The Stanford Startup and the MIT Startup
fpgacomputing.blogspot.com
fpgacomputing.blogspot.com
The other company has great marketing.
One of them is playing a positive-sum game. The other is playing a zero-sum game: http://en.wikipedia.org/wiki/Zero-sum_game
I don't know that one of these really represents MIT and the other really represents Stanford, but I can tell you that the technology company is adding value to the world in a way that the marketing company is not.
As pg puts it, writers who need to come up with startup ideas "come up with an idea that sounded plausible, but [is] actually bad."
The obvious case is a commodity, like wheat or soybeans or oil, where nobody bothers branding it at all until they get to a retail product. In the retail marketplace, the primary differentiation between one kind of cleaning product and another is the marketing, advertising, and associated superficial differentiations: coloring, fragrance, and so forth. That's why the largest advertising budgets are for companies with similar competitive products: GM/Honda/Ford/Nissan/Toyota/Hyundai; Procter&Gamble, L'Oreal, Estee Lauder; Samsung/Sony/HP; Progressive/Allstate/State Farm.
They aren't advertising to get new customers into the market: they're advertising to shift customers to competing suppliers.
Samples pulled from adbrands.net list of top 100 US advertisers. There are many, many more clusters there to illustrate the point.
Going back to the stanford example, great marketing companies understand user problems and design products that solve that problem in a scalable way. Awareness (via advertising) is only one aspect of marketing.
So while advertising in a commodity market might be a zero sum game, I contend that true marketing is not.
Think about it like this: In economics it's useful to assume perfect information so the market prices and clears and generally works. In the real world perfect information is extremely false, there's a cost to getting new information, a maximum processing speed for humans to digest new information, and a maximum amount of information humans can store (at or below max processing speed * hours lived).
Marketing (the advertising side) is the mechanism by which information is disseminated to consumers.
Marketing (the product-design side) is the mechanism by which information is gathered by producers.
Marketing is the mechanism by which the market attempts to overcome the lack of perfect information.
Marketing (ceteris paribus) reduces transaction costs, and thereby adds real value to the world.
Marketing (in its current form) does not attempt to overcome a lack of information. It exploits the lack of information.
See: The consumer driven debt-march of the American people.
"When I walk past a shop, I am amazed at all the things that man can do without"
It might not be the most efficient net outcome, as a better solution might otherwise exist, but if there is an exploitable lack of information that is exploited then by definition the information in the system has increased.
Imagine a world where there is all of this wonderful technology, but no one knows it exists or it is not applied to their problems.
I think both companies are valuable and great companies should strive to do both well (tech and market development)
To me, the crux of this article seems to be that investors need to be more scientifically literate.
Costs for sustainably produced chemicals are higher, but
the founders maxed out their credit card buying a
wholesale shipment and were able to sell a premium retail
product at a small profit.
It is, but it's worth noting two important aspects here:
1. Unlike most of the audience of Hackernews, most of the people to whom you pitch a product don't know what a zero-sum game is, and learning about it is well beyond their abstract reasoning abilities.
2. Even if they knew, they would simply continue to unconsciously hope they'll end up with one of the positive terms.
Impact should be measured after companies have existed for years, not based on what they first set out to accomplish (company goals inevitably change). The Stanford company built a marginally improved product, and became sustainable. Since marginally improved products aren't really defensible (without a strong brand name), they'll likely continue to iterate on their product and in the long run will improve the tech drastically. Whether the intention or not, essentially they make money and use profits to invest in long term R&D, while the MIT company takes investor money up front to invest in R&D.
I think both approaches are suitable for different circumstances (based on background of founders, current state of market, whether radical or incremental improvements are needed to solve the core problem, etc). I don't like how people call something more or less worthwhile to work on, because wide distribution for a marginally improved product is better for the world than poor distribution for a revolutionary technology that never leaves a research lab.
In the original story the "New Jersey" guy (but from Berkeley) was Bill Joy and the MIT guy was Daniel Weinreb (who was an HN user). If you're interested, you should see his blog post about it:
http://web.archive.org/web/20121107034606/http://danweinreb....
EDIT: Changed second paragraph for clarification.
Here's a small subset:
* Alveo Energy
* Amprius
* Blue River Technology
* C3Nano
* Momentum Machines
* QuantumScape
* Solum
Although if the OP's simple model had a category for startups simultaneously validating customer demand and driving technological advantages, most of these companies would be there.
That website is...erm...bare-bones.
Not saying there isn't much "pure technology" coming out of Stanford (there most certainly is) but you get the feeling that Boston-area startups are wired to find, as Peter Thiel would say, the 0 to 1 markets vs. the 1 to n markets of globalization (the West Coast).
In Boston, there appears to be a bigger emphasis on hi-tech vs. the latest subscription service or cloud provider coming from SV. But the SV culture encourages that, VC's cut checks for products with "traction" and more often than not, skip pure technology (be it medical devices, therapeutics, chemicals, hardware, and hard/expensive/timely ventures).
This is rather unfortunate since we are in need of better startup companies and technologies, and they are in need of funding. There are many reasons for this (one being the "easy way out biz model replication" phenomenon used in the ammonia subscription model example in the post). And another is the lack of talent, particularly great talent going to well-funded SnapChat #4 and other similar, non-groundbreaking companies.
Unfortunately, I don't see the face of venture capital changing much with regards to funding innovative and risky technologies. They of course are wired to big returns on 1/100 social company exits (most at least) and "technology" doesn't really matter (again, to most).
Can you imagine SV putting money into Fairchild Semiconductor or Intel today? Won't happen.
Don't let the excuses of "we infested in Barracuda and Cisco" fool you - the focus is on "how many users do you have for the latest hype driven app on the App Store."
I think the major problem is the lack of risk being taken on great technology vs. the easy way out. AKA, some VCs should grow some balls and invest in the future vs. the very-near future. Being an MBA out of a top-10 school and having check writing power to fund "startups" is also a negative since a lot of these guys know the spreadsheets - NOT the silicon.
But I digress.
Both schools are unique in their culture but share similarities in brilliance. The best outcome is too see great business as marketing intersect with powerful technology. Then we can have real great companies on the horizon again. As a West Coast founder myself, it's actually an honor and a humbling experience to be working with people from both schools and it's clear as day just how different the mindset is.
"... reducing investors' appetite for risk doesn't merely kill off larval start-ups, but kills off the most promising ones especially. Startups yield faster growth at greater risk than established companies. Does this trend also hold among startups? That is, are the riskiest startups the ones that generate most growth if they succeed? I suspect the answer is yes. And that's a chilling thought, because it means that if you cut investors' appetite for risk, the most beneficial startups are the first to go..."
Understanding this could mean a lot to SV VCs but then I'm sure they understand it already.
Plenty of stupid, useless (or even just nonprofitable) companies experience meteoric growth until a fad passes.
Much of VC actually grew out of Fairchild. Eugene Kleiner of KPCB was one of the Fairchild founders, and Don Valentine was a Fairchild sales+marketing exec before starting Sequoia. Those two firms essentially invented modern day VC investing, and you could say Fairchild was the first experiment that worked.
It's also worth noting how much VC has grown since then. Intel only raised 2.5MM before going public. Technology is now much more broad, and there's more money to fund diverse companies. Back then, tech was a niche industry.
Hard science is certainly still getting funded out here -- see Counsyl, LightSail Energy, etc. -- but I think lots of investors are now looking at existing industries that can be flipped via software (cue pmarca's "software is eating the world" neo-adage). The "Stanford startup" that Amir describes is moreso a tech-enabled company, rather than a hard tech company. They're both valuable, but should be seen as apples and oranges.
Maybe it's because I went to MIT and am perhaps foolishly ambitious, but I personally get frustrated when I see super talented people go after easy industries. The one that comes to mind is house cleaning services. It's possible I just don't get the potential, but it's disheartening to see ultra-educated young people trying to figure out how to disrupt a low-margin industry dominated by mostly illegal immigrant workers. Not something I'd want in my epitaph I guess.
The sweet spot is probably somewhere in the middle. Work with wicked smart people, have audacious goals, and focus on shipping and being relevant today.
As an aside, I don't know why everyone is hating on Snapchat. They are riding a massive wave -- the social behavioral shift in how people share online -- which is really not to be underestimated. It's strategic for them to fly under the radar and be disregarded as a stupid toy right now. But if you look at the team they've gathered, see their metrics, or actually talk to Evan, you immediately know that it's something special on the order of YouTube or Twitter.
It's the same reason people hate on claims of perpetual motion machines. Their entire selling point is known to be impossible.
The analogy which is usually made with DRM for media is flawed because of the mass audiences for ripped films and music. If one person of the hundreds of thousands who pays for a DRM protected music track cracks the protection off and posts the raw file, the protection is substantially devalued. A snapchat sent to a few people will not be forwarded unless one of the recipients has put forethought into doing so.
Also, once your friends realize you're using the "Snapchat recorder" they'll stop sending you snaps entirely.
That's strange, I'm pretty sure I own my own phone.
they read bits from the camera api, transmit them, and draw them on the recipients screen. they don't need to be compatible with any 3rd party like DRM-restricted systems.
Which means it's available, unencrypted, at the framebuffer.
https://itunes.apple.com/us/app/snaphack-pro-for-snapchat/id...
https://play.google.com/store/apps/details?id=de.innovationz...
Mike Markkula did, and that's the classic story of VC as in Venture Capital: He provided critical funding in a very risky business attached to also very specific conditions (part of the funding was a loan, for example), and in exchange he asked for a high return.
What we have today is NOT venture funding: We have funding without risk. And that's the OP's point.
We need more VC's like Mike Markkula today, no doubt.
Both Founders Fund and DFJ seem to take larger bets. I wouldn't count on AngelList to fund the next Fairchild.
What is the possible business models? Just because something is popular doesn't mean it is a good business. No one has figured out how to effectively monetize chat going all the way back to the IRC / ICQ days (or even farther back to BBS systems). Chat has always been very popular, but no one has figured out how to make a business out of chat. If snapchat was working on that problem, it might be more respected in the startup world.
(Not sure why the original TC link is broken. Weird.)
http://webcache.googleusercontent.com/search?q=cache:2Eqnnea...
That market just hasn't been cracked in non-Asian markets.
Less technical, consumer-focused startups from Stanford tend to get more press. But there are a lot of successful, high-tech startups from Stanford (often founded by engineering PhDs and professors) that you will never read about on TechCrunch, HN or any of the mainstream "tech news" sites.
http://online.wsj.com/news/interactive/NBT092012?ref=SB10000...
VCs take a bigger chunk at a lower price, because the risk/reward equation is different and they may need a lot of money, whereas unless you're racing for the scale of Facebook or Twitter, your mobile app doesn't need to raise that kind of money.
That sounds like an exciting setup to me. Also gives me hope for all the fresh PhDs you hear about lamenting their career options outside of academia. Then again, I suppose these ventures are quite limited, and risky, so you can't expect "MIT Startups" to serve as sanctuaries for Ivory Tower escapees.
More seriously: what these two companies need to do is merge. One has no product, and the other is just intellectual property assets no one is selling. Together, they'd make a mean business.
As a side note, Boston is about to be overtaken by NYC as the leading VC market on the East Coast.
http://www.theatlantic.com/magazine/archive/2013/10/the-boom...
This is such a perfect example.
[edit: And they've already proven their technology. And, it's easy for investors to vet their product, because a lot of investors are coffee snobs.]
Last but not least, the YC embassadors in the Boston area and alumni (Dropbox etc.) did a great job in sharing their experiences.
This so-called "Stanford model" looks more likely something about old companies. I think for most older generation of people, for whom "marketing" is really what's it's all about(most of the executives in old companies back in the 70s 80s were promoted from marketing I believe), that is their view of "starting a new business". However, they didn't innovate, not exactly because they didn't want to, but because they didn't know how to, in an already stagnant field. This is why I don't really believe this is the case for today's tech-oriented companies. In Peter Thiel's terms(I don't really like him, just using a reference here) this would not even be a "startup" at all(0 to 1). I mean, I don't believe there exist more tech startups which make themselves big by just extensively marketing existing technologies, than those which lead the innovation of new ones. Correct me if I'm wrong. I don't know a lot about these things anyways.
My point though is this. Even if these MIT PhD's don't understand marketing and can't raise enough to keep their company afloat, this hypothetical technology would be worth many many billions of dollars to a deep-pocketed chemical supply company. So the comparison is a bit lopsided.
Even if a technology company can raise funding and grants, sustainable revenue from licensing or contracts should be the first priority for the company or it won't be able to stay together.
The technology is 20 years ahead of the need, or it takes 20 years to work out the kinks in the process. So the patent expires just as it becomes valuable.
But once the patent expires, no company can get a sustainable advantage. Thus, the net incremental producer surplus gets competed down to zero. However, consumer surplus is increased by zillions of dollars, because the technology has effectively been "donated" to humanity through patent expiration.
But this would be a terrible story for VCs, because they didn't make any money off of it. Instead, 100% of the benefits went to humanity.
Big Chemical might make a huge capex investment to build a new plant. And then some Chinese company eats their lunch, because that Chinese company has access to low-interest loans from a state-owned bank.
This happened in batteries. This happened in solar panels. What's to say it won't happen in ammonia?
If enough LNG export terminals get built, that advantage goes away. Our prices will go up, and their prices will go down.
The oil refining industry demonstrates how bad things get in a capital-intensive industry when both the inputs and outputs are fungible. They had flush times, too, in the mid-2000s. That was also temporary.
In a twist ending, they later decided to enter the US market starting by Boston (but it might be for geographical reasons).
The difference is more of a function of the nature of business being built.
Some fields require a lot of investment to come up with the product to sell. I worked at IBM research years back, They spend a lot of money on a lot of things that didn't pan out to get the profitable invention. Labs are expensive.
I understand the author's intent. Just stating...
Stanford does pure tech too. Look up SRI.
Rightly or wrongly, I interpreted the article as being about current startups, e.g. since the last bubble. It seems a lot more accurate in that context. Of course there are exceptions - "MIT" startups on the west coast and "Stanford" startups in the east - but the more important point is about the models themselves rather than the catchy monikers.
MIT approach, according to OP: we're smart, so let's do something badass and hard where being smart is maximally advantageous (because dumbasses wouldn't even understand the problem or the work).
Stanford approach, according to OP: we may be smart, but most people are self-important morons, including many of the gatekeepers we have to impress, so let's do easy shit so we can focus 100% on the marketing.
I'm not going to make better vs. worse comparisons, as both approaches have value, but I think that adept engineers (which most of us are, or want to be) fare better under the MIT approach. Under the approach ascribed to Stanford by the OP, engineers are just commodities of low importance-- it's the connection-peddlers and marketers who actually matter. (And that's what we actually see in the VC-funded incarnation of the Valley.)