27 ms·

Why building a startup is probably your most sensible career path

270 points by dn2k 10y ago

dluan 10y ago
Seems like the author has a motivation to push this kind of storyline, that startups are sexy for scientists. Yet another early-stage fund with tons of investments and none that are notable successes (yet).
dboreham 10y ago
Glug glug glug. The sound of Koolaid being drunk..
klipe 10y ago
"Your PHD has..." closes tab
- doingmyting 10y ago
  Haha, I think I closed the tab after reading the same. It's like creating a startup is like a loaf of bread you pick up at the store on the way home.
- jack9 10y ago
  Exactly what I did.
pavel_lishin 10y ago
... if you've just gotten your Ph.D., know how to get some specific grants, have no interest in remaining in academia, and consider it against only one alternative: becoming a professor.
- pc86 10y ago
  Well if you have no interest in remaining and academia and you only compare it to becoming a professor, anything would be better, right?
  - stevenwiles 10y ago
    I'm having trouble understanding your logical association between having no interest in doing something and being willing to do literally anything else. Would you mind explaining what you mean by that?
dberg 10y ago
"That clearly sounds dismal, but it doesn’t matter – as long as you can cover your costs while you try then your upside is nearly unlimited, whilst your downside is virtually non-existent" Um, yes except for the part about paying for food and shelter ??
- nibs 10y ago
  Cover your costs includes mortgage and food in this context.
- sharemywin 10y ago
  guess the author skipped his economics class on opportunity cost.
sremani 10y ago
I really do not understand the cynical posts here, but given the path for Ph.D.s trying to get tenure is getting more difficult. Deep Science ventures are providing a path ahead. I understand the writer is vested in his venture. I was watching the ultra-endurance documentary the one in Tennesse, he was telling its usually people with Masters or PhDs who excel at this because they are used to planning and preparation. A Ph.D means you have higher than average IQ and you have grit, honestly, without will-power no body completes Ph.D. which have great correlation of success. You can learn marketing (which is very important), improve you EQ with seminars but Depth of Ph.D. and grit to the finish line is not easy to get. Its not automatic, but I agree with the essence of the article. edit: I am not a Ph.D.
- sidlls 10y ago
  That "grit" is entirely within an academic context and does not necessarily translate outside of that context. It's also debatable whether a PhD absolutely implies an above average IQ (or even whether that is a meaningful trait). As for correlation with success, I wonder if you could elaborate? What measures "success"? What studies are you referring to? What confounding issues (if any) are accounted for?
  - throwaway729 10y ago
    > That "grit" is entirely within an academic context and does not necessarily translate outside of that context Do you have any particular reason for stating this? Having been in both environments, I actually think that academic grit translates outside of academia more fluidly than the other way around. Conjectured reason: most academic projects have a horizon of 5-20 years to market and require large amounts of very mundane and uninspiring work. There are infinite distractions within academia. A successful academic project isn't going to make anyone rich, and even if it does, that person probably won't be you. Also, the pay is terrible. Startup and even corporate life are very different in almost all of those dimensions. Projects succeed or fail fast. There are few distractions, especially for individual contributors. There's still routine work, but usually routine in the sure-to-succeed way rather than the sure-to-fail-and-be-repeated-10x-times way. A successful project can easily make you a valuable person -- either on the job market or in terms of actual net worth. And even if everything hits the fan, you're receiving a reasonable salary and building your career. > It's also debatable whether a PhD absolutely implies an above average IQ That is certainly true.
Nalta 10y ago
As a Ph.D. student, I found this to be an interesting outlook, albeit a bleak look at the current job market. My research is in machine learning, and I'm hoping my job search will be easier :P
- gautamdivgi 10y ago
  I thought the outlook for a PhD w/ specialization in ML would be bright, no? At least in industry. The academic world is something I wouldn't even consider.
baristaGeek 10y ago
"Now, the most successful investors are increasingly coming to the realisation that searching for the low hanging fruit in web-tech is nearing the point of diminishing returns." This is true in the sense that the amount of VC poured into things such as social networks or SoLoMo startups is decreasing and the amount of VC poured into things such as machine-learning based startups is increasing. Even though the amount poured into the first is still larger than the amount poured into the second. However, this article shouldn't discourage people with just an undergrad (or even without an undergrad) to start startups. A group of 22 year olds might know about Deep Learning and might just apply it as a medium to a solution in a different narrow domain than a group of 40 year olds will.
jondubois 10y ago
I didn't go as far as PhD, but when I came out of uni, I knew nothing about life or work. I really wouldn't recommend anyone to try to launch their own startup right out of uni (especially not alone). When I think back to it; if I had succeeded with my first startup right out of uni, I would have had a seriously distorted view of life. My idea of 'hard work' has changed a lot since then. Staying up till 3am a few times a semester to finish an assignment is not hard work. Writing a 100 page thesis is not hard work either.
- amelius 10y ago
  > Writing a 100 page thesis is not hard work either Except if you stubbornly decide to fully understand LaTeX in the process.
  - pyromine 10y ago
    Dear lord, I've started writing all of my mathematics homework in LaTeX, it's not even remotely reasonable. Though interestingly, having to be more deliberate in my operations do make me have a much more rigorous understanding of what I'm doing.
    - arcanus 10y ago
      > it's not even remotely reasonable. What is more scary: I exclusively write LaTeX equations in emails, raw to my colleagues. We have all been working with it for so long it is a native language. Math does funny stuff to your brain.
      - tjl 10y ago
        Most of my colleagues did this too. I know I've done it with my supervisor. It's just easier when you're trying to write out something complex and everyone who you're writing to knows how to read it without difficulty.
      - 50CNT 10y ago
        This is your brain \this{is your brain on Math} As a less poetic example, playing Dwarf Fortress for prolonged periods of time reportedly has similar effects. The ASCII fades, and mental images take its place. It's freaky when decoding slips into the hardware layer.
ianai 10y ago
Usually a bubble is just about to burst when the last hold out becomes convinced and buys into the hype. The enthusiasm stems from a belief that it will always continue and apply to everyone. Then destruction and chaos.
- smallnamespace 10y ago
  > last hold out becomes convinced and buys into the hype Interestingly, this can be justified by economic theory, because it's the marginal price of the next transaction that everyone sees, so you always need someone who's willing to buy in for a higher price. Once everyone has already bought in and there's no fresh money to prop up the bubble, everyone can immediately see that, and things fall apart just like you said.
  - ianai 10y ago
    The telling stuff always happens on the margin ;)
- sp527 10y ago
  If we end up with a bubble in this case, it will be due to the nature of hard science contrasted against the prevailing Theranosian assumption that R&D can be brute forced by entrepreneurs.
Animats 10y ago
This is addressed to new PhDs. Look what they're comparing it with: "According to the Start-up Genome Report 92% of high-tech start-ups fail. ... If you can get as far as investment this number falls to around 40%, and your chance of making it big within that sits at around 10% (Pitchbook). This adds up to a 0.48% success rate overall. Interestingly, this is the same proportion of PhD students that reach professorship."
convolvatron 10y ago
if you succeed, great. if you raise, fail, and decide that it isn't the life for you - you aren't left with a lot of options.
bluebeard 10y ago
Aka science PhD's need jobs because there's not enough research funding to go around.
DenisM 10y ago
Every PhD student should read this book [1] about academic job perspectives. It's a bit of a cold shower, but it also tells you how to deal with it, if that's what you really want after all. [1] https://www.amazon.com/Professor-Essential-Guide-Turning-Ph-D/dp/0553419420 https://www.amazon.com/Professor-Essential-Guide-Turning-Ph-...
kayhi 10y ago
I completed a PhD in chemistry and went straight into the start up world in 2008. I focused on building tools that would have been helpful in my research lab[1]. My feeling was I can work a lot of hours for little pay and questionable benefits as a post doc or start my own thing. I use 'start my own thing' since I hadn't ever heard of doing a start-up being in the Midwest. [1] https://labspend.com/ https://labspend.com/
- ginger_beer_m 10y ago
  Do you get enough paying customers for your startup? I imagine labs tend to be hesitant before they approach startups for solutions to their problems..
lumberjack 10y ago
Well sometimes you still need that Ph.D, though. I recently enquired about a possible career at a small (<50 people) consulting firm. They do mathematical modelling and consulting and have scored a few high profile contracts. The work is interesting, easier than the typical Ph.D stuff and yet still challenging and varies a lot. BUT! They pretty much only hire Ph.Ds. Why? It's mostly a reputation thing. When they send you to help out a client they don't want to send somebody who "only" obtained a masters. The firm was started by two Ph.Ds out of school and could essentially be considered a startup. Barely 10 years old I believe.
- dejv 10y ago
  There is distinction between consulting company, new business and startups.
matrix 10y ago
While the author does quote the very low odds of success, I think he does readers a disservice by minimizing that risk. Very few people have the right disposition, the right partners, the right idea, and the right skill set to launch a startup. Being a freshly minted PhD doesn't change that. Failing at a startup sucks, and sucks really hard. But of course, its not in a VC's interest to tell you the pain that comes with a startup. Better advice: if you want to do a startups, get some experience, find some good partners, test out ideas in your own time, and when you have enough runway and something that looks solid, only THEN make the leap.
- __mbm__ 10y ago
  > Failing at a startup sucks This. I have a PhD and a failed startup. It felt real bad, and I didn't get that far with it (9 months before calling it quits). Creating a successful startup requires significantly more energy and stress than most science jobs, even for pre-tenure professors. It requires different skills than science. Few introverts will succeed at founding a company, for example. Moreover, you have to produce actionable results quickly, and you have very different success metrics. I'm one degree of separation from a PhD startup founder that literally killed himself as his startup imploded, when the company was 5(?) years old. Severe depression, loss of all semblance of a life... these are standard in the world of startup founders. The risk that you destroy your life is real.
  - pb 10y ago
    Many of the most successful founders are introverts. Mark Zuckerberg, Larry Page, Bill Gates, none of them are very extroverted, but they've started some of the most valuable companies in the world. Of course if you are so introverted that you are unable to make eye contact or talk to people it will be a problem, but most of us are not that extreme.
    - gohrt 10y ago
      Why do you think they are not extroverted? Bill Gates was a famously in-your-face person when dealing with employees, partners, and competitors.
gohrt 10y ago
Putting blogs and "news" on nature.com is continually degrading the quality of the "Nature" brand. It's poor form to mix this editorial and un-reviewed journalistic content under the same banner as peer-reviewed scientific articles of the Nature journal.
carsongross 10y ago
Dr Mark Hammond is one of the founders of Deep Science Ventures, a fully funded six-month programme designed to turn scientists in to founders. Hmmm. I'm going to go out on a limb and suggest ulterior motives are in play here.
Obi_Juan_Kenobi 10y ago
0.5% getting 'professor' positions is just nonsense. Just click the link so see some useful discussion, but it's ridiculous at its face: the average professor is not graduating 200 PhD students. The figure is probably closer to 20 for most fields, which gives a more reasonable 5%. Then you have to consider that many programs offer tenured research positions, but don't take on doctoral students. Finally, consider that, upon entering a doctoral program, many students don't intend on academia at all. That number drops by the time they complete their dissertations, too. Ultimately, I'd give a rough figure of 1 in 4 qualified and motivated PhDs getting a suitable tenure-track position in a reasonable timeframe (~4-5 years). Maybe one in five or six will get tenure, in the end. Which is still terrible when you think about it, but the odds aren't nearly as ridiculous as some would claim.
- patall 10y ago
  Might not be in physics or mathematics but in Life Science it certainly almost is. At the institute I am working right now has about 20 PhD students per full professor (every 3 years anew, so a Professor has 10 times that number in 30 years). Its all a question of associates and other types of group leaders and obviously delegation.
- awiedmer 10y ago
  That's assuming that tenured positions are not on the decline (they are), and that the number of PhD graduates has not been increasing (it is), and that the funding available to maintain a lab has not gone down (it has).
biounit 10y ago
Interesting perspective. Certainly get a good feel for your potential before striking out on a startup. And have as much savings as possible. I did the PhD and while doing my Postdoc I started a company. I did many years as Research Professor before quitting and focusing on my company. Comparing the two situations I'd argue the business option is far better for a number of reasons. The most important being freedom. To survive in academics now is an awful game of proposals and nonstop stress. Success rates are very low and there is intense competition. And all for very little (no) payoff. And it is a sales job. At the end, before I finally pulled the plug I often likened academics to a situation where you're selling something that people may like, but they don't want to pay for it. If you do it right with a company, people are begging to pay you. And the reality is the business route is much more challenging, just in a different way. Everything is on the line. I do think promoting a "startup" in the conventional sense where you have VC funding and explosive growth is disingenuous. There are many opportunities where a PhD scientist can carve out a niche where you have a company with 1-10 million in revenue. It affords an excellent situation that, if sustainable, is orders of magnitude better than the academic lifestyle that is so often idolized.
- monkmartinez 10y ago
  Does a startup have? Proposals and Stress = Check low success rates and intense competition = Check Need for sales = Check I can't help but think you essentially have the same situation in academia as one has in a startup, only "it's different." The difference to me is the illusion of control. One may think that they are in control of the startup vs. the bureaucracy of government/academia/established business. I posit that most things that are required by business are, in fact, not under your control. The illusion of control is there and one may feel that actions are leading to certain outcomes, but the truth is they have no direct control of the outcome. "Luck" plays the same role in the bureaucratic or private enterprise affair... some have it and others don't.
  - jedberg 10y ago
    Chance of extremely high returns = Check (not so much in academia)
hodgesrm 10y ago
I'm a huge start-up fan but this article makes doing a start-up sound more fun than they actually are on the ground. Here are a few things the article leaves out. * Years of grunt work. You just did that on your Ph.D. Don't kid yourself, start-ups are just as hard and can take longer before they are successful. As they say in show biz, it takes years to become an overnight success. * Business smarts are non-optional. If you are in a leadership position you need good business acumen. (I.e. like Andy Grove, not like Bill Shockley.) * You need to be an all-rounder who can roll with the punches and pitch in to do whatever needs to be done. Chances are only some of that will be science and some fraction will be downright unpleasant. My personal favorite: the head honcho in a small company is the top of the product support escalation chain. By the time users get on the phone with you they are often downright steamed. Finally, if you don't fit the mold there's a reasonable chance you'll either just flame out nothing to show for it or get fired. That's in addition to all the normal stuff that goes wrong when you have a good idea, good funding, good co-founders, etc. Doing a startup is not for everyone.
bluetwo 10y ago
Screw the statistics. If you feel compelled to start up a company, I say do it. Put yourself 100% in. You might be 100% successful, or, you might be 100% a failure. But do not think the statistics make you one or the other. Only advice I would give is to get yourself on decent financial footing before making the leap. It helps you from making impulsive/bad decisions later.
shoyer 10y ago
> Later stage investment is increasingly focused on science > > Even five years ago it was incredibly hard to get funding for science-based ventures... This seems wrong to me. This may be true for consumer facing products or tech investors like YC, but it misses that biotech investors have been investing $1B+/quarter into science based startups for over a decade now [1]. I suspect that as far as absolute numbers go traditional biotech investing still dominates in this space. [1] http://www.forbes.com/sites/brucebooth/2015/07/23/the-venture-funding-boom-in-biotech-a-few-things-its-not http://www.forbes.com/sites/brucebooth/2015/07/23/the-ventur...
astazangasta 10y ago
I deplore the model of the world where acquiring a PhD is a choice made by an individual, and what you should do is cash in on your credentials. When we ignore the social component of our lives, big surprise, it atrophies and dies. I have a PhD because many people (a society) invested in me to get one. My PhD is not something I should use for my own enrichment by building a "career" and a "startup" to satisfy my job requirements, there should be a substantial portion of my effort that is socially directed, that gives back the enormous amount that was given to me to get me here. This is why I will remain stuck in academia forever, desperately seeking that shrinking universe where someone can attempt to produce pure public goods.
- contingencies 10y ago
  Admirable! This sort of socially conscious personal life decision making rarely survives very long after contact with capitalist realities... a quote from Peter Thiel (of whose apparent social perspectives and path-to-riches I am not a fan, ) summarizes the late capitalist perspective on this: The eccentric university professor is a species that is going extinct fast. May I ask what your field is?
sporkologist 10y ago
Why work 40 hours a week for someone else and sleep at night, when you can work 16 hours/day 7 days a week for 10 years for yourself, and age a lot in the process?
- bbcbasic 10y ago
  Err. Why do you need to work 112 hours a week for 10 years?
  - sporkologist 10y ago
    To get to the other side.
thirstysusrando 10y ago
haha what
- sctb 10y ago
  Please comment civilly and substantively here or not at all. https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html
leroy_masochist 10y ago
This strikes me as an oddly process-based approach to a problem (how to start a successful company) that is never solved by following a set process. I think "startups require almost the exact same skills as science" is true only in the broadest sense....like, doing work of scientific value is hard and requires intelligence and discipline, and this is also true of startups. It is also true that there are many people who are exceptionally good scientists who are not good entrepreneurs (and wouldn't be if they tried), and vice versa.
ci5er 10y ago
This is interesting. Some of the worst-run startups and companies I know were started and run by one or more academics. You see this quite a bit in pharma-tech and bio-tech. Yes - they have the smarts. Yes - they know how to learn (but are sometimes arrogant enough that they won't). Yes - they know how to work hard. But there's something about the academic culture that makes for ... I don't know. Something. I haven't nailed it down, and I wouldn't go so far as to claim "advanced degrees considered harmful", but startups are hard in a different way than academia prepares one for. That said: you learn a lot in the first one. And get to learn more about the wrong lessons from the first on the second. Why not give it a whirl?
dxbydt 10y ago
I agree with the article not because I've drunk the kool-aid but because I have seen the math work out on numerous occasions. Say you average $x per year for 3 years, then your startup fails, so you join bigcorp at $y. So at the end of 4 years you have 3x+y. Your conservative classmate joins a bigcorp and stays there 4 years averaging $z per year. So he now has 4z. It's a simple exercise to enumerate tuples (x,y,z) where 3x + y > 4z. Literally in every case I have seen up close and personal, the LHS exceeded the RHS. And I'm not going to argue with math.
- nxc18 10y ago
  That equation makes it a little too easy to have a rosy picture of things. A failed startup can be very expensive. You might break even. You might not. Just from the financial perspective on it, when you start including negative and near-zero values on the left side of that equation, conservative job at bigcorp starts looking good. Then think about the rest of your life. Surely there is some value in getting enough sleep, having free time, work-life balance (because your work isn't 100% of your life), and a basic level of security. Startup life sucks; there are far fewer people who have what it takes (including luck) than there are people who think they do.
LargeCompanies 10y ago
Can't agree more! If you are not a skilled developer or designer yet have a major passion about seeing your app ideas built go for it and learn those skills along the way. Your startup journeys could succeed but the odds are against you. Though, whatever its a ton of fun and the stories that come from it make you a more interesting candidate when going up for your first Developer job. I had two startups fail yet I was able to raise incubator money for both, invited and had a demo meeting with Google, invited and filmed for a reality tv show about makers(I think I maybe a good marketer hence these opportunities yet I suck at people skills) and a lot of other random out of nowhere stuff. Overall, great times with many lows that are now great life stories even if I'm not a millionaire I now earn a good living coding.
kriro 10y ago
I'm not sure the headline is warranted as natural science career paths that involve R&D departments of established companies are a very good choice (as is academia in these fields). Taking slow and steady wins is actually a good choice in these fields (imo) as is access to "veterans". But I could be wrong, I only see natural scientists' lives from the outside. But the overall point is interesting. I feel there's a slight bias against academics from "the startup scene" or maybe I'm just imagining it being a mediocre scientist thinking about starting by own business every now and then (so not the super duper elite MIT guy). I've always felt that some of the stuff you learn by practicing science is fairly valuable even in a web based startup. Depending on the field you're more or less forced to get a better understanding of statistics and rigorous testing (with all the pitfalls involved) and the general hypothesis/test-it cycle is really helpful for startup life as well. For me the sad trade of is that my programming skills (mostly the secondary skills like architecture, being up to date on databases, refining the vcs workflow, testing etc.) erode slowly. I also have a lot less time for side projects these days which kind of sucks. [I haven't taken the traditional academia career path and actually worked at a software company before so I feel like I have a decent grasp on both fields]
andkon 10y ago
TLDR: start a startup if you are in such a terrible job market that you have, by default, a 95% chance of failure.
- j7ake 10y ago
  But the 1/20 that do "succeed" get a salary of ~$150000 over ~30 years, which is worth $2-4 million today dollars depending on how you set your interest rates. Are 1 in 20 founders today exiting with $2-4 million in their pockets?
  - andkon 10y ago
    No.
known 10y ago
"The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." --George Bernard Shaw
lsc 10y ago
eh, I am in infrastructure, not science, but looking back, I have a really hard time coming up with something that I might have done (I mean, something that the person I was in my 20s would likely have done.) that could have caused me more financial damage than starting a company did. On top of that, nobody tells you, but when you go to get a real job after starting a company? If you didn't have big-name VC, "Oh, that was your company?" the interviewer asks... but what the interviewer clearly means is actually "Oh, so you were unemployed" I mean, I'm not saying that you shouldn't start a company, I'm just saying that you need to be mindful of the risks. Yes, you feel free when you are young, but any money you put into retirement then is extremely impactful. The same is true of anything you put into covering your long-term housing short. Even if I could talk to myself in my early 20s, I probably could not talk him out of starting a company, but... i would be way better off now if he could have been talked into being more conservative... into maxing out all available tax-deferred retirement options, into limiting the amount of money and the amount of my self that I was willing to sink into the thing. Into buying a fucking condo, at least. Something.

270 points by dn2k 10y ago

dluan 10y ago

Seems like the author has a motivation to push this kind of storyline, that startups are sexy for scientists. Yet another early-stage fund with tons of investments and none that are notable successes (yet).

dboreham 10y ago

Glug glug glug. The sound of Koolaid being drunk..

klipe 10y ago

"Your PHD has..." closes tab

doingmyting 10y ago
Haha, I think I closed the tab after reading the same. It's like creating a startup is like a loaf of bread you pick up at the store on the way home.
jack9 10y ago
Exactly what I did.

pavel_lishin 10y ago

... if you've just gotten your Ph.D., know how to get some specific grants, have no interest in remaining in academia, and consider it against only one alternative: becoming a professor.

pc86 10y ago
Well if you have no interest in remaining and academia and you only compare it to becoming a professor, anything would be better, right?
- stevenwiles 10y ago
  I'm having trouble understanding your logical association between having no interest in doing something and being willing to do literally anything else. Would you mind explaining what you mean by that?

dberg 10y ago

"That clearly sounds dismal, but it doesn’t matter – as long as you can cover your costs while you try then your upside is nearly unlimited, whilst your downside is virtually non-existent" Um, yes except for the part about paying for food and shelter ??

nibs 10y ago
Cover your costs includes mortgage and food in this context.
sharemywin 10y ago
guess the author skipped his economics class on opportunity cost.

sremani 10y ago

I really do not understand the cynical posts here, but given the path for Ph.D.s trying to get tenure is getting more difficult. Deep Science ventures are providing a path ahead. I understand the writer is vested in his venture. I was watching the ultra-endurance documentary the one in Tennesse, he was telling its usually people with Masters or PhDs who excel at this because they are used to planning and preparation. A Ph.D means you have higher than average IQ and you have grit, honestly, without will-power no body completes Ph.D. which have great correlation of success. You can learn marketing (which is very important), improve you EQ with seminars but Depth of Ph.D. and grit to the finish line is not easy to get. Its not automatic, but I agree with the essence of the article. edit: I am not a Ph.D.

sidlls 10y ago
That "grit" is entirely within an academic context and does not necessarily translate outside of that context. It's also debatable whether a PhD absolutely implies an above average IQ (or even whether that is a meaningful trait). As for correlation with success, I wonder if you could elaborate? What measures "success"? What studies are you referring to? What confounding issues (if any) are accounted for?
- throwaway729 10y ago
  > That "grit" is entirely within an academic context and does not necessarily translate outside of that context Do you have any particular reason for stating this? Having been in both environments, I actually think that academic grit translates outside of academia more fluidly than the other way around. Conjectured reason: most academic projects have a horizon of 5-20 years to market and require large amounts of very mundane and uninspiring work. There are infinite distractions within academia. A successful academic project isn't going to make anyone rich, and even if it does, that person probably won't be you. Also, the pay is terrible. Startup and even corporate life are very different in almost all of those dimensions. Projects succeed or fail fast. There are few distractions, especially for individual contributors. There's still routine work, but usually routine in the sure-to-succeed way rather than the sure-to-fail-and-be-repeated-10x-times way. A successful project can easily make you a valuable person -- either on the job market or in terms of actual net worth. And even if everything hits the fan, you're receiving a reasonable salary and building your career. > It's also debatable whether a PhD absolutely implies an above average IQ That is certainly true.

Nalta 10y ago

As a Ph.D. student, I found this to be an interesting outlook, albeit a bleak look at the current job market. My research is in machine learning, and I'm hoping my job search will be easier :P

gautamdivgi 10y ago
I thought the outlook for a PhD w/ specialization in ML would be bright, no? At least in industry. The academic world is something I wouldn't even consider.

baristaGeek 10y ago

"Now, the most successful investors are increasingly coming to the realisation that searching for the low hanging fruit in web-tech is nearing the point of diminishing returns." This is true in the sense that the amount of VC poured into things such as social networks or SoLoMo startups is decreasing and the amount of VC poured into things such as machine-learning based startups is increasing. Even though the amount poured into the first is still larger than the amount poured into the second. However, this article shouldn't discourage people with just an undergrad (or even without an undergrad) to start startups. A group of 22 year olds might know about Deep Learning and might just apply it as a medium to a solution in a different narrow domain than a group of 40 year olds will.

jondubois 10y ago

I didn't go as far as PhD, but when I came out of uni, I knew nothing about life or work. I really wouldn't recommend anyone to try to launch their own startup right out of uni (especially not alone). When I think back to it; if I had succeeded with my first startup right out of uni, I would have had a seriously distorted view of life. My idea of 'hard work' has changed a lot since then. Staying up till 3am a few times a semester to finish an assignment is not hard work. Writing a 100 page thesis is not hard work either.

amelius 10y ago
> Writing a 100 page thesis is not hard work either Except if you stubbornly decide to fully understand LaTeX in the process.
- pyromine 10y ago
  Dear lord, I've started writing all of my mathematics homework in LaTeX, it's not even remotely reasonable. Though interestingly, having to be more deliberate in my operations do make me have a much more rigorous understanding of what I'm doing.
  - arcanus 10y ago
    > it's not even remotely reasonable. What is more scary: I exclusively write LaTeX equations in emails, raw to my colleagues. We have all been working with it for so long it is a native language. Math does funny stuff to your brain.
    - tjl 10y ago
      Most of my colleagues did this too. I know I've done it with my supervisor. It's just easier when you're trying to write out something complex and everyone who you're writing to knows how to read it without difficulty.
    - 50CNT 10y ago
      This is your brain \this{is your brain on Math} As a less poetic example, playing Dwarf Fortress for prolonged periods of time reportedly has similar effects. The ASCII fades, and mental images take its place. It's freaky when decoding slips into the hardware layer.

ianai 10y ago

Usually a bubble is just about to burst when the last hold out becomes convinced and buys into the hype. The enthusiasm stems from a belief that it will always continue and apply to everyone. Then destruction and chaos.

smallnamespace 10y ago
> last hold out becomes convinced and buys into the hype Interestingly, this can be justified by economic theory, because it's the marginal price of the next transaction that everyone sees, so you always need someone who's willing to buy in for a higher price. Once everyone has already bought in and there's no fresh money to prop up the bubble, everyone can immediately see that, and things fall apart just like you said.
- ianai 10y ago
  The telling stuff always happens on the margin ;)
sp527 10y ago
If we end up with a bubble in this case, it will be due to the nature of hard science contrasted against the prevailing Theranosian assumption that R&D can be brute forced by entrepreneurs.

Animats 10y ago

This is addressed to new PhDs. Look what they're comparing it with: "According to the Start-up Genome Report 92% of high-tech start-ups fail. ... If you can get as far as investment this number falls to around 40%, and your chance of making it big within that sits at around 10% (Pitchbook). This adds up to a 0.48% success rate overall. Interestingly, this is the same proportion of PhD students that reach professorship."

convolvatron 10y ago

if you succeed, great. if you raise, fail, and decide that it isn't the life for you - you aren't left with a lot of options.

bluebeard 10y ago

Aka science PhD's need jobs because there's not enough research funding to go around.

DenisM 10y ago

Every PhD student should read this book [1] about academic job perspectives. It's a bit of a cold shower, but it also tells you how to deal with it, if that's what you really want after all. [1] https://www.amazon.com/Professor-Essential-Guide-Turning-Ph-D/dp/0553419420 https://www.amazon.com/Professor-Essential-Guide-Turning-Ph-...

kayhi 10y ago

I completed a PhD in chemistry and went straight into the start up world in 2008. I focused on building tools that would have been helpful in my research lab[1]. My feeling was I can work a lot of hours for little pay and questionable benefits as a post doc or start my own thing. I use 'start my own thing' since I hadn't ever heard of doing a start-up being in the Midwest. [1] https://labspend.com/ https://labspend.com/

ginger_beer_m 10y ago
Do you get enough paying customers for your startup? I imagine labs tend to be hesitant before they approach startups for solutions to their problems..

lumberjack 10y ago

Well sometimes you still need that Ph.D, though. I recently enquired about a possible career at a small (<50 people) consulting firm. They do mathematical modelling and consulting and have scored a few high profile contracts. The work is interesting, easier than the typical Ph.D stuff and yet still challenging and varies a lot. BUT! They pretty much only hire Ph.Ds. Why? It's mostly a reputation thing. When they send you to help out a client they don't want to send somebody who "only" obtained a masters. The firm was started by two Ph.Ds out of school and could essentially be considered a startup. Barely 10 years old I believe.

dejv 10y ago
There is distinction between consulting company, new business and startups.

matrix 10y ago

While the author does quote the very low odds of success, I think he does readers a disservice by minimizing that risk. Very few people have the right disposition, the right partners, the right idea, and the right skill set to launch a startup. Being a freshly minted PhD doesn't change that. Failing at a startup sucks, and sucks really hard. But of course, its not in a VC's interest to tell you the pain that comes with a startup. Better advice: if you want to do a startups, get some experience, find some good partners, test out ideas in your own time, and when you have enough runway and something that looks solid, only THEN make the leap.

__mbm__ 10y ago
> Failing at a startup sucks This. I have a PhD and a failed startup. It felt real bad, and I didn't get that far with it (9 months before calling it quits). Creating a successful startup requires significantly more energy and stress than most science jobs, even for pre-tenure professors. It requires different skills than science. Few introverts will succeed at founding a company, for example. Moreover, you have to produce actionable results quickly, and you have very different success metrics. I'm one degree of separation from a PhD startup founder that literally killed himself as his startup imploded, when the company was 5(?) years old. Severe depression, loss of all semblance of a life... these are standard in the world of startup founders. The risk that you destroy your life is real.
- pb 10y ago
  Many of the most successful founders are introverts. Mark Zuckerberg, Larry Page, Bill Gates, none of them are very extroverted, but they've started some of the most valuable companies in the world. Of course if you are so introverted that you are unable to make eye contact or talk to people it will be a problem, but most of us are not that extreme.
  - gohrt 10y ago
    Why do you think they are not extroverted? Bill Gates was a famously in-your-face person when dealing with employees, partners, and competitors.

gohrt 10y ago

Putting blogs and "news" on nature.com is continually degrading the quality of the "Nature" brand. It's poor form to mix this editorial and un-reviewed journalistic content under the same banner as peer-reviewed scientific articles of the Nature journal.

carsongross 10y ago

Dr Mark Hammond is one of the founders of Deep Science Ventures, a fully funded six-month programme designed to turn scientists in to founders. Hmmm. I'm going to go out on a limb and suggest ulterior motives are in play here.

Obi_Juan_Kenobi 10y ago

0.5% getting 'professor' positions is just nonsense. Just click the link so see some useful discussion, but it's ridiculous at its face: the average professor is not graduating 200 PhD students. The figure is probably closer to 20 for most fields, which gives a more reasonable 5%. Then you have to consider that many programs offer tenured research positions, but don't take on doctoral students. Finally, consider that, upon entering a doctoral program, many students don't intend on academia at all. That number drops by the time they complete their dissertations, too. Ultimately, I'd give a rough figure of 1 in 4 qualified and motivated PhDs getting a suitable tenure-track position in a reasonable timeframe (~4-5 years). Maybe one in five or six will get tenure, in the end. Which is still terrible when you think about it, but the odds aren't nearly as ridiculous as some would claim.

patall 10y ago
Might not be in physics or mathematics but in Life Science it certainly almost is. At the institute I am working right now has about 20 PhD students per full professor (every 3 years anew, so a Professor has 10 times that number in 30 years). Its all a question of associates and other types of group leaders and obviously delegation.
awiedmer 10y ago
That's assuming that tenured positions are not on the decline (they are), and that the number of PhD graduates has not been increasing (it is), and that the funding available to maintain a lab has not gone down (it has).

biounit 10y ago

Interesting perspective. Certainly get a good feel for your potential before striking out on a startup. And have as much savings as possible. I did the PhD and while doing my Postdoc I started a company. I did many years as Research Professor before quitting and focusing on my company. Comparing the two situations I'd argue the business option is far better for a number of reasons. The most important being freedom. To survive in academics now is an awful game of proposals and nonstop stress. Success rates are very low and there is intense competition. And all for very little (no) payoff. And it is a sales job. At the end, before I finally pulled the plug I often likened academics to a situation where you're selling something that people may like, but they don't want to pay for it. If you do it right with a company, people are begging to pay you. And the reality is the business route is much more challenging, just in a different way. Everything is on the line. I do think promoting a "startup" in the conventional sense where you have VC funding and explosive growth is disingenuous. There are many opportunities where a PhD scientist can carve out a niche where you have a company with 1-10 million in revenue. It affords an excellent situation that, if sustainable, is orders of magnitude better than the academic lifestyle that is so often idolized.

monkmartinez 10y ago
Does a startup have? Proposals and Stress = Check low success rates and intense competition = Check Need for sales = Check I can't help but think you essentially have the same situation in academia as one has in a startup, only "it's different." The difference to me is the illusion of control. One may think that they are in control of the startup vs. the bureaucracy of government/academia/established business. I posit that most things that are required by business are, in fact, not under your control. The illusion of control is there and one may feel that actions are leading to certain outcomes, but the truth is they have no direct control of the outcome. "Luck" plays the same role in the bureaucratic or private enterprise affair... some have it and others don't.
- jedberg 10y ago
  Chance of extremely high returns = Check (not so much in academia)

hodgesrm 10y ago

I'm a huge start-up fan but this article makes doing a start-up sound more fun than they actually are on the ground. Here are a few things the article leaves out. * Years of grunt work. You just did that on your Ph.D. Don't kid yourself, start-ups are just as hard and can take longer before they are successful. As they say in show biz, it takes years to become an overnight success. * Business smarts are non-optional. If you are in a leadership position you need good business acumen. (I.e. like Andy Grove, not like Bill Shockley.) * You need to be an all-rounder who can roll with the punches and pitch in to do whatever needs to be done. Chances are only some of that will be science and some fraction will be downright unpleasant. My personal favorite: the head honcho in a small company is the top of the product support escalation chain. By the time users get on the phone with you they are often downright steamed. Finally, if you don't fit the mold there's a reasonable chance you'll either just flame out nothing to show for it or get fired. That's in addition to all the normal stuff that goes wrong when you have a good idea, good funding, good co-founders, etc. Doing a startup is not for everyone.

bluetwo 10y ago

Screw the statistics. If you feel compelled to start up a company, I say do it. Put yourself 100% in. You might be 100% successful, or, you might be 100% a failure. But do not think the statistics make you one or the other. Only advice I would give is to get yourself on decent financial footing before making the leap. It helps you from making impulsive/bad decisions later.

shoyer 10y ago

> Later stage investment is increasingly focused on science > > Even five years ago it was incredibly hard to get funding for science-based ventures... This seems wrong to me. This may be true for consumer facing products or tech investors like YC, but it misses that biotech investors have been investing $1B+/quarter into science based startups for over a decade now [1]. I suspect that as far as absolute numbers go traditional biotech investing still dominates in this space. [1] http://www.forbes.com/sites/brucebooth/2015/07/23/the-venture-funding-boom-in-biotech-a-few-things-its-not http://www.forbes.com/sites/brucebooth/2015/07/23/the-ventur...

astazangasta 10y ago

I deplore the model of the world where acquiring a PhD is a choice made by an individual, and what you should do is cash in on your credentials. When we ignore the social component of our lives, big surprise, it atrophies and dies. I have a PhD because many people (a society) invested in me to get one. My PhD is not something I should use for my own enrichment by building a "career" and a "startup" to satisfy my job requirements, there should be a substantial portion of my effort that is socially directed, that gives back the enormous amount that was given to me to get me here. This is why I will remain stuck in academia forever, desperately seeking that shrinking universe where someone can attempt to produce pure public goods.

contingencies 10y ago
Admirable! This sort of socially conscious personal life decision making rarely survives very long after contact with capitalist realities... a quote from Peter Thiel (of whose apparent social perspectives and path-to-riches I am not a fan, ) summarizes the late capitalist perspective on this: The eccentric university professor is a species that is going extinct fast. May I ask what your field is?

sporkologist 10y ago

Why work 40 hours a week for someone else and sleep at night, when you can work 16 hours/day 7 days a week for 10 years for yourself, and age a lot in the process?

bbcbasic 10y ago
Err. Why do you need to work 112 hours a week for 10 years?
- sporkologist 10y ago
  To get to the other side.

thirstysusrando 10y ago

haha what

sctb 10y ago
Please comment civilly and substantively here or not at all. https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html

leroy_masochist 10y ago

This strikes me as an oddly process-based approach to a problem (how to start a successful company) that is never solved by following a set process. I think "startups require almost the exact same skills as science" is true only in the broadest sense....like, doing work of scientific value is hard and requires intelligence and discipline, and this is also true of startups. It is also true that there are many people who are exceptionally good scientists who are not good entrepreneurs (and wouldn't be if they tried), and vice versa.

ci5er 10y ago

This is interesting. Some of the worst-run startups and companies I know were started and run by one or more academics. You see this quite a bit in pharma-tech and bio-tech. Yes - they have the smarts. Yes - they know how to learn (but are sometimes arrogant enough that they won't). Yes - they know how to work hard. But there's something about the academic culture that makes for ... I don't know. Something. I haven't nailed it down, and I wouldn't go so far as to claim "advanced degrees considered harmful", but startups are hard in a different way than academia prepares one for. That said: you learn a lot in the first one. And get to learn more about the wrong lessons from the first on the second. Why not give it a whirl?

dxbydt 10y ago

I agree with the article not because I've drunk the kool-aid but because I have seen the math work out on numerous occasions. Say you average $x per year for 3 years, then your startup fails, so you join bigcorp at $y. So at the end of 4 years you have 3x+y. Your conservative classmate joins a bigcorp and stays there 4 years averaging $z per year. So he now has 4z. It's a simple exercise to enumerate tuples (x,y,z) where 3x + y > 4z. Literally in every case I have seen up close and personal, the LHS exceeded the RHS. And I'm not going to argue with math.

nxc18 10y ago
That equation makes it a little too easy to have a rosy picture of things. A failed startup can be very expensive. You might break even. You might not. Just from the financial perspective on it, when you start including negative and near-zero values on the left side of that equation, conservative job at bigcorp starts looking good. Then think about the rest of your life. Surely there is some value in getting enough sleep, having free time, work-life balance (because your work isn't 100% of your life), and a basic level of security. Startup life sucks; there are far fewer people who have what it takes (including luck) than there are people who think they do.

LargeCompanies 10y ago

Can't agree more! If you are not a skilled developer or designer yet have a major passion about seeing your app ideas built go for it and learn those skills along the way. Your startup journeys could succeed but the odds are against you. Though, whatever its a ton of fun and the stories that come from it make you a more interesting candidate when going up for your first Developer job. I had two startups fail yet I was able to raise incubator money for both, invited and had a demo meeting with Google, invited and filmed for a reality tv show about makers(I think I maybe a good marketer hence these opportunities yet I suck at people skills) and a lot of other random out of nowhere stuff. Overall, great times with many lows that are now great life stories even if I'm not a millionaire I now earn a good living coding.

kriro 10y ago

I'm not sure the headline is warranted as natural science career paths that involve R&D departments of established companies are a very good choice (as is academia in these fields). Taking slow and steady wins is actually a good choice in these fields (imo) as is access to "veterans". But I could be wrong, I only see natural scientists' lives from the outside. But the overall point is interesting. I feel there's a slight bias against academics from "the startup scene" or maybe I'm just imagining it being a mediocre scientist thinking about starting by own business every now and then (so not the super duper elite MIT guy). I've always felt that some of the stuff you learn by practicing science is fairly valuable even in a web based startup. Depending on the field you're more or less forced to get a better understanding of statistics and rigorous testing (with all the pitfalls involved) and the general hypothesis/test-it cycle is really helpful for startup life as well. For me the sad trade of is that my programming skills (mostly the secondary skills like architecture, being up to date on databases, refining the vcs workflow, testing etc.) erode slowly. I also have a lot less time for side projects these days which kind of sucks. [I haven't taken the traditional academia career path and actually worked at a software company before so I feel like I have a decent grasp on both fields]

andkon 10y ago

TLDR: start a startup if you are in such a terrible job market that you have, by default, a 95% chance of failure.

j7ake 10y ago
But the 1/20 that do "succeed" get a salary of ~$150000 over ~30 years, which is worth $2-4 million today dollars depending on how you set your interest rates. Are 1 in 20 founders today exiting with $2-4 million in their pockets?
- andkon 10y ago
  No.

known 10y ago

"The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." --George Bernard Shaw

lsc 10y ago

eh, I am in infrastructure, not science, but looking back, I have a really hard time coming up with something that I might have done (I mean, something that the person I was in my 20s would likely have done.) that could have caused me more financial damage than starting a company did. On top of that, nobody tells you, but when you go to get a real job after starting a company? If you didn't have big-name VC, "Oh, that was your company?" the interviewer asks... but what the interviewer clearly means is actually "Oh, so you were unemployed" I mean, I'm not saying that you shouldn't start a company, I'm just saying that you need to be mindful of the risks. Yes, you feel free when you are young, but any money you put into retirement then is extremely impactful. The same is true of anything you put into covering your long-term housing short. Even if I could talk to myself in my early 20s, I probably could not talk him out of starting a company, but... i would be way better off now if he could have been talked into being more conservative... into maxing out all available tax-deferred retirement options, into limiting the amount of money and the amount of my self that I was willing to sink into the thing. Into buying a fucking condo, at least. Something.