I'm anticipating that some people will misunderstand Brin's advice which is caused by 2 factors:
(1) Page and Brin started Google Inc successfully in SV so it seems contradictory to advise others not to start a company there
(2) the meaning of the word "start" as Brin is using it
As for (1), Larry and Sergey had the luxury of building Google (the product -- not the formalized "incorporated" Google Inc) inside of Stanford and therefore using the university's computers and bandwidth for free. Piggybacking on Stanford's resources to create an "mvp" was like getting "AWS" for free. A lot of startups don't have that situation. Therefore, a founder may be better off working out the parents' garage in Kansas to keep expenses low. The free room & board is money that would have been spent on a crappy apartment in SV. Instead take those savings and use it to pay for the AWS hosting costs, or a 2nd programmer.
For (2), I think Brin is talking about "start" as in start from zero with an idea on a paper napkin, before an MVP, and before traction from others (users and investors). Zuckerberg started Facebook in Boston and after Sean Parker got interested, he moved Facebook to the more expensive SV. Bill Gates started Microsoft in Albuquerque, New Mexico and made profits before he moved the company to the more expensive Seattle. In other words, you can have a "low profile startup" in your local town that's under the radar which lets you keep expenses very low. You can later have a "high profile startup" in Silicon Valley where the very high costs are worth it to hire from the talent pool and network with other entrepreneurs and investors. That delayed move to SV is what Brin called "that second step".
Brin's advice rings true to me. Silicon Valley has changed since Google started. It has become a tough place to bootstrap a company from scratch especially if you are focussed on getting to profitability at an early stage.
Expenses in Silicon Valley are high and show no sign of declining. Even more important it is hard to find qualified staff in SV who are interested in working on start-ups that don't have strong funding. Most companies therefore look for a somewhat lower cost place with qualified engineers and have done this for a while now. At some point you have to ask why not just put the whole company somewhere else.
I had a contract run out last fall and was surprised by the number of companies who have moved to Seattle and brought their tech stacks with them. It's got a broader reach than ever now, I think.
And your choice of neighborhood has relaxed a bit, if you hate commuting. Even the east side/west side split is breaking down (.Net downtown now, and Java in Bellevue)
Just out of curiosity is it easier in Seattle to meet the conditions I mentioned, namely keeping expenses reasonably low and getting qualified staff? Housing and technical employment currently look like seller's markets in Seattle.
For example, if you take numbers from comparison tools like www.numbeo.com cost of living in Seattle is not much less than SV locations like San Jose.
It's certainly getting worse. I get sticker shock in my own neighborhood now and I only bought 5 years ago.
But I do think it's possible to make half the salary in Seattle and still not spend 40+% of your income on housing. But I haven't rented in a while and I've heard it's getting harder.
The thing is that if you know how to live frugally, being in an expensive town for a while can work in your favor, as long as you promise yourself you'll take the money and run. If you meet Mr or Ms Right before you get out though you may well be in trouble.
when i moved back to seattle from LA a few years ago, rent in issaquah was on par with the nicer LA neighborhoods(brentwood, beverly hills, santa monica).
a nice apartment close to work in greater seattle area is 2.4 - 4k a month for a two bedroom.
buying will cost you $5k+ unless you are putting a big chunk down, and that just seems to keep going up without missing a beat.
i would say its not possible to take a 50% cut and come here, unless you want to spend a lot of time commuting or hearing lots of sirens at night.
I've always wondered how much of the money YC seeds to its founders eventually ends up in a landlord's pocket? $3000/month on a studio apartment. That's crazy burnout for a startup.
"So startups are just an inefficiency imposed on wealth transfer from VCs to landlords. What if someone disrupted the industry and connected VCs directly to landlords?"
Seems like VCs would do well to buy some housing. A few more years of rent hikes down the line, "far below-market housing" could be a seriously compelling employee benefit. Apartments big enough to raise a family in would do wonders for employee retention.
That doesn't make sense though. They could still provide the same benefit by renting the apartment and giving it to the employee. Unless you think VCs should be betting on real estate, because that's the only advantage - they get the upside if the property goes up in value.
Currently, VCs must fund salary increases to keep pace with rent hikes. Renting apartments for employees just shuffles the numbers around. VCs could buy their way out of exposure to this phenomenon by owning their employee's homes and not increasing the rent.
Yes, this is a bet on rising home prices in the Bay Area, but that seems basically equal to a bet on the Bay Area tech ecosystem, which is already their core business.
I dunno, maybe it makes sense, but there seems to me to be something incredibly demeaning about living in housing owned by your employer. Very....19th century. Is there a company store?
I've lived in 'man camps' on remote jobs where it was easier to just put up trailers for workers than put them all in hotels. Or sometimes there isn't any temporary housing options available so you get to stay in the man camp. Pretty sure programmers aren't going to fall into line for that option though.
This wouldn't be remote, though - there'd be a few dozen such company towns on overlapping geography. True, you'd need to move to change employers, but it could just be across the street.
> Seems like VCs would do well to buy some housing.
Incubators yes. But exchanging a place to stay for equity (as OP alluded to) is mostly in the realm of the show although it has happened a few times in real life as well...
The median rent on a 1-bedroom in SF has increased from $2500 in 2010 to $3600 in 2016.
I don't accumulate as much stuff as some do, but I'd say $12k/year in extra discretionary cash (not just throughput to a landlord) is more than adequate compensation for the risk of a cross-town move.
Sounds like company towns all over again. VCs could open their own stores and sell food and only accept their own currency. And they can pay you in their own currency so you don't have any exchange rate costs. Oh, until you want to get out.
Samuel Brannon established his tool monopoly by diverting church money and using thugs to keep out the competition. Once he made his money he lost it all in real estate speculation and a divorce.
A number of successful industries have owed a deal of their success to the fact that they slowly turned into real estate magnates. And not just their property but property around them.
Maybe that suggests that these incubators should be owning rather than renting. Your investment in property is generally going to be recoverable if you have to kick out a failed startup.
One of the richest people in Silicon Valley is John Arrillaga.[1] He bought farmland in Silicon Valley and began converting it to office space. About a square kilometer of building floor space.
Sorry, wasn't trying to bash Texas. Maybe I should have said "Dallas oil dynasty". I haven't seen, read or viewed anything from or with Marc after he left Netscape, so I can't comment on what he's doing these days. All I know is that he's part of a VC.
Yes, except the sale price will probably have a good portion of expected future rent increases "priced in". (I'm not saying you won't make any money at all)
I can't imagine a lot of it. At some point you make the conscious decision to quit your current job and work on your startup. If you've prepared financially for that joining YC shouldn't really change that. You're spending your own money on your own rent.
Granted, if you can't afford rent in SV you might want to reconsider joining a YC class. It's a tough decision for sure.
Brin built his company on a kind of ageism. Several kinds in fact.
So that he and his peers think that continuing to live like you did in college isn't that big of a logical leap when you consider that their interview process feels exactly like a final exam in college, despite the fact that you and I and indeed Google's own success stories know the difference between theory and practice.
Days when "ramen-profitable" meant you actually ate ramen are over. Now everyone jumped on the fitness bandwagon, and eating various types of salads & green stuff is hip. And also expensive - "natural"/organic food is a big business these days.
Well starting a business has traditionally involved personal risk and sacrificing lifestyle for future gains. Only recently and mostly in American tech has entrepreneurship been "give away control of your company as fast as possible so you can pay yourself a comfortable salary while you build a company for someone else."
I think that was targeted towards 20somethings that don't have a wife/kids yet. Obviously, a start-up with extra baggage is an entirely different ballgame.
The idea that you should make livestyle tradeoffs while making $140K+ a year is absurd. But the alternative, spending $4k a month on rent, is also absurd.
Not sure, I believe San Jose, but was talking with a guy last week who was looking into wireless service since he was in a broadband desert. He was shaking his head as I have Google, ATT, and CCI all competing for 1G/1G service in my area of KS.
People have taken your example and run with it, but as someone from Kansas, I implore you not to start a business in Kansas. The governor has run what little infrastructure existed into the ground. There is nothing for you there. I'd suggest somewhere like Austin, where taxes are sufficiently low, the city has a supply of tech workers, and the city has working infrastructure.
For now! I'm not a cheerleader for Phoenix (I've been once and had a pleasant visit), but I do think it's well-positioned given its anchor university's commitment to experimentation, weather, cost of living, and reasonably good access to outdoorsy activities.
Markets figure these things out. Not sure why people feel the need to convince markets / market participants, unless they somehow don't like the decisions that those have made.
Markets figure these things out in favor of what makes money. There is no guarantee that what does get figured out is of any benefit to the people living there.
I agree. Though, Kansas City, MO has a decent tech base (for the midwest) and you can still get much cheaper housing costs than anywhere in SV, or even Austin.
From a Machiavellian perspective, Google competes in hundreds of market niches with competitors constantly nipping at its heels. It certainly doesn't want more competition. Brin doesn't need a 5 man startup hurting Google, so he's politically motivated to give advice that benefits him best. That Kansas startup with zero dollars isn't a threat to Google, and makes for a cheap acquisition if they do develop something interesting or threatening. It certainly is a threat when its backed by SV money, can scale quickly, and isn't available to buy for peanuts.
I've always been skeptical of the whole "advice from billionaires" format. Brin isn't motivated to help startups the same way Warren Buffet isn't giving away hot stock tips or how Gates never gave the FOSS community an easy way to reverse-engineer the doc format. These people know how their bread gets buttered. Or simply have internalized protectionist attitudes and give away just enough to not be dangerous to themselves.
edit: I'm not necessarily projecting malice, but there are other reasons why "advice from billionares" is kinda a bullshit format. Brin may be out of touch or, if you read the source of the article, made an off-hand comment he probably doesn't worry about all night and is blown out of proportion by the 'give us the latest hot investing tip' crowd.
> I've always been skeptical of the whole "advice from billionaires" format.
Me too. But it's not like it's unfounded advice. The things you need to start a company (time, knowledge, morale) are all significantly more expensive, and therefor less accessible, in SV.
Clearly investors are biased to having someone nearby. Even HN forces you to temporarily move to SV. How do you overcome this bias in Kansas especially when your competitors are in SV and chatting up people and making connections?
The problem with "how I got successful" advice is, even when its honest, is fairly dependent on that one person's experience. Most success has more to do with getting lucky or being at the right place at the right time, than purely merit. That Kansas startup may have a lot of merit, but a less meritorious startup actively "working the room" in SV will have a strong advantage.
I suspect you'd learn more from the guy who had five failed startups in a row than the guy who hit a homerun on his first time to bat. Again, the "advice from billionaires" format is really a bizarre thing. Its the closest thing secular society has to the Delphi oracle and from what I can tell is usually borderline dishonest or just a lot of pleasantries/PR for said billionaire. It does get ad impressions though, so I imagine this is why we keep seeing this type of format over and over again.
Maybe Brin's advice made sense in the 90s but is questionable today. Or maybe he's woefully out of touch. Or maybe he's being dishonest. Or maybe its just a throwaway comment that he never really thought about too strongly and is being sold by the tech press as "The New Hot Tech Tip!" Unfortunately, its impossible to tell.
Could it be that now that Google is no longer able to retain staff through a no-poach collusion, its CEO is instead trying to discourage other businesses from setting up shop in a place where they might be able to hire developers away?
That's cynical, but I'd say we have the right to be somewhat cynical after reading those emails.
Excellent point. Another nail in the coffin for the "altruistic billionaire." Those poaching emails are one of the rare behind-the-scenes glimpses we get of the world that people like Jobs and Brin live in.
I would not be surprised if poaching concerns are foremost for the higher ups at these giant tech companies.
A five-man startup that competes with Google can be very simply mitigated by Google, though.
There is an advantage for a billionaire to help startups get better - someday they can acquire the talent, product, etc. Remember that acquisition is a form of recruitment, too.
You don't speak to this exactly, but Brin might as well just suggest to not start a company in California, which demands a lot of legal and tax paperwork to be done right from a business's beginning.
For tech companies that paperwork is really not that big a deal. First, you need to get tax and governance stuff right no matter where you start if you plan on taking funding later on. Second, basic company setup is kind of a wash between California and Delaware. (At least that's the advice I have gotten in the past.)
Complexity of basic company setup is low enough that it's basically a wash. Delaware has benefits later, due to legal environment and tax, but the main benefits of Delaware won't matter to you until you get big.
I'm sure free access to 1998 Stanford servers and bandwidth was nothing like getting AWS for free. And a lot of startups these days do actually get AWS for free -- or at least the first $100K.
>free access to 1998 Stanford servers and bandwidth was nothing like getting AWS for free.
I wasn't comparing 1998 Stanford technology to 2016 AWS as if it was apples to apples. The "AWS" was a modern analogy to give an idea of what they didn't have to spend money on compared to other startups. In 1997, if you didn't have a university to host your (potential business) website, what were the options? You could call up one of the IBM/HP/Origin managed data centers and have them rack some servers. But that costs money, and requires a contract, etc, etc.
When NCSA was bringing you the World Wide Web and the University of Illinois millions in NSF grants and industry partnerships, and accolades and thank you's from vice presidential candidate Al Gore, all the work was being done in the nearly windowless basement of the old Oil Chemistry Building.
The university charged NCSA for the privilege of occupying the tenth shittiest building on campus. And we joked about what other crappy building they'd gift us when they finally got around to giving us the space we had requested. Their definition of building us new space was to build a building for someone else and give us their hand me downs.
And of course they owned all the IP and charged a pretty big tax on all of the income. I'm fairly certain we brought in almost as much money as the football program (who got a stadium upgrade the same year class sizes were increased to save money).
I have no illusions of Stanford having a more generous program. All that free stuff isn't actually free.
You're not disagreeing with him. Thing is, even if you get AWS for free, you're better of minimizing your living expenses and spending that extra money on other aspects of your business to get it going.
There's a lot of talk about investors not being as easily available outside SV, but having worked for startups outside the valley most of my career my experience is that it's actually the rich customer bases you're missing by not being in the valley. I think it's easier to evangelise a product where there's a huge receptive audience, and doing it in person is much more effective. Trying to sell people on some new tech in a 1 million person or less city is hard, even when that city has decent official support for startups.
Businesses are more cautious and there are harder to cross barriers to word of mouth spread for consumer oriented businesses. For example, I worked for a social network started before Facebook that was used by nearly every teenager in a large region, but we could not break out of that region no matter what we tried. The kids in the schools in our region just didn't interact outside the region much, so there were no network effects.
In the valley, though, nearly everyone you meet is excited about tech somehow. Many people are from somewhere else in the world and are still in some way connected to where they're from, both in terms of businesses and people. It's just fertile ground for network effects.
Certainly by the time Facebook became open to the general public it was. Before that, until we saturated our geographic area, it was definitely viral growth.
I'm not saying we didn't have other problems that would have led us to fail eventually anyway (especially to facebook), but we observed this as a very real problem.
This is the kind of thing that's relevant when it's relevant. If you are making twilio or stripe, it is probably important to be in SV. If you're making tinder, you should probably live in a big college town. Facebook? You can live anywhere. Twitter? LA might be a good idea.
If you're making an ebay killer, you can live anywhere.
I don't think Facebook could ever have been as successful anywhere else. At the very least, its incubation on a major college campus was a huge benefit, and the students themselves evangelised it to the rest of the world.
The folks who disagree seem to be aligning with the reasoning that it's easier to raise capital and be closer to investors in SF.
But raising money != starting a company. Sure, it's important at some point, but this is the growing problem with the industry as a whole, no? People show up with just a pitch deck and want a million bucks.
Most responsible (and technically-capable) founders are trying to save as much money as possible and get on the investing boat as late as possible.
I have wondered this for a few years now. It seems to me the best way to start a company is to actually build your idea somewhere cheap and fly to where the investors are. I don't see how someone moving to San Fran to start a company is a good idea when a large amount of your angel investment is spent on rent.
I think a better idea would be to move to a college town. I think the guys at id did this when they first started though they moved to somewhere in the midwest, but cheap nonetheless.
My take would be to move to college town with a large student population and good science programs. Somewhere like pittsburgh or even state college, PA
I'm actually not from PA, but have been to much of PA. I purposely left out Philly because I think its too expensive and not a smart financial move for a startup. In my mind, you are better off moving to DC if you are going to pay Philly rent.
It's all relative. Philly can be affordable, cheap even, if you aren't picky about location. I live near Scranton now, so maybe that has changed in the ~4 years since I've lived in Philly.
I'm in a state whose economy is mostly manufacturing-based and I've wondered why there aren't more startups here because it would be so much cheaper even if you were in the state capital.
I have decided it is kind of like the saying "Why did I rob banks? Because that's where the money is." SF is where the money is. Where the networking possibilities are. Where it is easier to get coverage in the tech press because that's where so many of the tech reporters are.
All of those can be overcome if you aren't in SF but that means you are spending time overcoming those things - time that could be spent on making your product. You may spend tons on rent and have to pay your employees a lot more but it is easier to get more funding than it is to get more time.
Agreed, but once you have the money, execute in a place I describe above.
Most tech startups don't make it so why not use the extra capital (because rent is much lower) as insurance to succeed.
I have worked in numerous places throughout the US. My most successful venture has been to licence my tech to a commercial company. I live in a small town and travel around the world each year for fun. Its a nice pace of life. I visit some friends in San Fran and find the talent there honestly not that great. Sure, there are lots of good programmers but you need a lot more than good programmers to succeed.
"I'm in a state whose economy is mostly manufacturing-based and I've wondered why there aren't more startups here because it would be so much cheaper even if you were in the state capital."
What has the state done to attract startups? Do they still enforce non-competes? Do they have well funded schools focused on tech? And, quite frankly, does your state have something that makes people actually want to live there?
If you do the math, you will find out that its better for you to take a few trips to San Fran every few months than to actually set up shop there. You can easily save tens of thousands of dollars over the course of a year by using this method.
Most VC's I have talked to really like traveling anyways. They can write the trip off as a business expense.
Depends a lot on the type of company you start I suppose. If in doubt start it where the initial customers are going to be. Even if you intend to build the next network-effect based super customer product....maybe pick one specific niche first and be around those people for feedback (FB/Harvard etc.).
My gut would say that location (close to key customers) is very important for B2B for example. For some startups I can't help but think that some of their (scaling) problems are directly related to being started in SV (food delivery or cleaning services started in SV for example).
I also think you should double check your mindset if your reasoning revolves around getting financing and thus moving to the place that optimizes for that. Solve problem, ignore capital (software isn't very capital intensive) >> think about funding first.
Lots of top valley folks say this, but then the VCs still want someone to be nearby. Anyone know if Brin is investing his own money more broadly (geographically)?
What Brin actually said was to start somewhere cheap, then move to Silicon Valley when you need VC funds. So he isn't dismissing the need for proximity when raising funds.
I'm not sure that conflicts with Brin's statement here; he seems to be encouraging people to start outside the Valley when they're pre-funding. A lot of VCs (for instance, Y Combinator) are happy to fund non-Valley founders as long as they use the money to move to SF.
Question out of interest - is it likely he is talking about companies moving from a location from within the US to SV?
Lots of young programmers/entrepreneurs in Europe seem to dream of going to SV as soon as possible. At the same time there are lots of great cities for startups around over here - London, Stockholm and Helsinki to just name a few... SV might be the best/a great choice later on in the process but I'm quite sure Europe has lots to offer as well (not even considering potential visa issues here, which obviously restrict many or most Europeans from joining startups over in the US).
As someone who has only visited SV, I wonder: In order to avoid the crazy living expenses in SV and SF, why don't more people start companies within 100 miles of SV, in cheaper places like San Ramon/Castro Valley, Livermore, Sacramento, Pleasanton, Modesto, etc.? That way you have reasonable proximity without having to suffer the living quality issues associated with SV and SF.
> In order to avoid the crazy living expenses in SV and SF, why don't more people start companies within 100 miles of SV, in cheaper places like San Ramon/Castro Valley, Livermore, Sacramento, Pleasanton, Modesto, etc.? That way you have reasonable proximity without having to suffer the living quality issues associated with SV and SF.
No, instead, you have the much greater living quality issues associated with the places you mention, which is why SF/SV are more expensive places to live -- people are willing to pay a lot more to live in SF/SV than in, e.g., the Central Valley.
That still involves moving to Sacramento. Most of the people in SV are there because they want to be in SV, not Sacramento. They want to be in SV to be in the center of what's happening, not to mention the cultural fusion of SF.
Partying definitely loses it's luster (at least for me) but the opportunity for meetups, learning about new tech, etc. hopefully doesn't. Clearly it's a balancing act but everything in life is. How much time do you want to spend with a spouse, kid, hobby, exercise, work, meetups, relaxation, sleep, etc. People are here because they want to have access to all these options. I'm 41 and still excited about all that is available here. So maybe I don't want to grow up. :-)
I absolutely do get to complain. I also vote against the nimby council members who refuse to build residential in lock step with commercial in Mountain View which led to half the board being replaced a few years ago... Things still have not gotten much better and now they are collecting signatures to get rent control on the ballot which in general I am opposed to but clearly something needs to change.
Everyone has different ideas about how much space is necessary. Up until recently homes were really tiny and it varies greatly by country (ref: http://shrinkthatfootprint.com/how-big-is-a-house) I agree that the kids should be outside playing in parks and riding bikes and all that jazz and not in a tiny apartment. But then we have the "safety" fears when the world is far safer than when and where I grew up.
In my life I have lived in townhouses, an efficiency, rented a room in 2 houses, rented a whole house, lived in a dorm with roommates and lived in a big house with an acre back yard. They all have their plusses and minuses. I have not yet owned a house myself though. I would argue it's better the kids live in a small apartment than their parents have 2 hour round trip commutes because they had to switch jobs after they bought a house. I put a lot of value on the ease with which I can move closer to my job.
How about people just start building your idea, wherever you are. Rather than worrying about where you are when doing it? If the thing has merit, the rest should fall in line.
Why would Sergey Brin advocate startups being founded where all the action and high valuations are?
He knows that it's hard to move your company once you've settled in somewhere. Much more economical for Google if you stay on the fringe outskirts and let them acquire your baby for pennies on the dollar.
He literally is. The whole point of the article was that SV is great when you need VC money, but if you are just trying to create a product it's expensive and unnecessary. Create your product, develop a user base and THEN when you start to need cash, move to SV.
Being headquartered in San Francisco, CA is "negatively correlated" (standard deviation when pooled with all cities: -0.5261) with having a successful exit.
Best cities/towns/etc (most correlated with experiencing an exit, rather than failing/floundering (shut down, or founded before 2004 + no exit/acquisition/IPO)) to create a startup:
stddev Exits Total City
8.4678 143 207 Mountain View, CA USA
6.4855 130 356 London, H9 GBR
5.7360 15 25 Berlin, 16 DEU
4.5423 23 56 Tokyo, 40 JPN
4.5105 17 32 Richardson, TX USA
4.3559 56 159 Paris, A8 FRA
4.2118 44 99 Vancouver, BC CAN
4.1381 28 62 Montreal, QC CAN
4.0152 60 99 Waltham, MA USA
3.9462 57 138 Atlanta, GA USA
3.9414 14 21 Arlington, VA USA
3.8747 15 20 Foster City, CA USA
3.8528 14 41 Copenhagen, 17 DNK
3.7484 32 42 South San Francisco, CA USA
3.7131 14 31 Baltimore, MD USA
3.6728 39 79 Fremont, CA USA
3.4364 9 17 Irving, TX USA
3.3919 8 11 Chelmsford, MA USA
3.3833 5 7 Surry Hills, 2 AUS
3.3637 19 59 Bangalore, 19 IND
3.3335 11 24 Hamburg, 4 DEU
3.2568 19 36 Morrisville, NC USA
3.2101 10 29 Seoul, 11 KOR
3.1476 17 32 Bethesda, MD USA
3.1191 17 42 Cambridge, C3 GBR
2.9591 7 18 Zurich, 25 CHE
2.9566 7 7 Chapel Hill, NC USA
2.8746 11 32 St Louis, MO USA
2.8625 39 83 Boulder, CO USA
2.8225 35 69 Portland, OR USA
2.8000 45 128 Beijing, 22 CHN
2.7117 15 23 Aliso Viejo, CA USA
2.6170 13 19 Los Gatos, CA USA
2.4679 12 30 Helsinki, 13 FIN
2.4569 4 12 Istanbul, 34 TUR
2.4385 22 49 Minneapolis, MN USA
2.3187 21 72 Dublin, 7 IRL
2.3046 5 6 Burlington, ON CAN
2.2251 4 6 Kennesaw, GA USA
2.1569 8 12 Sterling, VA USA
2.1564 111 206 San Jose, CA USA
2.0461 37 58 Cupertino, CA USA
2.0432 20 36 Pasadena, CA USA
2.0180 7 8 Itasca, IL USA
1.9883 23 51 Ottawa, ON CAN
1.9772 5 6 Oak Brook, IL USA
1.9468 7 8 Westford, MA USA
1.9386 3 3 Gent, 8 BEL
1.9285 4 7 Delft, 11 NLD
1.9251 9 16 Beverly Hills, CA USA
1.8902 8 18 Oslo, 12 NOR
1.8785 4 7 Costa Mesa, CA USA
1.8746 13 37 Tampa, FL USA
1.8346 9 21 Jacksonville, FL USA
1.8340 13 18 San Bruno, CA USA
1.8246 7 10 Venice, CA USA
1.8117 16 50 Stockholm, 26 SWE
1.8010 8 15 Wilmington, DE USA
1.7720 7 19 Burnaby, BC CAN
1.7050 5 7 Kitchener, ON CAN
1.6856 6 16 Gurgaon, 10 IND
1.6729 398 807 New York, NY USA
1.6474 3 3 Fuzhou Shi, 3 CHN
1.6362 15 41 Madrid, 29 ESP
1.6145 2 5 Utrecht, 9 NLD
1.6135 5 11 Minnetonka, MN USA
1.5937 21 35 Milpitas, CA USA
1.5785 14 28 Mclean, VA USA
1.5671 6 11 Orem, UT USA
1.5587 3 8 Prague, 52 CZE
1.5505 3 3 Lake Forest, IL USA
1.5461 9 12 Alameda, CA USA
1.5459 4 14 Rio De Janeiro, 21 BRA
1.5402 9 14 El Segundo, CA USA
1.5240 8 17 West Hollywood, CA USA
1.5223 4 5 Doylestown, PA USA
1.5127 9 38 Sao Paulo, 2 BRA
1.4928 4 6 Arlington Heights, IL USA
1.4881 15 27 Marlborough, MA USA
1.4800 4 10 Stuttgart, 1 DEU
1.4516 7 11 Kfar Saba, 2 ISR
1.4352 5 13 Dubai, 3 ARE
1.4241 2 3 Blackrock, 7 IRL
1.4220 3 5 Roncade, 20 ITA
1.4157 3 8 Espoo, 13 FIN
1.3833 7 24 Vienna, 9 AUT
1.3667 5 6 Solana Beach, CA USA
1.3442 7 14 Mississauga, ON CAN
1.3139 7 10 Belmont, CA USA
1.3073 13 29 San Antonio, TX USA
1.2982 6 14 Buffalo, NY USA
1.2742 4 5 Ames, IA USA
1.2476 5 11 Abingdon, K2 GBR
1.2381 19 50 Washington, DC USA
1.2296 102 171 Cambridge, MA USA
1.2282 7 13 Westminster, CO USA
1.2127 5 12 Annapolis, MD USA
1.2056 2 2 Odense, 21 DNK
1.1670 2 2 West Des Moines, IA USA
1.1492 2 2 Notting Hill, 7 AUS
1.1480 3 7 Schaumburg, IL USA
1.1403 13 35 Plano, TX USA
1.1257 3 3 Sunrise, FL USA
1.1201 13 31 Ann Arbor, MI USA
1.0986 4 9 Halifax, NS CAN
1.0967 2 3 San Marcos, TX USA
1.0963 2 2 Mi Wuk Village, CA USA
1.0841 4 7 Newtown, PA USA
1.0752 3 4 Zug, 24 CHE
1.0658 37 74 Tel Aviv, 5 ISR
1.0643 10 34 Shenzhen, 30 CHN
1.0558 5 10 Munchen, 2 DEU
1.0379 5 5 Branford, CT USA
1.0283 3 3 Fredericton, NS CAN
1.0099 2 3 Pune, 16 IND
1.0033 4 8 Charleston, SC USA
Worst:
-9.2963 55 178 Los Angeles, CA USA
-5.8472 13 135 Moscow, 48 RUS
-4.2334 2 16 Quebec, QC CAN
-4.1071 9 24 Berkeley, CA USA
-3.9312 4 15 Lucerne Valley, CA USA
-3.7260 63 124 Redwood City, CA USA
-3.5625 2 10 Netanya, 2 ISR
-3.3962 8 20 Newton, MA USA
-3.3100 10 20 Princeton, NJ USA
-3.2994 9 21 La Jolla, CA USA
-2.8129 6 13 Petaluma, CA USA
-2.5981 12 42 Raleigh, NC USA
-2.3986 14 41 Brooklyn, NY USA
-2.3916 4 10 Addison, TX USA
-2.3777 59 155 Toronto, ON CAN
-2.3297 2 6 Allentown, PA USA
-2.2695 2 9 Edison, NJ USA
-2.1975 1 5 Champaign, IL USA
-2.1557 114 240 Austin, TX USA
-2.0160 0 7 Livermore, CA USA
-2.0154 127 244 Palo Alto, CA USA
-1.9803 1 5 Blacksburg, VA USA
-1.9321 4 20 Melbourne, 7 AUS
-1.9214 13 34 Charlotte, NC USA
-1.8469 0 4 City Of Industry, CA USA
-1.8068 5 19 Rochester, NY USA
-1.7758 1 4 Lod, 2 ISR
-1.7431 0 3 Eatontown, NJ USA
-1.7269 4 12 Wilmington, MA USA
-1.7250 0 5 Owings Mills, MD USA
-1.6628 5 13 New Haven, CT USA
-1.6467 1 5 Golden, CO USA
-1.6458 2 16 Memphis, TN USA
-1.6433 0 5 Cedar Park, TX USA
-1.6354 3 15 Santa Ana, CA USA
-1.6351 9 17 Gaithersburg, MD USA
-1.6102 35 99 Houston, TX USA
-1.6033 0 3 Superior, WI USA
-1.5757 0 10 Saint Petersburg, 66 RUS
-1.5656 0 4 Tacoma, WA USA
-1.5626 8 25 Orlando, FL USA
-1.5451 0 9 Little Rock, AR USA
-1.5421 1 7 Galway, 10 IRL
-1.5226 9 41 Cleveland, OH USA
-1.5176 1 6 Liverpool, H8 GBR
-1.5007 5 30 Columbus, OH USA
-1.4983 19 48 Philadelphia, PA USA
-1.4838 0 4 Toledo, OH USA
-1.4825 0 6 Newark, NJ USA
-1.4783 26 83 Pittsburgh, PA USA
-1.4602 15 35 Oakland, CA USA
-1.4537 0 6 Sausalito, CA USA
-1.4369 1 8 Kista, 26 SWE
-1.4364 0 5 New Orleans, LA USA
-1.3845 6 20 Newport Beach, CA USA
-1.3755 1 8 Manchester, I2 GBR
-1.3457 29 84 Dallas, TX USA
-1.3180 0 8 Centennial, CO USA
-1.2865 3 13 Charlottesville, VA USA
-1.2839 0 5 Morgan Hill, CA USA
-1.2704 1 4 Lawrenceville, GA USA
-1.2478 1 6 Burbank, CA USA
-1.2446 0 8 Tallinn, 1 EST
-1.2440 87 157 San Mateo, CA USA
-1.2188 4 13 Plymouth, MN USA
-1.1966 0 3 Laguna Beach, CA USA
-1.1938 26 64 Salt Lake City, UT USA
-1.1711 0 9 Jakarta, 4 IDN
-1.1672 43 106 Irvine, CA USA
-1.1634 1 5 Guangdong, 5 CHN
-1.1536 0 2 Orsay, A8 FRA
-1.1421 0 2 Cherry Hill, NJ USA
-1.1222 4 13 Longmont, CO USA
-1.1221 0 3 Columbia, SC USA
-1.1055 0 4 Laval, QC CAN
-1.0837 0 2 Mountain, WI USA
-1.0790 0 3 Gilbert, AZ USA
-1.0743 4 7 Boxborough, MA USA
-1.0369 0 2 Pittsburg, CA USA
-1.0302 0 4 Napa, CA USA
-1.0275 2 7 Clearwater, FL USA
-1.0176 9 30 Indianapolis, IN USA
-1.0122 0 9 Porto Alegre, 23 BRA
Being headquartered in San Francisco, CA is negatively correlated [..] with having a successful exit
On the other hand, Silicon Valley appears to be well-represented on that list with four SV locations in the top 20 and ten locations overall. That said, the list isn't very useful without absolute numbers (Mi Wuk Village? A town with 941 people?)
I suspect maybe you're showing (X exits)/(N funded companies). You can't just compare binomial proportions like that, between proportions of different N. You'll have lots of random fluctuation. If Waltham had a couple of exits out of few tries, it will look better than Cambridge, even though the error bars on Cambridge would be much smaller. Maybe do a lower bound of a confidence interval? Sorry for the stats nitpicks.
Yeah, while you can solve it that way, you're not getting any insight from the analysis. Logistic regression assumes all data are known exactly, while error bars are important for this case. Take a look at my comment to the parent for a better way.
Using the "error bar" approach (Wilson score), the following ranking results (need at least 7 startups to make the list):
Good:
stddev Exits Total Place
3.17127075776 7 7 Chapel Hill, NC USA
3.01843089923 143 207 Mountain View, CA USA
2.94392395713 32 42 South San Francisco, CA USA
2.33101381278 15 20 Foster City, CA USA
2.31494411999 7 8 Itasca, IL USA
2.31494411999 7 8 Westford, MA USA
2.29564449793 539 966 San Francisco, CA USA
2.2599048739 102 171 Cambridge, MA USA
2.16920869781 37 58 Cupertino, CA USA
2.15675433868 60 99 Waltham, MA USA
2.14760333093 107 185 Santa Clara, CA USA
2.03695302614 13 18 San Bruno, CA USA
2.02879100247 8 10 Redwood Shores, CA USA
1.96889748562 50 85 Menlo Park, CA USA
1.92488214848 87 157 San Mateo, CA USA
1.88561243541 111 206 San Jose, CA USA
1.87088199283 12 17 Bedford, MA USA
1.863724487 9 12 Brisbane, CA USA
1.863724487 9 12 Alameda, CA USA
1.81215794651 31 52 Burlington, MA USA
1.80795794099 13 19 Los Gatos, CA USA
1.79841328745 398 807 New York, NY USA
1.79232968296 127 244 Palo Alto, CA USA
1.78670614594 86 161 Boston, MA USA
1.76603102228 130 252 Seattle, WA USA
1.76116206965 14 21 Arlington, VA USA
1.725668452 15 23 Aliso Viejo, CA USA
1.69835027929 16 25 Emeryville, CA USA
1.65992493077 115 228 San Diego, CA USA
1.6288650578 21 35 Milpitas, CA USA
1.62697011397 106 211 Sunnyvale, CA USA
1.61878237122 8 11 Chelmsford, MA USA
1.54045615814 9 13 Watertown, MA USA
1.54045615814 9 13 Lowell, MA USA
1.52765254628 23 40 Campbell, CA USA
1.52184678546 63 124 Redwood City, CA USA
1.45979095374 114 240 Austin, TX USA
1.44275765525 12 19 Burlingame, CA USA
1.43452996236 6 8 Calabasas, CA USA
1.42071168128 15 25 Berlin, 16 DEU
1.34271490531 7 10 Belmont, CA USA
1.34271490531 7 10 Venice, CA USA
1.34271490531 7 10 Louisville, CO USA
1.33560665511 20 36 Pasadena, CA USA
1.31683773572 18 32 Lexington, MA USA
1.3079978015 35 69 Portland, OR USA
1.29748756451 8 12 Sterling, VA USA
1.28502636002 37 74 Tel Aviv, 5 ISR
1.27560983396 9 14 El Segundo, CA USA
1.26780216551 12 20 Bothell, WA USA
1.26575403151 39 79 Fremont, CA USA
1.25989899302 42 86 Santa Monica, CA USA
1.16897845975 15 27 Marlborough, MA USA
1.15041829453 30 61 Bellevue, WA USA
1.14641742472 19 36 Morrisville, NC USA
1.12663799508 18 34 Pleasanton, CA USA
1.12663799508 18 34 Woburn, MA USA
1.11830928242 39 83 Boulder, CO USA
1.1055382851 17 32 Bethesda, MD USA
1.1055382851 17 32 Richardson, TX USA
1.0729130146 10 17 Malvern, PA USA
1.06468478349 5 7 Kitchener, ON CAN
1.06468478349 5 7 Surry Hills, 2 AUS
1.06468478349 5 7 Bridgewater, NJ USA
1.04813789306 27 56 Durham, NC USA
1.0299173951 6 9 Petah Tiqva, 2 ISR
1.02697161367 7 11 Tucson, AZ USA
1.02697161367 7 11 Kfar Saba, 2 ISR
1.00242961826 44 99 Vancouver, BC CAN
Bad:
stddev Exits Total Place
-3.03488442406 13 135 Moscow, 48 RUS
-2.24461906751 0 10 Saint Petersburg, 66 RUS
-2.24054924343 2 22 Edinburgh, U8 GBR
-2.1019848289 0 9 Porto Alegre, 23 BRA
-2.1019848289 0 9 Little Rock, AR USA
-2.1019848289 0 9 Jakarta, 4 IDN
-1.99941240406 1 14 Lexington, KY USA
-1.93525993713 0 8 Tallinn, 1 EST
-1.93525993713 0 8 Centennial, CO USA
-1.93525993713 0 8 Reno, NV USA
-1.86110616417 5 30 Columbus, OH USA
-1.74075145471 1 12 Glasgow, V2 GBR
-1.73777813442 0 7 São Paulo, 27 BRA
-1.73777813442 0 7 Taipei, 3 TWN
-1.73777813442 0 7 Livermore, CA USA
-1.73777813442 0 7 Brisbane, 4 AUS
-1.73777813442 0 7 Turku, 15 FIN
-1.73777813442 0 7 Lima, 15 PER
-1.69885420792 2 16 Quebec, QC CAN
-1.69885420792 2 16 Memphis, TN USA
-1.65383249608 9 41 Cleveland, OH USA
-1.62527946863 4 23 New Delhi, 7 IND
-1.61418876295 6 30 Buenos Aires, 7 ARG
-1.56634866608 55 178 Los Angeles, CA USA
-1.53427233432 4 22 Chennai, 25 IND
-1.48872091756 9 38 São Paulo, 2 BRA
-1.40898404732 1 10 Colorado Springs, CO USA
-1.40299548518 21 72 Dublin, 7 IRL
-1.33066279643 4 20 Melbourne, 7 AUS
-1.32829693893 130 356 London, H9 GBR
-1.30824843556 26 83 Pittsburgh, PA USA
-1.2681757003 13 46 Mumbai, 16 IND
-1.24409522273 56 159 Paris, A8 FRA
-1.21002433161 6 25 Hong Kong, HKG
-1.20538559097 1 9 Troy, MI USA
-1.20094804147 12 42 Raleigh, NC USA
-1.18989409055 24 74 Shanghai, 23 CHN
-1.18887014907 45 128 Beijing, 22 CHN
-1.11938393751 2 12 Fayetteville, AR USA
-1.11331820129 19 59 Bangalore, 19 IND
-1.10037104532 5 21 Phoenix, AZ USA
-1.09606612444 35 99 Houston, TX USA
-1.09541919563 29 84 Dallas, TX USA
-1.09414918118 3 15 Santa Ana, CA USA
-1.09414918118 3 15 Hyderabad, 2 IND
-1.05291108276 16 50 Stockholm, 26 SWE
-1.04636163325 59 155 Toronto, ON CAN
-1.0386757642 13 42 Miami, FL USA
-1.02924375409 10 34 Shenzhen, 30 CHN
-1.00129935812 9 31 Calgary, AB CAN
I downloaded the Crunchbase database (startups, acquisitions, exits, investments, ipos, etc) as a CSV file, then calculated the importance of each city based on exits vs. failure. The left column is the standard deviation of the value of that city relative to all other cities considered. That is, Mountain View, for example, is 8.4678 standard deviations above the norm.
Being from the area Waltham is more desirable for business for a lot of the reasons Brin mentions in the article. It's cheaper than living/working in Cambridge and local ordinances have done well to make that a really desirable place to be. For Boston most of the startup tech industries linger most around the ring roads around Boston rather than in Boston properly like it seems SF/SV tends to be. It's also much more practical to attract talent that lives outside Boston and inside Boston in Waltham.
I've worked in Waltham myself, so I'm not dissing on it. I just don't really buy that his statistic was 4 times better. In fact, doing it my way says that neither Waltham or Cambridge is statistically better than the other, although the point estimate for Waltham is a bit higher.
Richardson being in the top 5 makes me very happy, because I have such strong personal and professional ties there... it's basically "my city". I went to both high school and college in Richardson (UTD alumna here), I've worked in two different companies in Richardson (and I'm currently at one in a suburb next door to Richardson, which is also on the list), and it's a great place to both work and live. Much better than anything in SV, IMO.
One of the companies I worked at in Richardson had a very successful exit, too.
Its interesting to see that cities in the Bay Area that are separated by a few miles fall either in the best or worst cities list.
Some examples:
Redwood City, CA vs. Foster City, CA
San Francisco, CA vs. South San Francisco, CA
Mountain View, CA vs. Palo Alto, CA
Its one of those "Love me or hate me, but don't ignore me" moments.
Regardless of whether you succeed or fail, you have to first get into the game. And there are few better places to get into the game than the Bay Area.
But it doesn't break down other cities in a similar fashion. Metro London is absolutely huge, it's larger than all of those listed cities (plus San Jose) combined.
It looks like the small- or mid-sized cities on the best list are almost all part of much larger metropolitan areas though. It looks like it's dominated by cities in the Bay Area (including Silicon Valley) and the regions surrounding LA.
Probably because the data is US-centric. So they will break out neighboring American cities into several 10-100sq mi. chunks but leave European cities that are 600sq mi. in tact.
Define exit? because I'm sure most exits aren't necessarily good/great for founders and employees. A list of successful/profitable exits would be more appropriate
Using logistic regression that way isn't going to do much, since each city has a single nonzero independent variable.
I'd try something like this, stats wise. It is just 10 minutes of effort, but accounts for the different sample sizes. I added one success and failure to each city, i.e. Laplace smoothing. Then I computed a 95% percent confidence interval for each, using binom.test in R (i.e. binom.test(8,9)$conf.int). I sorted by the lower bound.
If the lower bound of one city is above the upper bound of another, then you can say it is a better place, using your metric.
x+1 n+2 lb x1/n2 ub
Mountain View, CA USA 144 209 0.621 0.689 0.751
South San Francisco, CA 33 44 0.597 0.750 0.868
Cambridge, MA USA 103 173 0.518 0.595 0.669
Chapel Hill, NC USA 8 9 0.518 0.889 0.997
Waltham, MA USA 61 101 0.502 0.604 0.700
Cupertino, CA USA 38 60 0.499 0.633 0.754
Foster City, CA USA 16 22 0.498 0.727 0.893
San Mateo, CA USA 88 159 0.473 0.553 0.632
San Jose, CA USA 112 208 0.468 0.538 0.608
New York, NY USA 399 809 0.458 0.493 0.528
San Bruno, CA USA 14 20 0.457 0.700 0.881
Palo Alto, CA USA 128 246 0.456 0.520 0.584
Itasca, IL USA 8 10 0.444 0.800 0.975
Westford, MA USA 8 10 0.444 0.800 0.975
Los Gatos, CA USA 14 21 0.430 0.667 0.854
Arlington, VA USA 15 23 0.427 0.652 0.836
Aliso Viejo, CA USA 16 25 0.425 0.640 0.820
Branford, CT USA 6 7 0.421 0.857 0.996
Milpitas, CA USA 22 37 0.421 0.595 0.752
Alameda, CA USA 10 14 0.419 0.714 0.916
Redwood City, CA USA 64 126 0.417 0.508 0.598
Austin, TX USA 115 242 0.411 0.475 0.540
Berlin, 16 DEU 16 27 0.388 0.593 0.776
Chelmsford, MA USA 9 13 0.386 0.692 0.909
Portland, OR USA 36 71 0.386 0.507 0.628
Tel Aviv, 5 ISR 38 76 0.383 0.500 0.617
Pasadena, CA USA 21 38 0.383 0.553 0.714
Fremont, CA USA 40 81 0.381 0.494 0.607
Boulder, CO USA 40 85 0.361 0.471 0.582
Morrisville, NC USA 20 38 0.358 0.526 0.690
Marlborough, MA USA 16 29 0.357 0.552 0.736
El Segundo, CA USA 10 16 0.354 0.625 0.848
Sterling, VA USA 9 14 0.351 0.643 0.872
Richardson, TX USA 18 34 0.351 0.529 0.702
Bethesda, MD USA 18 34 0.351 0.529 0.702
Burlington, ON CAN 6 8 0.349 0.750 0.968
Oak Brook, IL USA 6 8 0.349 0.750 0.968
Solana Beach, CA USA 6 8 0.349 0.750 0.968
Venice, CA USA 8 12 0.349 0.667 0.901
Belmont, CA USA 8 12 0.349 0.667 0.901
Vancouver, BC CAN 45 101 0.347 0.446 0.548
Atlanta, GA USA 58 140 0.332 0.414 0.501
Montreal, QC CAN 29 64 0.328 0.453 0.583
London, H9 GBR 131 358 0.316 0.366 0.418
Kfar Saba, 2 ISR 8 13 0.316 0.615 0.861
Ottawa, ON CAN 24 53 0.316 0.453 0.596
Irvine, CA USA 44 108 0.314 0.407 0.506
Mclean, VA USA 15 30 0.313 0.500 0.687
Minneapolis, MN USA 23 51 0.311 0.451 0.597
Beverly Hills, CA USA 10 18 0.308 0.556 0.785
Toronto, ON CAN 60 157 0.306 0.382 0.463
Surry Hills, 2 AUS 6 9 0.299 0.667 0.925
Kitchener, ON CAN 6 9 0.299 0.667 0.925
Doylestown, PA USA 5 7 0.290 0.714 0.963
Ames, IA USA 5 7 0.290 0.714 0.963
Salt Lake City, UT USA 27 66 0.290 0.409 0.537
Irving, TX USA 10 19 0.289 0.526 0.756
Gaithersburg, MD USA 10 19 0.289 0.526 0.756
Tokyo, 40 JPN 24 58 0.286 0.414 0.551
Gent, 8 BEL 4 5 0.284 0.800 0.995
Fuzhou Shi, 3 CHN 4 5 0.284 0.800 0.995
Lake Forest, IL USA 4 5 0.284 0.800 0.995
Sunrise, FL USA 4 5 0.284 0.800 0.995
Fredericton, NS CAN 4 5 0.284 0.800 0.995
Princeton, NJ USA 11 22 0.282 0.500 0.718
Baltimore, MD USA 15 33 0.281 0.455 0.636
Paris, A8 FRA 57 161 0.280 0.354 0.433
Wilmington, DE USA 9 17 0.278 0.529 0.770
San Antonio, TX USA 14 31 0.273 0.452 0.640
Beijing, 22 CHN 46 130 0.272 0.354 0.442
Oakland, CA USA 16 37 0.271 0.432 0.605
Hamburg, 4 DEU 12 26 0.266 0.462 0.666
Westminster, CO USA 8 15 0.266 0.533 0.787
Philadelphia, PA USA 20 50 0.264 0.400 0.548
Houston, TX USA 36 101 0.264 0.356 0.458
Cambridge, C3 GBR 18 44 0.263 0.409 0.568
Ann Arbor, MI USA 14 33 0.255 0.424 0.608
> I'd try something like this, stats wise. It is just 10 minutes of effort, but accounts for the different sample sizes. I added one success and failure to each city, i.e. Laplace smoothing. Then I computed a 95% percent confidence interval for each, using binom.test in R (i.e. binom.test(8,9)$conf.int). I sorted by the lower bound.
Could you enlighten the statistically-impaired on what this has accomplished, compared to the raw data?
This is a called a binomial process. We want to estimate the true proportion of exits for a given city. A city has x exits out of n trials, so one estimate of this proportion is just x/n. However, if you have a bunch of cities like this list, you're going to have cities that by random chance end up with a fraction close to 1. That doesn't mean that startups there are guaranteed to succeed. It means if you flip a coin 5 time, and replicate it 1000 times, you're doing to have some runs of 5 heads, and some runs of 5 tails. If you kept going, you would find the proportion approaching 1/2 (for a coin) for all cases.
So the problem is how you compare a city with 5/6 exits like Branford, CT USA with 143/208 like Mountain View. Is Branford that much better because 5/6 > 143/208 ? Mostly all you know is that the error in your estimate is much larger for Branford than for Mountain View, because your value of n is 6 vs 208. You can't say with statistical confidence that Branford is better.
So one trick to punish the little n locations is to do some smoothing. Laplace smoothing is to add 1 for all outcomes, so 1 success to x and one failure, meaning we add 2 to n. That also means that nothing gets exactly to 0.0 or 1.0. The odds in Saint Petersburg Russia aren't really 0.0 because the were 0/11. There is some chance you could succeed, so it gives you a better estimate of "unseen events".
The next thing you want to do is look at confidence intervals, rather than our point estimate of x/n or even (x+1)/(n+2). There are a number of formulas you can use, I used one built into R, a statistical modelling language. This gives you a lower and upper bound on your true estimate of the proportion. If the bounds is exact, then 95% of the time the interval will contain this true, unknown proportion.
The bounds on my smoothed counts are:
x+1 n+2 lb x1/n2 ub
Mountain View, CA USA 144 209 0.621 0.689 0.751
Branford, CT USA 6 7 0.421 0.857 0.996
Los Angeles, CA USA 56 180 0.244 0.311 0.384
So the true estimate of Mountain View is somewhere between 0.621 to 0.751, while Branford CT is between 0.421 and 0.996. Since these estimates overlap, we can't really say one is better than the other. Also consider LA, which has range of 0.244 to 0.384. Since 0.384 < 0.421, we could say that LA has a worse exit ratio than either Brandford or Mountain View, with 95% confidence.
To sort, it is often good to be conservative and use the lower bound. I used a 95%, which is good for saying Branford is better than LA, but might be a bit large for sorting. You could use a 90% or even 80% interval for that, if desired.
It is really crucial to take into account what you don't know when comparing fractions based on different values of n.
I'm very pleased to see Los Angeles beat Mi Wuk Village in this list. The LA vs. Mi Wuk startup rivalry is legendary and, I must admit, I was quite embarrassed about LA's performance when I saw the first list.
Interesting that Austin TX is not present. Neither is Los Angeles.
So ... I guess what you're putting here is the number of standard deviations from the mean for each city, but only for cities above the mean? And San Francisco is below the mean?
It's entertaining how badly some cities do compared to other cities that are only a few minutes away (Raleigh and Morrisville). Could this be because the startups founded in Raleigh are NC State offshoots?
I would argue "Surry Hills" should be "Sydney" and "Notting Hill" should be "Melbourne" since there are no other suburbs listed for those cities in the list.
Updated (min total startups to be considered = 7):
Worst:
stddev Exits Total City
-8.0733 55 178 Los Angeles, CA USA
-5.5089 13 135 Moscow, 48 RUS
-4.6686 63 124 Redwood City, CA USA
-4.1965 12 42 Raleigh, NC USA
-3.5345 9 21 La Jolla, CA USA
-3.4526 2 16 Quebec, QC CAN
-3.3204 11 36 Rockville, MD USA
-2.9826 4 15 Lucerne Valley, CA USA
-2.9103 2 22 Edinburgh, U8 GBR
-2.8953 2 10 Netanya, 2 ISR
-2.8038 0 7 Livermore, CA USA
-2.7531 2 9 Marina Del Rey, CA USA
-2.7115 9 24 Berkeley, CA USA
-2.7108 3 13 Torrance, CA USA
-2.6849 26 83 Pittsburgh, PA USA
-2.6366 3 15 Santa Ana, CA USA
-2.6334 0 8 Centennial, CO USA
-2.6226 5 30 Columbus, OH USA
-2.5629 2 9 Edison, NJ USA
-2.5578 1 10 Colorado Springs, CO USA
-2.4961 1 14 Lexington, KY USA
-2.3659 2 16 Memphis, TN USA
-2.3645 1 8 Sacramento, CA USA
-2.2869 1 8 Kista, 26 SWE
-2.2535 11 33 Santa Barbara, CA USA
-2.2321 2 10 Lake Forest, CA USA
-2.1807 3 12 Concord, MA USA
-2.1566 0 8 Reno, NV USA
-2.1472 0 9 Little Rock, AR USA
-2.1286 59 155 Toronto, ON CAN
-2.0520 4 12 Wilmington, MA USA
-2.0374 4 12 Westborough, MA USA
-1.9802 9 30 Indianapolis, IN USA
-1.8886 0 10 Saint Petersburg, 66 RUS
-1.8813 0 8 Tallinn, 1 EST
-1.8615 2 8 Worcester, MA USA
-1.7769 14 41 Brooklyn, NY USA
-1.7189 1 12 Glasgow, V2 GBR
-1.6803 1 8 Huntsville, AL USA
-1.6410 1 9 Troy, MI USA
-1.6338 0 7 Taipei, 3 TWN
-1.6337 0 7 Brisbane, 4 AUS
-1.6184 6 20 Newport Beach, CA USA
-1.5786 0 7 Turku, 15 FIN
-1.5739 2 7 Blue Bell, PA USA
-1.5707 0 9 Porto Alegre, 23 BRA
-1.5668 0 9 Jakarta, 4 IDN
-1.5612 9 41 Cleveland, OH USA
-1.5600 4 12 Jersey City, NJ USA
-1.5029 6 13 Acton, MA USA
-1.4916 2 9 Lausanne, 23 CHE
-1.4875 0 7 Sao Paulo, 27 BRA
-1.4645 1 8 Fort Worth, TX USA
-1.4637 1 8 Manchester, I2 GBR
-1.4199 4 20 Melbourne, 7 AUS
-1.4106 1 8 Honolulu, HI USA
-1.4096 8 25 Orlando, FL USA
-1.4001 1 8 Sarasota, FL USA
-1.3922 2 7 Maple Grove, MN USA
-1.3811 5 13 New Haven, CT USA
-1.3667 5 19 Albuquerque, NM USA
-1.3379 2 8 Wellington, G2 NZL
-1.3361 0 7 Lima, 15 PER
-1.3238 2 9 Bristol, B7 GBR
-1.3162 25 64 Denver, CO USA
-1.3004 4 13 Tempe, AZ USA
-1.2051 19 48 Philadelphia, PA USA
-1.2004 2 12 Fayetteville, AR USA
-1.1954 2 9 Markham, ON CAN
-1.1660 4 11 Maynard, MA USA
-1.1381 2 8 Saint Louis, MO USA
-1.1364 1 7 Chattanooga, TN USA
-1.1294 2 7 Richmond Hill, ON CAN
-1.1272 8 20 Newton, MA USA
-1.1053 3 9 Goleta, CA USA
-1.0719 1 7 Glendale, CA USA
-1.0718 6 17 Englewood, CO USA
-1.0635 2 10 Hartford, CT USA
-1.0437 5 19 Rochester, NY USA
-1.0119 6 22 Omaha, NE USA
-1.0092 10 21 San Carlos, CA USA
Updated (min total startups to be considered = 7):
Best:
stddev Exits Total Place
10.9946 143 207 Mountain View, CA USA
7.1669 130 356 London, H9 GBR
6.7790 56 159 Paris, A8 FRA
6.5101 398 807 New York, NY USA
6.0217 32 42 South San Francisco, CA USA
5.4159 14 21 Arlington, VA USA
5.3030 15 25 Berlin, 16 DEU
5.0688 44 99 Vancouver, BC CAN
4.9659 60 99 Waltham, MA USA
4.5533 19 36 Morrisville, NC USA
4.5464 130 252 Seattle, WA USA
4.4976 23 56 Tokyo, 40 JPN
4.4164 39 79 Fremont, CA USA
4.2893 37 58 Cupertino, CA USA
4.1535 15 20 Foster City, CA USA
4.1358 57 138 Atlanta, GA USA
4.1111 107 185 Santa Clara, CA USA
3.8654 8 12 Sterling, VA USA
3.8488 9 17 Irving, TX USA
3.7888 7 8 Itasca, IL USA
3.7762 35 69 Portland, OR USA
3.6685 9 21 Jacksonville, FL USA
3.6062 17 32 Bethesda, MD USA
3.5609 17 32 Richardson, TX USA
3.5188 114 240 Austin, TX USA
3.5133 31 52 Burlington, MA USA
3.5031 28 62 Montréal, QC CAN
3.4952 20 36 Pasadena, CA USA
3.4738 13 19 Los Gatos, CA USA
3.4369 7 7 Chapel Hill, NC USA
3.2760 7 8 Westford, MA USA
3.2017 12 23 Eden Prairie, MN USA
3.1571 5 7 Surry Hills, 2 AUS
3.1014 14 41 Copenhagen, 17 DNK
3.0194 15 23 Aliso Viejo, CA USA
2.9798 17 42 Cambridge, C3 GBR
2.9354 13 18 San Bruno, CA USA
2.9326 10 29 Seoul, 11 KOR
2.9201 18 34 Woburn, MA USA
2.8992 8 11 Chelmsford, MA USA
2.8750 16 37 Las Vegas, NV USA
2.8428 6 12 Oxford, K2 GBR
2.8319 21 35 Milpitas, CA USA
2.6148 7 11 Tucson, AZ USA
2.6053 7 18 Zurich, 25 CHE
2.5816 6 16 Gurgaon, 10 IND
2.5450 7 10 Louisville, CO USA
2.5402 6 10 Cologne, 7 DEU
2.5291 8 18 Oslo, 12 NOR
2.5089 11 27 Cincinnati, OH USA
2.4926 12 30 Helsinki, 13 FIN
2.4697 16 50 Stockholm, 26 SWE
2.4594 9 16 Beverly Hills, CA USA
2.4512 50 85 Menlo Park, CA USA
2.4479 15 41 Madrid, 29 ESP
2.4423 12 17 Bedford, MA USA
2.3961 102 171 Cambridge, MA USA
2.3345 6 8 Calabasas, CA USA
2.3127 9 12 Alameda, CA USA
2.2776 14 31 Baltimore, MD USA
2.2581 4 12 Istanbul, 34 TUR
2.2397 11 24 Hamburg, 4 DEU
2.2114 19 50 Washington, DC USA
2.1985 14 28 Mclean, VA USA
2.1772 8 10 Redwood Shores, CA USA
2.1754 37 74 Tel Aviv, 5 ISR
2.1736 6 14 Buffalo, NY USA
2.1232 22 49 Minneapolis, MN USA
2.1157 6 11 Orem, UT USA
2.1067 19 59 Bangalore, 19 IND
2.0824 45 128 Beijing, 22 CHN
2.0262 27 56 Durham, NC USA
2.0083 10 27 Boca Raton, FL USA
1.9988 5 7 Bridgewater, NJ USA
1.9870 111 206 San Jose, CA USA
1.9746 4 7 Delft, 11 NLD
1.9710 8 15 Wilmington, DE USA
1.9640 6 10 Stamford, CT USA
1.9471 9 12 Brisbane, CA USA
1.9421 8 19 Greenwood Village, CO USA
1.8905 5 13 Dubai, 3 ARE
1.8893 5 13 Lund, 27 SWE
1.8747 7 17 Brentwood, TN USA
1.8449 5 11 Minnetonka, MN USA
1.7944 5 7 Kitchener, ON CAN
1.7868 16 25 Emeryville, CA USA
1.7221 23 40 Campbell, CA USA
1.7141 4 7 Alachua, FL USA
1.7124 86 161 Boston, MA USA
1.6797 4 8 Duluth, GA USA
1.6639 87 157 San Mateo, CA USA
1.6414 539 966 San Francisco, CA USA
1.6382 5 8 Conshohocken, PA USA
1.6130 4 7 Costa Mesa, CA USA
1.6076 14 43 Barcelona, 56 ESP
1.5818 9 14 El Segundo, CA USA
1.5665 4 8 Charleston, SC USA
1.5409 5 11 Reading, K7 GBR
1.5380 4 9 Halifax, NS CAN
1.5129 5 10 Silver Spring, MD USA
1.5129 12 34 Sydney, 2 AUS
1.5071 7 11 Kfar Saba, 2 ISR
1.4988 9 13 Lowell, MA USA
1.4875 30 61 Bellevue, WA USA
1.4408 3 9 Belfast, R3 GBR
1.4257 7 13 King Of Prussia, PA USA
1.3873 6 9 Petah Tiqva, 2 ISR
1.3862 5 12 Annapolis, MD USA
1.3560 12 20 Bothell, WA USA
1.3433 39 83 Boulder, CO USA
1.3420 13 29 San Antonio, TX USA
1.3402 11 32 St Louis, MO USA
1.3264 23 51 Ottawa, ON CAN
1.3214 18 41 Reston, VA USA
1.3128 5 10 Munchen, 2 DEU
1.3105 4 7 Red Bank, NJ USA
1.2965 3 7 Chatsworth, CA USA
1.2788 5 11 Abingdon, K2 GBR
1.2683 5 8 Westlake Village, CA USA
1.2628 3 7 Chaoyang, 19 CHN
1.2391 3 8 Prague, 52 CZE
1.2274 4 14 Rio De Janeiro, 21 BRA
1.2240 10 17 Malvern, PA USA
1.2201 7 14 Mississauga, ON CAN
1.2171 13 37 Tampa, FL USA
1.1791 3 8 Guildford, N7 GBR
1.1749 3 7 Westport, CT USA
1.1657 18 43 Munich, 2 DEU
1.1547 6 11 Cary, NC USA
1.1455 4 10 Stuttgart, 1 DEU
1.1425 11 34 Amsterdam, 7 NLD
1.1243 3 7 Boise, ID USA
1.1113 7 13 Westminster, CO USA
1.1013 13 31 Ann Arbor, MI USA
1.0974 3 10 Sofia, 42 BGR
1.0831 4 9 Melbourne Beach, FL USA
1.0641 7 10 Venice, CA USA
1.0534 8 20 Providence, RI USA
1.0360 7 10 Belmont, CA USA
1.0331 4 7 South Jordan, UT USA
1.0307 7 24 Vienna, 9 AUT
1.0274 4 7 Newtown, PA USA
1.0259 4 15 Milan, 9 ITA
1.0135 3 7 Fort Collins, CO USA
1.0079 6 13 Broomfield, CO USA
I think there's a lot of factors at play when you're founding a company and what Brin is hopefully saying is that given the current conditions SV is difficult for early stage (bootstrapping or seeding), which seems to be true.
In 1998 things were very different so of course they could afford to be in the menlo park area and reap the rewards of the VC network. Now, to stay in the area you need to fund heavily up front which leaves you less time to build that awesome product before funding dries up (or worse, you have to sell too much of your company in order to survive).
The reality though is you can try to start up a company in a place like Boca Raton and unless you have some previous connections and major capital you're not going have the funding network and excess of high tech workers your company needs to really take off. Magic Leap basically went this route and seem to be doing fine but most companies aren't that high profile (and well funded).
This is fine advice if you're talented enough to startup without a co-founder. It's also fine advice if you're lucky enough to find a co-founder in your local talent pool.
However, if you find that starting up is beyond your individual skill set, and the people you enlist to help you come over to your garage and spend 10 mins programming and the rest of the time on YouTube and Reddit, you may want to move to a better startup ecosystem to find a more ambitious co-founder.
I don't think there is universal answer. It depends a lot on your personal situation (connections/resources available/citizenship/lifestyle preferences).
I think a remote, distributed company is by far the best option. There are drawbacks, but the increased size of the labor pool, opportunity to hire talented folks with lower salary requirements, lower overhead costs, and (arguably) higher productivity/ work life balance just to name a few.
> Chicago had just 31 total exits during the period, equal to Raleigh. Austin, Texas, had 86; Washington, D.C., saw 87; New York posted 98; and the San Francisco Bay Area recorded 613. While Chicago had 14 deals that produced returns of more than 10-fold, San Francisco had 153 such deals.
If your goal is to start an enduring & profitable business, number of exits is the wrong metric. I get annoyed when total number of exits is assumed to be the only metric that counts. It's clearly a metric VCs care about, though.
If you're a founder, not a VC, you would probably want another metric for the rate of return on your sweat from founding to exit.
There's a TVM calculation in there, to normalize the dollars you got paid on exit to the years that you did the work. And there's an opportunity cost for not being a salaried employee of an existing business while your were building your new business.
In the end, you probably want to reduce it to $X/week, run the calculation for all founders, and then aggregate by metropolitan area to see which city is best.
If your business becomes self-sustaining without a buyout, merger, or IPO, you can probably run the calculation using a TVM for a presumed perpetuity after the day you finally start paying regular dividends, along with your equity value as a lump sum. And if it goes bankrupt, you could end up with a negative value if you chose to pay yourself a salary lower than you could have earned elsewhere.
I dunno. Maybe if you're already around there. Remember the Superfan sketched on Saturday Night Live? There's a reason those guys were big. It gets cold in the winter.
Atlanta maybe? the south east has a disproportionately low number of tech companies relative to number of people (whether that's equivalent to talent, who knows?)
Quite a few of the tech jobs in the SE US are clustered around military bases. If you want to start a military-industrial complex contractor, and for some bizarre reason you cannot do it in the DC Sprawl, what you can do is find a {choose 1 or more: disabled, native American, veteran, female, or other lawfully preferred minority} figurehead to own 51% of your business, grab some office space near a base, and start submitting bids, which you then subcontract to someone with actual employees. Then you start hiring employees and have them work on-site. Then you get acquired by a giant like SAIC or CSC or Raytheon or Boeing. That makes the local labor pool employed by a company that is probably headquartered elsewhere.
The downside of this model is that you never get to actually build anything cool, so you either can't attract top talent, or they hate working for you and your neurotic government customer. So you have to slap on the silver handcuffs (high salary) to keep them from moving away.
It sort of poisons the well for other kinds of tech businesses. Top talent either doesn't want to be there or costs more than you might expect. Mediocre talent is happy to warm a chair to pad out a contract for the same salary you would pay to do actual work. And the dregs are the same there as everywhere else.
If you're itching to start a company out of a garage, then you shouldn't pick up and move to Silicon Valley, says Russian immigrant Sergey Brin, who cofounded Google at Stanford University, after immigrating to the United States with his family at the age of 6. With help from an advisor who put him in touch with Larry Page while they were studying at Stanford, they implemented a new algorithm, called Page Rank, on top of a data mining system Brin was already developing. With further help from the Stanford community's network, they soon received a check for US$100,000 from Andy Bechtolsheim, co-founder of Sun Microsystems, written to a company that did not yet exist. About a year later, they announced closing a $25 million round from Sequoia Capital, who suggested hiring an external CEO in the form of Eric Schmidt. Within 4 years Microsoft started making bids for the company, though Google eventually went public instead. Today Sergey Brin is one of three people listed as 11th richest in the world, with a net worth of US$39.2 billion.
Recently speaking at the Global Entrepreneurship Summit, he said onstage, "It's easier to start a company outside the Valley than in it."
In a follow-up interview, he was asked to reflect on the $100,000 check he received from Andy Bechtolsheim of Sun Microsystems, before Google existed. Now that he himself was worth billions, would he write a similar check based on a pitch from someone living in Kansas, the interviewer asked. Sergey Brin laughed.
"That would sort of contradict what I've been saying here," he said with a laugh. "Maybe try kickstarter? But if you ever do move to Silicon Valley definitely reach out."
He then got very serious:"Seriously though. We live in a connected world. It doesn't matter where you live. It matters what you build."
He then apologized saying he had to run, and drove off in a pre-production Tesla Model 3 in fiery red, produced by South African immigrant Elon Musk, who made his fortune at PayPal, based in San José, and now runs Tesla Motors based in Palo Alto. An odd choice for Brin to be driving, as Google is said to be developing its own driverless vehicle.
That is, if a few startup kids in Detroit don't beat them to the punch. You can read about the extraordinary results of the Detroit startup company (find link) . One thing's for sure: just as Brin eloquently stated, there's nothing stopping them.
That's great and all but access to capital, even angel capital, is much easier in the Valley than anywhere else in the world. You come here and live threadbare, but you still derisk your startup.
How on earth does taking capital derisk your startup? It's like throwing gasoline on a bonfire, yes it burns a lot faster, but also shorter. Early money is the most expensive money you'll ever take, and if you're reliant on it you're dead when it dries up.
Find a viable business with a skeleton crew, grow with your profits, and then pour on the gas if you want when a repeatable pattern is found. It's not as glamorous, but you're more likely to create a solid business instead of crashing and burning like most SV startups.
Imagine how "Silicon Valley" series would be boring, if Pied Piper founders started their company in cheap US state/country with their own savings. Distributed company would be even more boring :-)
(1) Page and Brin started Google Inc successfully in SV so it seems contradictory to advise others not to start a company there
(2) the meaning of the word "start" as Brin is using it
As for (1), Larry and Sergey had the luxury of building Google (the product -- not the formalized "incorporated" Google Inc) inside of Stanford and therefore using the university's computers and bandwidth for free. Piggybacking on Stanford's resources to create an "mvp" was like getting "AWS" for free. A lot of startups don't have that situation. Therefore, a founder may be better off working out the parents' garage in Kansas to keep expenses low. The free room & board is money that would have been spent on a crappy apartment in SV. Instead take those savings and use it to pay for the AWS hosting costs, or a 2nd programmer.
For (2), I think Brin is talking about "start" as in start from zero with an idea on a paper napkin, before an MVP, and before traction from others (users and investors). Zuckerberg started Facebook in Boston and after Sean Parker got interested, he moved Facebook to the more expensive SV. Bill Gates started Microsoft in Albuquerque, New Mexico and made profits before he moved the company to the more expensive Seattle. In other words, you can have a "low profile startup" in your local town that's under the radar which lets you keep expenses very low. You can later have a "high profile startup" in Silicon Valley where the very high costs are worth it to hire from the talent pool and network with other entrepreneurs and investors. That delayed move to SV is what Brin called "that second step".