Just How Important is the Valley? Let's Look at some Data. | Hacker News Reader

Just How Important is the Valley? Let's Look at some Data.

tonywright.com

83 pointswebwright26 comments

Very interesting...

>> "I took some public lists of technology acquisitions in 2007 and 2008"

Not all startups are after the 'Acquisition' event either. I'd say if anything that would bias results in favor of the Valley. I don't know how you can really get a handle on the millions of 'startups' that just make money reasonably quietly.

Communities like HN, IRC channels, etc make me wonder more what the Valley has to offer. There's a whole array of cool online communities, and if you build something cool, people will start to contact you. I think face to face contact is needed less and less these days, unless you're trying to raise money.

Yaw, given the source of the lists (TechCrunch, VentureBeat-- two American Valley Blogs), I imagine there is a bit of a Valley slant.

And, as you say-- tons of startups are actually aiming for a profitable business and don't care so much for exits.

I agree in some respects, but there definitely are still some real advantages to being in the Valley.

I've been doing a lot of work towards monetizing justin.tv lately, especially with video ads. These products tend to be very ad-hoc, and are usually quite poorly documented. Sometimes we've tried to get an integration working by having me try to follow the docs, and call/email when things don't work properly. It usually works out in the end, but it's frustrating and takes too long (sometimes several days of back-and-forth).

To contrast, with the ad companies that are in the Bay Area, we just ask them to send a developer to our office. Both of us sit on my couch, while I do the integration and make bad jokes about the quality of their documentation. We usually get it done in an hour or so this way.

I think it depends on how integrated vs. modular the business is.

Disruptive business ideas usually require many deals between different parties - there's the business integration, and the technical integration that you mention (as video ads become more common, technical and business standards will probably emerge, making it modular).

But some business ideas are more modular, in that customers can just slot the produce/service into an existing business role, and into an existing socket in their system. They are plug-compatible, in both senses.

This plug-compatibility also applies to partners. If you need to interact with content providers and hardware manufacturers (e.g. as Apple needed for the iPod and iTunes), some kind of Pond is essential, not just for access to those parties, but also access to helpers of the connection (e.g. lawyers who understand the issues in these kinds of deals). I think this is probably true whether you have a novel disruptive business, or just an ordinary one: if you need to make deals, a Pond makes things much easier.

But if you can modularize (systematize) your interactions with your "partners" to such an extent that they become "customers", you don't need to be in the same pond. An example is the "partners" of Google Adwords - there's a standard interface that people use all over the world, and it works well even though there's no customer service.

New ideas tend to not be modular - standardization comes later - so a Pond there.

It's true that some businesses are more nimble than others, but Bill is still right: the ability to sit on the couch here is a more efficient integration mechanism than any other. You're far more likely to complete a deal when the technical work can be done extremely quickly.

I've worked at startups in Seattle and San Francisco now, and in my experience, the startups here are faster. At the risk of sounding like a cliche, it's just something in the air here. Things are always on fire.

Is it worth the cost of living? I don't know. But there's definitely a difference.

With email, you can ignore it, and just drag things out. I had a pretty good experience with Google the other day though, they changed something on adsense which just happened to break Mibbit (Their js tried to access something, security permissions, threw an exception Mibbit wasn't expecting etc). I sent an email, and about 10 emails later and about an hour later, we'd sorted out exactly what was going on, and they'd fixed it.

I think it can work either way...

Even if I was in the same city as the other developer I would just talk to them on the phone. I think the 'in person' meeting would usually involve enough time wasting to defray the benefits of the small increase in direct communication. If you are going to be doing a lot of work together it can be nice to meet the person but it isn't necessary just to clarify something.

>"I don't know how you can really get a handle on the millions of 'startups' that just make money reasonably quietly."

There are millions of startups? There are millions of startups making money quietly? Does anybody know, because from where I am sitting I'd be surprised if there were 10,000. And yes, it does make a difference to the argument.

Well most 'affiliate' websites I know of don't call themselves 'startups', and make a lot of money. Quietly. I don't know if you want to call them startups, depends on your definitions, but they're online businesses that make a lot of money. Then what about all the ebay businesses out there making a lot of money.

A start-up isn't somebody making money on the Internet, or somebody using e-commerce. It is a technology company founded by a small group of people that, were its technological product prove popular, presents an opportunity for spectacular growth.

It is the potential huge return that makes start-ups of interests to VCs (and if you believe PG's thesis, of potential large economic impact). If I had an Internet business idea that would make me $250,000 for life guaranteed but fixed, that's a great business idea for me - but not for an investor.

Yeah, if you use an unduly narrow definition of "start-up", you can get some small number in the few thousands. But if you use a more normal definition (like the one at the top of http://en.wikipedia.org/wiki/Startup_company — “A startup company or start-up is a company with a limited operating history. These companies, generally newly created, are in a phase of development and research for markets,” or my usual rule of thumb, “A startup is a company that has not yet become profitable,”) you get a much larger number, perhaps in the millions.

This post is a result of this thread: http://news.ycombinator.com/item?id=563810

I did a bit more digging. As the previous commentors pointed out, the data is not perfect and I'd be careful about drawing any conclusions about it. But, I thought it was interesting nonetheless.

Regardless of whether the data is "perfect", I haven't seen anything more than anecdote supporting the SV hypothesis. Great article!

I posted a reply here:

http://news.ycombinator.com/item?id=567497

I thought this data was interesting, and I used it as an excuse to play around with some data analysis / visualization tools that I've been exploring lately, Many Eyes and DabbleDB.

I wrote a quick blog entry discussing how I used them to perform some further analysis on the data:

Visualizing and sharing data ( http://daniel.yipyip.com/77049.html )

One of the interesting things here was this graph (when you change it to $ per state); http://manyeyes.alphaworks.ibm.com/wikified/acquired_startup...

Texas brought in almost as much money as CA in acquisitions in this snapshot of data.

While this blog post certainly rings pleasantly in my ear (based in ATX), I think there's more to being in the valley than simply the success rate of a startup. I'm very envious of the enthusiastic community and people you meet in the area, whether you're at the bar, having a cup of coffee, etc. However, I'm not envious of the cost of doing business and basic living.

RE: 2008 stats, I think it's a reasonable assumption that the economic downturn is certainly being felt first by the east and west (capital capitols) before it hits other tech satellites like Austin, Boulder, etc. It will be interesting to see who bounces back first and at what velocity... Compared to Web 1.0 => Web 2.0, things could be drastically different this time around given the market's significant retraction, and displacement of VCs in web startups.

One thing about business here, we haven't had the luxury of big VCs so most startups here have a different challenge that generally results in a profitable web business, not big exits.

The Valley offers no edge here– in fact, some people wonder if the “echo chamber” doesn’t actually get in the way of understanding the rest of the world.

This was my experience at Startup School last year. I regularly run into business owners who need all kinds of technology, but no one I ran into at the meetups knew anything about it.

If another person had pitched me his flash card app, I would have jumped out the window (except no building had more than 2 stories).

Damn you for harshing on my flash card app for small business owners.

Without harshing on your data (esp the denominators of population size as you remark), you do show that it's far from absolutely necessary to be in the Valley.

But to put it in perspective: in Valley vs. World, it's doing disproportionately well.

This data isn't of much use for analysis without all the startups that failed or didn't get acquired. This is like trying to analyze average height by nationality using a sample consisting of professional basketball players for each country.

Actually, I think it might be a bit better than that. The question is - what biases are there to acquisition with regards to location? Let's say you are more likely to be acquired if your are successful (surely we can all agree on that). If SV is the navel of the world, then you would expect startups in SV to have a higher chance of being acquired, than say a startup in Philadelphia.

Now the question is - is being acquired a good trace marker for a not-particularly successful startup? The answer is it doesn't matter - because the original PG essay was talking about being a major economic driver, so in some ways it is irrelevant what the non-successful startups are doing.

What it comes down is, is acquisition a fair sampler of successful startups. I don't know the answer to that, but I think a fair number of people would be surprised if it was "Hell, no".

Back to the sporting analogy, it is like trying to figure out the fastest nationality by looking at their 100m record. If each country trains and sends to the Olympics its fastest people then yes - but you have no simple way of knowing whether that is true. However you may draw some conclusions about who the fastest race is, on the basis that you would expect any bias to work against them.

correlation does not imply causation

Thoroughly mentioned in the post. Real conclusions would require much more data. Still, I thought it was interesting.

Why do people routinely make this cliche comment, thinking its going to give them genius points, when it only kills their karma?

Well vaksel has more karma than most small countries, so I don't think that was a consideration for him. :-)

Reply on news.ycombinator.com