What if successful startups are just lucky? (opens in new tab)

All successful startups are lucky, but they're never just lucky. If they were, we could save ourselves a lot of work reading applications and doing interviews. We could just pick randomly instead. If we did that for one batch, you'd really be able to tell the difference, believe me.

This has been on my mind lately as well, for some reason. That, and the point of this story http://sivers.org/horses which is basically that even after the fact, it's really hard to know anything[1].

I don't know if it's just from growing older but I feel a lot more humble when looking at peoples ideas and actions than earlier. I literally feel that I have no clue whether they will succeed or not whereas I used to quickly get quite strong feelings in one direction or the other. In one way, this feels more mature and zen but at the same time, it annoys me a bit. I feel that conviction is the trait of a leader and something I need to possess if I want to convince anybody to engage in my projects. Maybe I'll just practice faking it and hope they didn't read this comment :-)

Edit: [1] One of my favorite examples of this is in Felix Denis' book How To Get Rich where he describes that the problem with Apple is Steve Jobs. This was written 7 years ago when Apple was struggling and it looked like Steve Jobs was driving it into the ground with his stubbornness. (The book is excellent by the way, in spite of the tacky sounding title)

One important point to realize for the author is that games are not services, they are works of art. As such, you can't say this is an incredible game with the same objectivity as this is an incredible service.

Ryzom was a great game, but MMOs are built on one thing: atmosphere. Blizzard made WoW precisely because WarCraft was its longest running and most popular brand to date. Blizzard already had a massive fan base of WC fans ready to play WOW. The lore was considered a masterpiece, which immediately immersed the players in the world before they even played it.

With services, things are much more objective, because we don't care about branding. Using a service like Dropbox is not a deeply emotional experience, but it does work very well. Apple is one of the few companies that can instill emotion into its services and hardware, and that is due to their marketing and design philosophies, which are considered the best in the world. However, this is nowhere near the level of emotion that art invokes.

As such, I would advise the author to carefully distinguish between art and service. When making art, you should never make it with the expectation that others will like it. You should make it such that it satisfies your own tastes. If others like it, you know that your taste is shared. If not, your tastes are unique. It's a win win.

Don't regret making works of art. It is all that succeeds us when we leave this realm.

"I am a great believer in luck, and I find the harder I work, the more I have of it."

-- Thomas Jefferson (apocryphal; see [0])

[0]: http://www.monticello.org/site/research-and-collections/i-am...

As a former player, and customer service representative of the game, i can say confidently that the reason Ryzom failed had nothing to do with bad luck, and everything with bad decisions and incompetence.

They had no qualms about changing large parts of the game. The prime example being that they changed the experience distribution in the game such that where it was first that any large group of players could play together and cooperate, after beta it was changed such that only squad-sized groups could cooperate and squads being in the same area only meant difficulties, since experience could not be shared anymore.

On the technical side they had a lot of terrible code as a result of much of their 50 man staff being students doing an internship for free there. They also had a very bad attitude towards volunteer support staff, where, for example, an issue on the web site would be reported, the dev in question would say nothing, and when asked five minutes later would say "What problem, i see none?" after obviously having fixed it.

Lastly, they also had a very bad attitude to players. The game had guilds and an ingame browser (kind of hidden, since it was the backend for 90% of the game UI), so some enterprising players had figured out how to use that to provide guild forums. The result being legal threats.

As much as i love the game still to this day, and still hold on fondly to the screenshots and memories i collected in its wonderful world, it wasn't luck that killed it. It was Nevrax' incompetence and shortsightedness.

Well, according to statistics (see Kahneman) CEO 'performance' has very low correlation with perceived CEO performance. Also, past success has low correlation with future success. In other words, there isn't really a 'CEO skill', or a 'successful manager' skill. In other words, company success is not repeatable given the same set of people. In other words, yes, luck is extremely important.

There is no moral of the story other than: failure does not mean that you didn't do well.

I know it's not the point of the article, specifically, but it's very difficult to draw any startup conclusions based on a game company. Games are notoriously difficult to make into successful products, even more-so than your average startup.

That said, I do think the author has a lot of good points. Surely there is a portion of "randomness"/luck, but there's also a healthy dose of focus & determination that make things work.

To play a bit of devil's advocate, I do feel that in the last few years the idea of "success" has somewhat diverged from what it used to mean. Is Twitter successful? To their founders & early investors, certainly they are. In 5 years, will Twitter still be relevant & a worthwhile investment to their millions of new stockholders?

Was Zynga "successful"? Again: to early investors, they were wildly successful. They were able to dump their shares onto a greater fool.

To me, those are two examples of "lucky" startups. A startup like Salesforce is less "lucky" and more a well-executed product. It's not as sexy as a Facebook/Twitter/Zynga, but it's a solid business addressing a specific need. It isn't reliant on "scale massively and figure out a sustainable business model later", which to me is what the aforementioned companies are.

To paint a picture, some startups have fisherman tactics: throw as wide of a net as possible, catch as many things as possible and some small portion will be valuable. These would be startups to whom luck would seem to play a larger role (in my opinion).

Others are more focused: identify a specific set of targets to go after, and hunt them down. This would be more like Salesforce. To me, these are the startups that don't rely as much on luck.

I used to think I was really smart and that my intelligence would carry me when I finally set out to start a company.

Wrong, while I still consider myself to be smart, I don't think I'm savvy enough to rely upon good decisions to carry me. There are plenty of people smarter than me that have failed in their ventures. So, I've reverted to a brute force strategy coupled with an unrelenting resolve to see projects to completion.

Just lucky? Not just. When you build your companies with a willingness to change, frequently self re-evaluating, you will amaze yourself with how much you grow as an entrepreneur. It is ridiculous what I know today when compared against my assumptions 2 years ago.

When the author was talking about the failed MMO, I took him at his word when he said his team created an amazing game. So the point of failure is obviously marketing. Who knows what could have been if those game developers had a comparable marketing team in their corner. What if they launched a beta 1 year into development to help develop a core user group as well as aid in gamer feedback which could then be redigested by the marketing team to help them create buzz.

I love this rumored quote assumedly from Jack Dorsey "It only took 10 years for me to become an overnight success." The entrepreneur that doesn't die is hardened and polished in failure over a number of years. When luck finally strikes, he will have the tools, experience, insight, vision to take full advantage.

I'm 2 years into this journey, hopefully I'll find luck in year 3, and hopefully I will have grown enough to take full advantage of it.

"Success in [indie] game development is lucky like poker, not lucky like the lottery."

I forget where I read this; it may have been on Twitter, but it could apply to startups and probably many other facets of life.

Edit: Found the source. https://twitter.com/MattRix/status/401072871323086848

Of course they are lucky. Luck is not the only thing, you really do need something useful and it has to work so you need to know your audience, but luck is a big part of it.

You have to be in the right place at the right time. That is all luck.

It keeps me up at night that you could work your entire life in startups, do everything right and even work on good ideas, and you could still never really succeed. Then I realize that success (in terms of an exit or something similar) can't be the goal: I work on startups because I like building things (edit: things that have the potential to have a tremendous impact) and I want to change the world.

No, not really.

About being too complex to be controlled, yes maybe, but there is very good people controlling complexity.

I know lots of entrepreneurs, as I am one myself.

The best entrepreneurs will learn from other people. Take for example Steve Jobs, he was not born knowing it all, he had fckng Robert Noyce as mentor, and all his life was spent learning from the best he could, like Polaroid inventor or Sony founder.(every year he will take 100 people from his company to learn from)

No matter were you are, there is always someone 10 to 20 years over you on a topic, and the best thing you could do is learn from them instead of thinking you are inventing gunpowder(Spanish expression that means you are inventing something revolutionary nobody knows about).

I personally know a man called Warren Buffet. This man, who is super rich old(normal old people believe they know it all) and famous will cold call you if he is interested in something you do for please explain it to him, and then he will start asking you deep questions about your business like crazy and if you ask him, he would do the same for you for anything you ask. This man has amazing amounts of valuable knowledge. In public he could not say what he really things because his word alone is capable of making markets go up or down.

Successful people do something called "masterminding", that is very similar to "pay it forward" in Silicon Valley. They learn form each other what works and what does not. They do it officially or by networking and just talking.

For anyone arguing that there is some causation behind a startup's success besides "luck", I challenge you to prove it.

It is not easy to prove causation.

For example, startup says "We did X and this was the direct cause of Y", where X is some random thing, and Y is "success". That is difficult to prove. And startups never say that. They just say "We did X." Maybe they also say they believe this was crucial somehow to their success. But they never try to prove causation.

Of course, no one demands anyone to prove causation when it comes to the actions and the success of any startup. The scientifc method does not apply here. We are happy to listen to someone tell us about what a startup did, about their eventual success and then let our imaginations make the supposed causal connections.

I do not think it's unreasonable to question, as does the OP, whether "luck" is a proper explanation for the success of a startup. If it were, then that leaves startups free to make their own rules instead of believing they must do X in order to produce Y. And I say that is fine. The fact is, doing X might not bring about Y for every startup that does X. Clearly there is more to this than a "formula", whether it is the YC's or someone else's.

This meme is particularly viral among game devs. They say "Notch was lucky", but what they really mean is that Minecraft resonated with the market more than they can comprehend. They say "my game failed, and I can't see why, so I must be unlucky", rather than saying "my game failed, because people didn't like it more than the alternatives".

Short of people being forced to purchase it, the only thing that makes a product successful is when people want it. How the product stands on its own merits is dependent on customers knowing about it, and choosing it over the alternatives.

When it comes to games, works in the same genre may not even be competing against each other because of the way gamers buy games. Plenty of gamers will buy Battlefield AND Call of Duty, or will buy Minecraft AND all of its derivative clones. They just won't buy things they don't want.

MMORPGs in the same genre, though, are often competing against each other. People can only play so many MMORPGs because of the time and social aspects.

To echo the monkeys writing Shakespeare analogy, if a hundred startups are throwing darts at a dartboard, there's a decent chance that at least one of them will hit the bullseye.

Where network effects are involved, they may not even need to hit the bullseye - just getting closer to the bullseye than anyone else may be enough.

IMHO, luck is only a factor in success. Failure is much more frequently pre-determined. For example, a startup with a great idea and a great execution still might fail, while a startup with a terrible idea or poor execution will almost definitely fail. I think of startups as needing to be not good but good enough: if the idea and execution are good enough, other factors start mattering. However, if they are not good enough, then no amount of luck will change that in the long run.

For example, the Color app had all the resources (funding, contacts), but the basic value proposition was just not there. They never had a chance. OTOH, Turntable.fm had a great product, but as luck would have it, they did not make it. Either the timing wasn't right or something else, but despite everything they are having to pivot.

It's doing-the-right-things plus luck.

You do need to be good at marketing. You do need to get product-market-fit quickly. You do need to have a good product, and so on.

The better you are at these things, the less luck is required. But you cannot get the amount of luck needed down to zero.

"Think of it like a lottery. In this case the obvious decision is to buy more tickets in order to get more chances to win."

I like what Seth Godin has to say on failing until you succeed:

http://www.youtube.com/watch?v=fDtkBsWgzWE

And this article by Scott Adams:

"What's the best way to climb to the top? Be a failure."

http://online.wsj.com/news/articles/SB1000142405270230462610...

If successful startups are just lucky, I have some questions:

- What's the expected monetary return for someone wishing to enter the startup game? In particular, is it positive or negative? For example, the usual platitude "you miss 100% of the shots you don't take" also applies to the lottery, where the expected return is negative. Could it be the case that rational people just shouldn't do startups?

- What's the expected return after adjusting for diminishing marginal utility of money? For example, a person with 2 million dollars is not 2x happier than a person with 1 million. This increases the impact of the previous point: even if the decision to a startup gives you positive expected money, it could still lead to negative expected change in happiness. Most of the monetary wins go to a lucky few, which doesn't raise their happiness enough to compensate for the unhappiness of the unlucky many.

- What's the expected return if you take the previous two points into account, and also compare against the next best thing you could've done instead of a startup? For example, is there anyone on Earth who's deciding whether to do a startup and who wouldn't get higher expected money, happiness... everything, just by getting hired at Google?

- Given the above points, is it ever morally okay to advise someone to do a startup? What advice would you give people if you were truly, honestly immune to survivorship bias?

Luck and skill tend to have the same distributions (generally a normal curve). As an outsider looking in, it is difficult to tell the difference. In fact, it can be difficult as an insider as well. We may have a plan, and attribute our success, or failure, to the plan rather than to luck. Or something could look like luck to an outsider because they don't understand all of the intricacies that go into making a system work.

Great blog Vianney. I just attended a conference last week full of silicon valley's best and brightest (http://warmgun.com/). I don't think that things are totally random, but I was struck by the million different ways to success (and failure). Chance definitely has a lot to do with it. It allows the really hard work to pay off.

Luck by definition is based on chance, something one cannot fully control. A startup of course involves luck, simplify because no company can control all the factors involved in becoming a success. So yes, I completely agree that successful startups are lucky. However, it's not "just" lucky as your title suggests. Just like playing lottery or gambling, there are ways you can maximize your chances of winning. Buy more lottery tickets as your blog post says, is one way to maximize the chance. In the startup world, obtaining the right skills, choosing the right features to implement, making the right connections and a million and one others are all the things you can do to maximize your chances of building a successful startup. I think your post illustrates the right idea that there is luck and chance involved at building a successful startup. But isn't that the case for everything in life? In the end, the rule of thumb is, "do whatever you can, and the rest is luck".

To put a little finer point on it, what if successful startups have enough randomly-uncontrolled and untracked things happen to them to make them successful as to be not reproducible?

So you do all the right things, and you still only have a 1-in-20 chance. Or the other guy does all the wrong things, and he sells out in two years for 20 Million. A third guy does some good things and some really bad things, still manages to grow and flip, then he writes a book on "How to have a successful startup" that claims to know the answers.

As far as I can tell, this is pretty close to the state of things, and that's okay. It just means 1) you can't give up, and 2) you need to choose your strategy based on this state of affairs.

What you don't want to do is to acknowledge there is randomness, then throw in the towel. Or to go on some jag where you hero worship Zuckerberg or some other rich schmuck simply because the startup lottery hit for them. The truth is much more complicated -- and involves a lot of hard work.

Obviously successful startups are lucky. I think the idea of releasing quickly/efficiently is to test the market to see if a larger, more complex product is desired (IOW can you get lucky?). In the example provided, a beta release of one level of Ryzom could have saved years of dev time if it was not well received.

If successful startups are just lucky then you should have as many startups as you can to get more chance at winning. That could mean focusing on quantity instead quality. Also that can be the reason behind MVPs: do something minimal that works then move onto the next project.

High growth start-ups are nothing but luck. How does a company built in a garage with practically no marketing budget get such high growth numbers in the initial stages (facebook, twitter, snapchat, instagram,etc)? They manage to go viral. Pure herd mentality.

"Even if it's not random, it's too complex to be controlled." - OP.

I think it's this realization that gives rise to quotes like, "Talent hits a target no one else can hit; Genius hits a target no one else can see." by Arthur Schopenhauer.

A start founders charisma points must be very high. You have to get others(co-founders, future employees, investors, ealry customers) to believe in your vision. And I'm not sure that's teachable. But, there are many ways to pull it off.

What if successful startup founders have high IQs? Or particular personalities making them persevere? It's entirely possible that all those methods and practices count for nothing or are completely eclipsed by IQs, luck, networks...

This argument is being made against education. How much education really matters and how much outcomes are just a matter of raw cognitive ability of students and (self-)selection?

Truth is that we don't know. So most people choose to follow in the footsteps of those who have demonstrable achievements and hope for the best.

BTW This might be the first time I'm seeing a ghost blog in the wild. Just browsing the Internet, not looking at ghost blogs specifically.

I think that it is some misunderstanding between be successful and have money. Money is the consequence, but the objective of a business should be create real value. In this sense, if you have only a few customers but you are delivering real value, you are successful. Take youtube, for example. They were a very successful company in the beginning, delivering a delightful experience. Now, if you consider money as objective, they are still successful, but they are not. They now deliver a awful experience, with Ads in the beginning, middle and the end of each video. For me, no matter how much money they are making, they just not successful anymore.

The problem is that it is hard to give each project any real chance of success without spending your full effort on seeing it through. A strategy like this will make you abandon your good projects at the first resemblance of a road block.

Honestly, this is poorly written. The author ends up giving advice even though he starts out claiming that is a problematic thing to do. His advice is not so different from common startup knowledge-- success is proportional to the number of swings of the bat. That I think any founder can take to the bank. It is true regardless of what your personal situation is. Obeying a simulated annealing algorithm rapidly through life keeps you from looking too far ahead and from getting stuck in ruts. There are many other generalizations largely independent of your personal story that are very helpful.

Just because OPs game floundered does not mean that startups must be lucky to succeed.

OP was quite proud of his game, is confident is was great because there were 15 talented people working on it... but ultimately the idea didn't resonate with the public in the way they implemented it.

Notch on the other hand resonated with the public incredibly well. By giving away completely free versions and listening to, even literally interacting with game players: He now heads a very successful software company where most of the original code was written by one man.

I don't think a successful company necessarily involves luck at all.

If you define luck as the happening of an event that the probability of not happening is greater than the probability of happening. Then all depends on how well you can calculate your probabilities of success and how much risk you want to take.

Of course you can't be 100% sure that you are going to be successful. But you must try to increase your probabilities as much as you can, trying to predict your probabilities as accurate as you can.

As Laplace said: "Life's most important questions are, for the most part, nothing but probability problems".

I think luck is a trivial factor in a startups success if you're smart. If you think about success more reductively, and ask yourself "what has to happen for me to succeed?" and "can I accomplish those things?", you'll see that luck isn't too big a factor. I wrote an article that talks about this: https://medium.com/i-m-h-o/cd29597704f9.

This argument compares apples to oranges. Your game was not the same as World of Warcraft. It had completely different graphics, different marketing, distribution, gameplay, etc. Facebook is not Google+. Windows is not OS X.

Lottery tickets are apples to apples. Yes, the problem domains for these start-ups may overlap, but these are not apples to apples. Your MMO was not the same as WoW.

The differentiating factor between your MMO and WoW was not just luck. They're different games.

It seems weird to be shocked that success in a startup depends on luck. Of course it does; we already know that from seeing how widely the returns fluctuate. But that doesn't mean successful startup founders aren't smart. A good way to put it would be that, while your chances of success are random, if you're smart and determined your chances are higher. If you aren't smart or determined, you don't qualify for the lottery at all.

"I am a great believer in luck, and I find the harder I work, the more I have of it." — Thomas Jefferson

"The harder I work, the luckier I get." — Samuel Goldwyn

Chance favors the prepared mind.

Great post and I'm totally agree. Once I got a idea, i do it quick and release it earlier, that's cool than just talking. but i usually lost after I release my first version of my project. I don't know how to promote my product, how to raise traffic and how to let people know my product, that makes me really confuse. I think marketing knowledge is also very important for engineer

The only startups that are "lucky" are those that never validate before development and still find traction.

I define validating as getting 3 paying customers before building a viable app.

You could have built an abridged or simpler version of your game and asked people to pay you to play it. If that posed problems, you could have pivoted to a different premise for the game.

Are famous artists such as musicians and actors just lucky or do they actually need the talent as well?

You can have all the talent in the world but not lucky to be in the right place at the right time. However, when you are at the right place at the right time you better make sure you bring the talent.

One doesn't exclude the other, you need both: talent + luck = success

1/3 Talent, 1/3 Effort, 1/3 Luck

My essay on the subject, cast in terms of climbing a fitness landscape and getting stuck in local optima: https://plus.google.com/u/1/+RayCromwell/posts/5bhdU8hF68t

I've noticed that people who win the lottery are the ones who bought tickets. Also the ones that win usually play every week, increasing their chances.

So at this point how are you defining lucky? They might be lucky but then they also worked harder and longer (i.e. bought more tickets)

Guy releases a unfun game with less then half the budget of his competitors and no name recognition along with poor marketing. Also program was technically ambitious. Then claims its random chance.

THIS IS precisely why its not random!!!

Among the best, luck is more important for winning than between best and mediocre (where skill is more important). I thought this was common knowledge after being described in so many pop.sci. books like black swan,etc.

It's similar to a video going viral. Luck is always a factor, but you need less of it as the quality of your video increases.

"I'd rather be lucky than good - but nothing beats lucky and good."

Obviously it can't be about luck, because that would deflate a lot of egos.

It all depends on what -- exactly -- you define as "luck."

If startups are luck, you should found as many as possible.

of course they are. But you have to have been working hard all those months/years, so you can capitalise on the luck when it comes.

First you are lucky, then you have a network.

There is no lucky

http://en.wikipedia.org/wiki/Outliers_(book)

This is what I always thought.

Interesting topic, but the blogpost lacks data.

Michael Jordan and Lebron James definitely get lucky... But I'd bet on them hitting that lucky shot more than 99% of other basketball players because they just know how to execute. One of my friends was a top 10 high frequency equity futures (snp minis) trader in the u.s. i used to think damn he gets lucky all the fking time. i mean at best it's like 70/30 ratio. just pure luck. how does he just call the tops and bottom and never read a sec filing or financials of a company?! nope he is really that good at execution, and that's why he's only right 70% of the time whereas others are just coin flipping.

a lot of times. timing is luck. but a lot of people get lucky on timing... it's like if the bubble bursted next year then people would have said there was never an easier market to raise money / take risks / make quick money in mobile,social, etc. but i'm pretty sure there's a ton of smart people with luck on their side that's trying to do that right now; and 100 other variables besides luck will make majority of businesses fail.

how do you define luck? break down the components of luck.

there is definitely an element of "luck" (timing, what other actors in the relevant space are doing, etc.)

but the way you hedge against that is by exposing yourself to lots of things with a potential net upside. both in parallel and serially. you might have bad luck with one thing, then good luck with another thing. just like the reasoning behind YC-style investment bets, it only takes one of your little bets to pay off huge in order to more than make up for all the other little failures. which leads to the other complimentary tactic: minimize your spend (money,time,energy) you're willing to risk on any given thing before you see an ROI from it. Also, you never have to completely give up on a thing --- in the software/digital world especially you don't have to destroy/delete/throwaway things, you simply passivate them (put on the backburner), press Pause, etc.

LOL... sorry to laugh, but casual perception should tell you that if you have a startup and your first product's release is a PC game with a date that is four years out (Four YEARS?) - You are going to need more than just luck to be successful.

If successful startups are just lucky, then getting an A on your project/exams in school was all luck. Can someone show us someone making money with shitty products today? Yeah, it's possible to get popular with an incomplete buggy product, but if it's filling a need, it's better than nothing. You will still have to work hard and improve to maintain your position.

Say what you might about how facebooks earlier/current code was/is horrible and in PHP, Facebook has done amazing things to be able to handle a billion users every second/every day. Say what you might about how simple instagram or snapshot is, they filled a need and grew. Their products doesn't "suck", sure they might be valued ridiculously by some groups of investors, but don't discount their work.

Successful startups are not JUST lucky. There is always an element of lucky, but there is also the hardwork. One mistake a lot of "failed" startups make is thinking and believing that they failed. If your goal is to create an application that solves a certain problem and you do so. You have succeeded. A lot of startups have this goal, they don't really have a goal of making money. After they achieve a goal of building their app, then they want to make money and when they fail on that end, they consider everything a failure. The truth is that making money is not just about your apps or how smart you are or hard you worked to designed.

It takes a very different set of skills to market, hustle, and sell. A lot of programmers don't have those business skills, and they whine and cry when the world don't rush to their door for the better mouse trap. Don't cry, the world didn't know because you failed to market or they have heavily vested in another one because you started late.

If you want to really realize that startups are not just lucky, find the other group we don't talk about on HN that make money off the internet. The hustlers, these are guys with little to zero tech knowledge. They do whatever they can, outsource via internet, basic wordpress sites, they sell actually products, their business is rarely based on ads, their business is physical products, software as a service or ebooks etc. Technology is usually a small part of their game, maybe 10%, the other 90% is pure sweat and hustling. We don't hear for them, but there are many of them out there, making very good money.

Successful starts are not lucky, they out hustled the rest and that's just what it is. So if you wish to make money, don't just stay a hacker, become a hacker and a hustler.