[maxklein]

The problem here is simple: The people buying pick-axes are much smaller than the people buying gold. If there are 1000 people buying gold, there are 100 people mining for gold, and 10 people selling pickaxes. So the pick-axe market is quickly going to be dominated by someone who executes well and can offer the best pick-axes at prices less than what anyone else can compete with.

It's like frameworks - only a few rise to prominence, and that's because they are constantly developed, well marketed, etc. There is little space for a lot of choice in the tools market. The winners take all.

In the consumer market, however, even if you don't make a blow-out hit, your chance of getting by are much bigger. Even if you find a little gold, you can always sell it off and survive. But if you're selling pick-axes and they are lower quality and more expensive than the axes of the other established players, you'll not sell anything and die quickly.

In both cases it just comes down to who is smarter and faster and can better manage their finances to expand wisely. Both the pick-axe business and the gold business need proper intelligence and good business sense, and those who can bring both will be successful in either business.

Another consideration is this: It's easy to find gold. It's not easy to sell pick-axes. For pick-axes you need a supply chain, a cheap source, a marketing strategy. Gold panning requires a pan and some luck. The people who go for the Gold rush are normal people hoping to get lucky. They lack the skills to actually produce and sell jeans. Even if they tried, most of them would not be able to sell pick-axes.

So you can't really advice them to go sell pick-axes, as that is a more skilled business than panning for Gold.

[jasonkester]

I think you missed the point of the analogy.

The "Miners" in this scenario are all the companies trying to cash in on the dot com boom. There are thousands of them, and only 14 are going to make it big.

Those thousands of companies will burn through tens of millions of consulting hours. Consider those hours to be picks, and hopefully you'll start to see what the analogy means.

If you sell 2000 hours of consulting this year, you're guaranteed to be ahead somewhere between 100k and 300k, depending on your rate. If you instead spend those hours building the next Twitter, chances are very good you'll end the year with nothing at all.

Just like 90% of the people who headed off to the Yukon with a shiny new pick in their hand.

[maxklein]

I believe you miss the point of the analogy.

The analogy is not about people selling services to the gold diggers, it's about selling tools to the gold diggers.

The services market would be akin to the people who sell food to the miners.

[jmm]

Perhaps you're both being a bit too literal? I think what Chris is trying to say is that while it may be more obvious to pursue a consumer facing startup because of the attention they get, entrepreneurs should still keep in mind the less shiny/immediately gratifying things. Because the existence of the first set creates an opportunity for the second. Things like Heroku.

Kind of apropos... I got stuck on a trip recently without a book, and grabbed a copy of Heyday off a shelf. It's a historical fiction novel about the gold rush period and how things in CA played out. Not a terrible read, if you're into that kind of thing.

[maxklein]

What I'm saying is that the numbers don't add up. There are not 10 herokous or similar - platforms are a winner take all game. The market is also much smaller than consumer goods.

It's not clever to think there is more opportunity in selling B2B products : the market is MUCH tougher and the competition is a lot more skilled.

This gold-digger analogy is seductive because it fits in so perfectly with the programmer mindset, but look outside the bubble: most money is made in selling goods to consumers, not to businesses.

[spokey]

> There are not 10 herokous or similar

1. Salesforce's Force.com

2. Amazon's EC2

3. Google's AppEngine

4. Microsoft's Windows Azure

5. VMWare/Salesforce's VMForce

6. Makara

7. CloudBees

8. PiCloud

9. PHPFog

10. DuoStack

I'm not sure it's obvious that PaaS is "winner take all", but even if it is, right now there are several deep-pocketed companies fighting to be that winner. Sounds pretty good if you're looking for an exit.

[maxklein]

Those are not hosting the same languages as Herouku, so there is no direct competition.

[joshu]

This is important. The relative frequencies mean that observer bias cuts all the harder.

Additionally, it's very difficult to take the winner and back-calculate the particular winning strategy. The pick-axe sellers won in this case. But one story is not statistics.

Finally, it's important to note that the story sounds good and that makes us believe it even more strongly...

(I bet it's even better to be a banker at the time of the gold rush, incidentally.)

[nl]

It's always better to be a banker.

[ojbyrne]

Except during a "gold rush." Then there's a 1000 people mining for gold, even though only a 100 are going to be successful. Which implies there's 100 people selling pickaxes.

[tastybites]

There is little space for a lot of choice in the tools market

Apparently not, as anyone in the enterprise software world can tell you. There's plenty of room for lots of shitty choices, almost all of which generate millions of dollars of "value".

I do agree with your final conclusion though, that it takes a higher level of competence and business acumen to start, say, an ISP than another run of the mill social network.