And although I applaud your efforts and accomplishments at Tarsnap, I must say that you base your argument on significantly less than 100 startups.
I wish I could agree with you here, but I'm afraid I've seen far too many comments along the lines of "you're violating pg's rule XYZ, so you must be doing something wrong". Yes, most people here probably understand that pg's guidance doesn't apply to every situation, but I still think it's worth reminding people occasionally.
I don't believe this sort of comment is helpful. For one thing, "X works" and "Y can work, too" are not two bits of advice which are in opposition to each other, and shouldn't be treated as such. (Even when X = "Startups should have multiple founders" and Y = "Actually, you can get by with one person if...")
For another thing, evaluating expertise by weight of incomparable circumstances is not logical. If it were, we'd all be taking advice from McDonalds on how to sell software, as the sum total of all customers of every startup company which has ever been mentioned on this site would not equal the number of cups of coffee McDonalds sold worldwide yesterday. Now there are many things to learn from McDonalds, but willy-nilly application of their ideas because they have mass on their side is a bad idea. (The sweet spot for pricing is below $5. Advertising is less about motivating people to buy in the near future than building your brand image for life. Literacy is a nice attribute to have in an employee but is not necessary if you exercise proper management.)
Scribd: 5M daily uniques
Justin.tv: 3M daily uniques
last.fm: 2M daily uniques
It might be satisfying to include some reference lines on your daily user graph, so you can celebrate when you serve more pages than, say, McDonald's sells fish sandwiches, or Wal-Mart sells Spiderman action figures, or whatever.
Of course, this is a meaningless comparison, because the number of people that Scribd or Justin.TV reach on a daily basis is more closely correlated to the number of users they have than the number of customers.
On the other hand, PG has also spent the past 3 years funding different startups and must have some decent idea about what is more likely to help a startup.
Same goes for you. I would assume that you are more experience in Bingo software then other programmers even if they are better programmers than you are. But in the field of Bingo, you would probably crush several highly talented developers.
Why?
"Weight of incomparable circumstances" probably.
I am trying to write a post on my blog about what I call the "solo entrepreneur blues", which has to do with one of the biggest mistakes I have seen by entrepreneur friends: start alone and trust no one. I'll post a link once I have the tone right.
Touché!
start alone and trust no one
I'd say that's a bit strong to describe my situation. I started alone, yes; but it's more a matter of not needing anyone than it is of trusting no one.
Those 3 are also quite related - while I still had cofounders, we'd kinda prop each other up during the low points, so we didn't get demoralized, so we didn't give up. But when my last remaining cofounder quit, I tried to keep it going for 2 months, but the emotional burden was far too much. I couldn't do it - if you can, good for you, but I think that there's a lot of truth in these rules.
Also - I think there's too much wiggle room in "Understand your users." I don't have it listed in either column for my startup, which tells me that we probably didn't, but most startups won't have the emotional distance to realize that. Even now, I'm not sure whether we did or not: since getting out to Silicon Valley, everyone I've told about my startup says "Oh, that's a cool idea, I wish you'd been able to finish it." Maybe we should've moved out west before quitting. ;-)
I think it would be very interesting to chart the success rates of sole-founder startups vs. whether or not the founder had a graduate degree: I get the feeling that the sort of struggles single founders often go through are very similar to the sort of struggles graduate students go through when they've been working for X years and still don't have a thesis in sight. If someone can work quietly in a corner and churn out a thesis after a few years, does that make them more likely to be able to churn out a startup by themselves?
I think there's a lot of truth to it, but it can also be misleading. A lot of these solo-Ph.D startups just chug along for years at ramen or small-business profitability levels. They work, they provide a nice living for the founder and a few employees, but they don't grow. Whether or not they survive seems to depend on how independent-minded the founders are, because the financial returns are often less than the founders can get elsewhere. Anyone with a Ph.D that has the work ethic to run a startup ought to be worth at least a couple hundred grand a year in industry.
There're a lot of other benefits to having multiple founders besides the "don't give up" aspect. Ideas tend to become better when they're bounced off multiple people, since the dumb parts of the idea get weeded out and the clever parts get amplified. Other people tend to take you more seriously when you can say "we" instead of "I". You learn how to compromise and accept other's input early, which is absolutely crucial once you have employees. You get a stronger company culture, which always seems to evolve out of the interactions between founders.
Actually, you seem a lot like my old boss - both math Ph.Ds with an independent streak, except you're half his age. He was a good guy and very, very smart (finished a Princeton Ph.D in 3 years). But the startup I worked at for 2 years, which is now 8 years old, is still limping along and hasn't taken off. And I really think that if it'd had 2-3 equal founders instead of 1 majority and 2 minority founders, it would've done much, much better.
Ah, but that requires a completely different work ethic. To an academic, living on ramen while solving interesting problems is far more interesting than getting rich.
Ideas tend to become better when they're bounced off multiple people
That's what users are for. :-)
But if that's your goal, then likely that will be what your startup gets. Interesting problems, few customers, and not much money.
There're lots of Ph.D-founded startups in that category. My previous employer was one. Symbolics is another high-profile example. Lots we haven't heard of, since they're working on interesting problems and not getting rich.
If that's what you want, go for it, as long as you're bootstrapping (it's irresponsible to take other people's money when you have no intention of making some yourself). Just be honest with yourself and any future employees about it. I'm still a bit bitter that I was sold on visions of Bloomberg and revolutionizing the financial industry while the reality was more about working on the boring parts of problems that were interesting to my boss and only vaguely interesting to customers.
Partly because "we" (HN) are biased toward consumer-facing web startups, which are typically not the sort of companies started by academics, or people with extensive knowledge of a non-web industry. There are many startups out there that "we haven't heard of", but still end up solving valuable problems, and quite a few of them get rich as a result.
I doubt there's a strong correlation here. When you're working on a thesis, you're still within some academic community, have a rough deadline, and have a pretty good sense of what's going to fly or not. Moreover, the progress you make on a thesis is inherently satisfying. You can stand back and look at even a page or a paragraph of work with pride.
When you're working on a startup, you spend much of your time working on things that don't work yet or will look and feel like crap for a while. A lot of the problems you solve aren't going to be intellectually interesting, as they would be for a thesis. Which means that there are precious few rewards to be had from the pre-launch development process by itself.
I tried to go solo for a while, but found myself crushed by the lack of feedback. Then I paired up with someone (remotely), and if we weren't pair programming, we were at least showing each other what we'd done several times a week and giving feedback. Not only is this immensely helpful in terms of sustaining motivation, but merely having someone else who thinks of things you don't is incredibly valuable.
Quality is a vector, not a scalar. What's the average of Steve Jobs and Steve Wozniak? I don't think that's a meaningful question. But the vector sum of their skills was a startup with amazing technical and artistic skill, product focus, business aspirations, charisma, and salesmanship.
And I'd like to help you become ramen profitable, so let me demonstrate what I mean by putting on my Steve Jobs hat [1] and telling you a story.
I'm in the market for something like tarsnap, so i was very excited to hear about it, for the first time, in this random HN comment 23 days ago:
http://news.ycombinator.com/item?id=458793
...because the product sounds great. I may sign up any day now. But I distinctly remember that, after reading this enthusiastic blurb, I spent several minutes trying to figure out if tarsnap had any kind of decent security model. I remember this because I literally laughed out loud when I finally managed to figure out the answer, after clicking through more than one link and reading a few blog posts -- tarsnap was designed by a cryptographer! Indeed, now (but only now) I learn that it's even better than that: You're the FreeBSD Security Officer, which presumably involves a great deal of practical expertise in reviewing and fixing insecure software.
Why isn't this information on the tarsnap homepage, next to a "buy it now" button?
---
[1] It's not nice to go all Steve Jobs on someone without being asked, so I'm deliberately toning down my Jobs impression. This is like a 2 on the Steve Jobs bluntness scale.
Because the tarsnap web site sucks. (Hey, you did ask!)
Until very recently the vast majority of people interested in tarsnap were people who knew who I was, had seen my work on FreeBSD, and read my blog -- so putting information about tarsnap onto my blog was only natural. Now tarsnap is starting to gain recognition as something beyond "that thing which Colin is doing", it's probably time to revisit where and how information about tarsnap is presented.
Most likely I'll do this myself, though -- even aside from the issue of money, I find that doing tedious things like web design myself (a) is a useful learning experience, and (b) allows me to think about issues a bit more clearly.
I would say you could easily get a web/design/biz cofounder who could dramatically change your ramen and also enjoy a few of those tasks you have been putting off for a while :).
I agree entirely. I did say that I was being a bit facetious. :-)
cperciva has a bit of a history of surprising people with his credentials: http://news.ycombinator.com/item?id=35079
PG says "The essential task in a startup is to create wealth", that pretty well rules tarsnap out as a startup, I guess.
It does? That's weird -- I thought tarsnap was creating wealth.
I think plenty of the 'wannabe' startups here are a bit like kids 'playing house'. If you say hey, it's just a blanket and two couches they get mad at you for breaking the illusion.
If you had a design/marketing/seo cofounder that was willing to hunt for sales/bizdev opportunities, domain knowledge for that person doesn't have to be THAT deep and would bring huge benefits.
"The vision behind tarsnap, from 2.5 years ago -- an online snapshotted backup service with a tar front-end, a heavy emphasis on security, and a linear pricing model -- hasn't changed at all."
Is that a good thing, do you think? I'm not saying it isn't, but I think founders who don't experiment/explore generally succeed more than those that don't. (See the 17 out of 26 stat here) http://www.unionsquareventures.com/2007/11/why_early_stage.h...
I think your rule #0 is dead on-- there are lots of exceptions and there's no formula. But I'm pretty convinced that your startup would be more successful if you hadn't broken the "rules" that you did.
I think you might be overstating things here. I've found my domain knowledge to be very important when selling tarsnap -- being able to convincingly explain why tarsnap is secure and efficient (and why the competition isn't!) has gained me a lot of customers.
From a personal perspective arrogance blinds you to opportunities and your own shortcomings. From an interpersonal perspective it makes otherwise excellent people less interested in working with you.
Who would you work with?
Someone who is smart; who doesn't hesitate to tell me if I'm wrong; and who doesn't get offended if I tell him that he is wrong.
If arrogance makes people more likely to speak up when they see something which is wrong or could be improved, I'd say that's a net positive. False modesty has merits in some circumstances; but a startup is not such a circumstance.
You're awesome. Your skill speaks for itself. Being arrogant gets you nothing except a reputation for being difficult to work with. That's all I'm saying.
Of course, YMMV. I'm just some guy on the internet.
False modesty is not the opposite of arrogance. Arrogance is not defined as that thing which makes someone more likely to speak up when something is wrong.
Arrogance (or modesty for that matter) is more to do with the mind then the mouth. By most people's definition, arrogance is a sort of emotional weakness that impairs judgement in various ways. Very few people who use the term equate it with outward honesty. The contrary, many would probably equate it with inward dishonesty.
I am not accusing you of arrogance myself. I do agree that many of those things that people use to diagnose arrogance might actually indicate steadfastness, confidence, straightforward honesty or some other mostly positive attribute. But as the argument turns in on itself, people's ability to diagnose each others' emotions is actually fairly evolved. Failure to recognise the possibility that they might have a point, might be..arrogant.
Most people should not have that arrogance because most people are not Caesar. Thing is, if you are Caesar, you sometimes have to be arrogant to mediocre people (especially bureaucrats) because they will crush you if you're not. But it's hard to draw the line because it's so few people are true geniuses and it's hard to know for sure if you are one.
Being a genius (as far as I can tell, because I don't know if I am one) seems to be tough in this world. It's not like being a star athlete, because in that case the best know they're the best, say so, and bet their future on it. That is not the case with geniuses. You can't take out huge loans against your future wealth in order to buy status and get laid in high school, for instance. I'm sure there are Silicon Valley billionaires who wish they could get back a part of their nerdy lost youth, but it's impossible. On the other hand, it's much less likely for a genius to have the balls to bet on his brains, challenge the curriculum, and get people excited about his future the way athletes do. I guess it's because, at the end of the day, society doesn't like nerds as much as it likes athletes, and because people like to think they're better than others. In this case, it's old people who think they are better than young people, rationalizing that they are smarter.
So.. 14. Sell as soon as possible. Find out what people will buy for. Never stop selling.
I feel guilty that this just doesn't do all that much for me. I do get some satisfaction from making a user happy, but about as much as I get from eating quality ice cream. I get far more enjoyment out of solving a difficult technical problem.
I just think there are a lot of barriers to entry for folks who didn't go to Top 25 schools (or drop out of one) and as a result, they read stuff like this, get excited and think "I can make it!" without realizing the folly of their poor pursuits.
hacker news, reddit etc
I think this is one of the most difficult rules to abide by because a lot of very talented people say they would like to start a company, but never will. When it comes down to it, they'd rather be unhappy at their job and receive a steady paycheck than take a risk and start their own company. Do you lower your standards and start a company with someone who may be talented, but may jump ship at the first job offer elsewhere?
It takes time to find out what people are really made of. From my experience, people who say they want to start their own business or be their own boss really just want to be a mid-level manager. Other people have the "Entrepreneurial ADD" and will never be able to stick to one idea from beginning to end. Others get scared of the prospects of late-night coding and early mornings, and decide to go to grad school instead (how does that make any sense????).
What's the solution for finding good cofounders? I think at some point everyone comes across someone who would be a perfect business partner, but has different ideas. Start working to form a business with that person regardless. It's likely that the business will evolve into something that neither cofounder originally anticipated. So getting everyone on board with the same idea may not be all that important, you just need to be able to work together.
This explains why someone like myself can start a startup and not feel embarrassed when they meet real wunderkinds (specifically CS gurus) along the way. On the other hand, maybe this means that good hackers aren't necessarily CS geniuses? http://www.paulgraham.com/gh.html
http://www.google.com/search?q=site%3Apaulgraham.com+%22most...
http://www.google.com/search?q=site%3Apaulgraham.com+%22prob...
etc
If the difference between expectations and reality is what makes us unhappy, then lowering aspirations to zero is what would make everybody happiest. I don't believe that it would.
Creating the impression that it is possible to be sucessful with your own startup cannot be false reality. Possibility is never reality, and founding a startup isn't taking part in a ponzi scheme. Some people being successful doesn't lower the odds of others to be successful as well.
There's no risk in failure either, you can always go and work for someone else, and personal bankruptcy isn't _THAT_ bad.
I don't think PG is giving anyone false hope that they'll hit the lottery - he's just offering an alternative. And as everyone has said tons of times before, your chances of making a million by starting your own business are orders of magnitude higher than winning the lottery.
most of them would be heartbrokenAnd the average age in YC is 25. These aren't young kids.
Funny, but this is how I feel about my 11 year career spent with 3 huge corporations - that it has been largely a waste of 40-50 hours of my life every week. I've made 6 figures most of my career, and I've had plenty of time for my family. I'd trade the salary and benefits for something I was passionate about if I felt I could do it without being irresponsible to my wife and child.
I suppose the grass is always greener on the other side.
My sentiment is that it's worth trying when you're young because you might not get a shot when you're older.
It's not false hope.
- Customers: Understand them, improve their lives, get them to pay regularly, get them to spread the word - Costs: Keep them as near to zero as possible. That will enable you to be profitable even with minimum number of customers
During the first year of your startup don;t worry about anything other than the above two. Special warning: Dont spend anytime on external fund raising, fund raising is the most distracting thing for startups and something where you have least control and very low odds, unless you accomplish something "of value".
If you can get thru the first year with a complete focus on Customers & Costs and hit break even by the end of first year. I can guarantee that you on the way to build a successful company (not just a good startup)
Even so there were several things that seemed new to me. Let me check.
I never quite realized till now that a version 1.0 was (or could be) nothing more than a pretext.
The rectangle metaphor and the point about one side being easier to change is new; I only thought of that a couple weeks ago.
The greater difficulty of lying to yourself when expanding userwise is something I don't think I've written about.
I don't think I've mentioned that founders' own standards for customer service are artificially low, or that customer service is mostly a way of studying users.
I don't think I realized before that being cheap was interchangeable with iterating, or mentioned that a culture of cheapness keeps companies young.
I don't think I've written about how paying distractions are so much worse than other kinds, or that it's specifically because they work like interrupts.
I didn't realize till recently that the right metaphor for a deal was a background process.
And I definitely didn't realize till I'd actually written out the list which would be the most important one, or that it was involved in half the others.
Not that bad for an essay only 2.5 pages long.
Absolutely superfluous comment: I heart pg.
a short essay
three anecdotes illustrating the point
one anecdote providing a counter-example (to provide context and also remind readers about fallability of generic rules)
a short list of literature (e.g. pointers to specific chapter of "founder@work", "4 steps to the epiphany etc")
For each aspiring hacker this could easily replace days worth of reading of assorted stuff out there, just trying to get up to speed. In turn, more hackers will jump in.
While I have done my best to abide by PG's rules there are some habits I just can't give up. I could easily ruin whatever prospects my startup has by my pig-headed determination to give people what I feel they need rather than what they say they want.
I would point out that sometimes they move away from you (competitors come in, new ideas, etc)
BUT, sometimes they move towards you, and that is the classic where the media says "they were an overnight success" but you see they had been working at it for ten years.
I'm sure they're cheap. The problem with "going big or going home" is that in many cases you are going home. The one thing I've learned is that in the vast majority of cases, the market moves much slower than you expect.
If you find yourself in the case of a market moving faster that's usually the problem to have....then you hire and spend like a maniac but it doesn't matter because if the market is truly moving that fast you're making money hand over fist.
The problem comes when you are spending like that and the market isn't growing that fast.....then you flame out.
I'm pushing premium over freemium.
It relates to point 5 heavily and we're getting the startups we work with to stop trying to be a freemium company and start with premium first. I've outlined my arguments here;
http://www.pollenizer.com/content/premiree-more-premium-less...
Thanks for posting this.
9 is a transient, since we only started charging last week. 10 is partially a result of 9, as we do consulting to fund ourselves.
This is a great list though. I can think of real-world examples from the last year that reinforced each of those things on the "have" list.
This is the real startup killer for our last startup. But it goes with other problems like get ramen profitable. Since you are not profitable and you need to eat, you will be very easy to be distracted..
Do you know, what are living expenses in SFO? There are about USD 500/month in Czech Republic :)
Compliments the article I read just prior to this:
"Working For Somebody Else Never Amounted To Anything - Wayne Huizenga "
Sam www.b2b2dot0.com
Someone I knew they launched startup and the guys also have some part-time jobs. What's your idea?
Until you know that...
Yeah, sometimes silently, as in the counter-party just stops calling you back one day. Not unlike the dating scene.
launch it fast... launch it early. This article is... perfect?