I also think the 90% figure is misleading - but I think it's too optimistic. The number of people that quit their job and use their savings and/or family-friends-fools money to start working on a startup and never get to a point where anyone knows they even exist is huge. These people are never counted in any of the official statistics, and don't talk much about their venture because it's never nice to talk about your failures.

If you include this very large shadow group I'd argue that less than 10% even make it to ramen profitability. You never hear about the failures - only the successes. This skews the picture.

There is a second shadow group, those that are quietly successful, and rake in the money while other (often smaller) parties are banging the drums as loud as they can.

Some have really good reasons for doing this, such as protecting a niche they've identified.

I might be a member of this shadow group.

I read the other day (HBR?) that 85% of venture-backed startups are dead within 3 years. Say what you will about VC firms, but the companies they fund tend to be a notch above the mean. And they are certainly the types who get (and maybe even ask for) advice. Creating a business (the kind that supports 2+ founders) is HARD.

Thinking your chances are better because you're (by your OWN assessment) smarter, more open to advice, and more hard working than the mean is a dangerous attitude. If you walk into a room and say, "raise your hand if you think you're an above-average driver", pretty much every hand goes up. 50% of them are wrong.

Normally I love Gabriel's posts, but I think this one smacks of Mr. Rogers ("you're special because of who you are").

50% would be wrong if you said above-median driver.

Median is better than mean as a metric for "average goodness of driver". To take a median, you only need a strict ordering; to take a mean, you need to quantify. Any reasonable method of quantifying how good a driver someone is will be heavily influenced by environmental factors. The worst driver on an infinite, featureless plane would be involved in fewer accidents than the best driver in, say, New York.

Still, it could be even higher than 50% wrong, since sometimes the most capable people underrate their ability: http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect

Yeah, if you take average=mean, then well over 50% of drivers are above average.

So you think someone entering a startup blind (no advice, no methodology on customer development/hiring/etc.) has an equal failure rate to someone who's spent years taking in good advice?

The other subgroups were capital intensive vs not and low marginal cost vs high. I think all three would decrease your chances of failure.

What counts as good advice? I would imagine that bloggers on the Internet (even if they've sold companies successfully before, and even though they're better than the information vacuum before the Internet) don't really compare to having face-to-face investments from a network of angels who've done this before.

and if you're VC funded your model is capital intensive, otherwise what's the point.

VCs expect a 10x return and put an incredible amount of pressure on a startup. getting VC money is not the right path for every startup in every market, which is why the numbers are so dire.

all the data on this is fuzzy, and may always be fuzzy. what type of startup it is is important, as well as the quality of the founders, none of which is easily quantifiable. I think the reality is that you make your own odds. a brilliant/determined/flexible founder can and will make something work.

I've always felt that what kills a startup are commitment/flexibility issues based on factors outside of the business itself - your spouse doesn't support you, you're only willing to live on peanuts for so long, children on the way, etc.

but I don't look at startup success as determined by Idea X, sink or swim. If you stay in the game long enough, and find a way to feed yourself in the meantime, you'll find a way.

also, as in life, you alone determine what is and isn't success. for many people, a lifestyle business that just allows you to work on your own terms (not too far past ramen profitability) is enough.

I see this discussed now and then, and I never really see the point. I don't care what a startup's chance to succeed is, I only care what MY startup's chance to succeed is. And since these estimates vary so widely even when they're completely general, they're not even that useful as a baseline.

Intuitively I would tend to agree, but being a physicist I would like to see some more quantitative data on this.

Both the links from the article address different sets of companies than what he is actually talking about, the one on tomuse.com is about the companies on that famous graphic, which of course had already gained a certain amount of traction, otherwise they wouldn't be on there. The other one covers a vast superset of that, and also covers a much larger timescale than what I imagine when I hear the 90% figure.

how do you define success?

I'd say even "breaking even + $30K/yr" does not apply to 90% of websites/web businesses.

no opportunity cost WRT a normal career path :)

The "normal career path" looks very different from ages 20-40 than 40-60: that's what creates a lot of "involuntary entrepreneurs." Not everyone starting work at Google today, for example, will enjoy the same kinds of returns, and even job security, that have been demonstrated over the last ten years.

by that criteria you can create a wordpress blog, slap some adsense on it, and make $10 a month with it.

And even that is harder than it seems.

Not really. Updating regularly with good content and some SEO and you could easily get $10/month with adsense.

> Updating regularly

> with good content

> and some SEO

> could

That's what I said.

Updating regularly for $10 / month is a pretty hefty commitment, that's going to work out to what ? $0.30 / hour ?

Good content requires writing skills and editing skills.

SEO requires SEO skills, you'll have to do that yourself because at $10 / month you are not going to pass it off to someone else to do it for you. That's a good bit of reading.

And finally, that's a 'could', not a 'will' or 'shall', it remains to be seen if it will work for that particular case of content, update frequency and SEO skills.

For less than $300 / month you're better off washing cars.

You're making it sound like you need to be a professional writer to earn $10/month. I can earn that amount with maybe 20 minutes of work every 2 or 3 days, which, by the way, anyone can unless they're computer illiterate.

Whether you're better off washing cars is another argument.

30 / 2 * 20 / 60 / 10 that comes out to $2 per hour.

So what's your point?

Well my point is that $2/hour is kind of low.

Whenever I see people giving these estimates about how their blog is earning $5 a day without doing anything, they always seem to be low-balling the amount of time they spent on it.

I've had multiple ad sense websites and they all net below $1/hour. I'm sure I can improve that, but not enough to be a worthwhile part-time job.

I would be happy to have someone prove me wrong by helping me build a low effort $10+/hour blog, but until then, I remain skeptical.

So writing something for 20 minutes every 3 days (which some might even enjoy) counts as a part-time job? It's like saying browsing on HN is a part-time job. Or were you talking about the $5 a day? In that case, yes, you might have to put a little more effort into it.

However, building a blog that earns $5 a day without doing anything(or let's just say, doing a lot less than you're supposed to) is possible, but it's just not ethical.

Let's not get hung up on definitions here. It doesn't really matter if $10/month is a job or just a hobby to you. The important question, is whether or not it is attainable with low effort as you seem to suggest.

I've built a few websites with pagerank 2-4 where I spent more than 20 minutes every 3 days and enjoyed it. All of them have netted less than $10/month over a span of 5 years.

This experience has led me believe it is not possible to even get to $10/month without treating it as a job...unless you happen to enjoy blogging about acai berry and prescription drugs.

Let me ask you a few simple question:

Do you have a blog ?

Url please ?

How much money does it make ?

How much time do you spend on keeping it up-to-date ?

How much time did you spend building ?

How much time did you spend gaining your SEO skills?

How long has it been running ?

With those figures we can work out exactly what your hourly rate was, can you provide this ?

So ~200 minutes per month for 10$

~3$/hr

"You're doing a software/Internet startup with low marginal costs."

I know this is where a lot of the media focus is on software and internet. But I think there is still a good number of service or product related start-ups, although I don't have actual numbers.

Same kind of misleading advice old folks tell the young to study hard, get good grades, work for a good company, and blah blah blah ...