A lot of people here are saying $100B is too high of a goal to accomplish.
The best advice of the video is to have a strong vision of the future. How will the world be fundamentally different in 10 years compared to how it is today. Think about that change, and then build a company that will be well positioned with the new change. The stronger your vision, and the more accurate it is, the better your chance of success is.
In 10 years there will be a dozen 1 trillion dollar companies. The number of $100B companies is going to explode as well. More and more commerce is happening on the internet which scales much easier than brick and motor commerce. More and more goods purchased are entirely virtual, and have huge margins. More and more value is in software, which can be copied and used without practical limits. If your goal is to start a company that grows to $100B there has never been a better time to do that than right now.
If you set your goal at an unfathomable amount of net worth, then you lose focus on the problem you're trying to solve, and basically set your self up for failure.
A good example of this is [1] marshmallow challenge, which is a challenge where you build a tower out of spaghetti and balance a marshmallow on top. The biggest tower wins. The organiser discovered that if he offered a $10,000 prize to the highest structure... No one could build anything, the pressure of a large incentive would overwhelm people and the marshmallow would collapse the hastily designed structure.
The lesson is to solve a problem iteratively and not to focus on the reward. This is how large companies like FAMG do it too. Not try to make a multi-billion dollar business from day one.
> In 10 years there will be a dozen 1 trillion dollar companies. The number of $100B companies is going to explode as well.
Devil's advocate, is that actually a good thing?
Might be good for the consumer but not great for the laborer. Working for someone else is a bummer. Increasing consolidation means increasingly more of us will be in that position.
Yeah, the future is the Dystopian-Cyberpunk-Anime future of the mega-corporation-government-states which own and control everything.
People may say that is too-dystopian and cynical of a perspective, but all data-points over the last several decades are very solidly pointing in this direction....
There will certainly be lots of 10B, 50B and 100B companies - then the 1-5T companies will consume them just as the 5B companies grabbed up everything they could to build their moats over the last few years.
There will be cyber-wars for privacy in the next 20 years. There will be digital outcasts who seek to hack and bring down large systems.
There will be plenty of cyber-terrorism to come.
Personally, I thin kthe US is sitting on its laurels way too much and will be completely unprepared for a real cyberwar when it actually happens.
Anyone know of some good authors with a contemporary grasp of where we currently are with technology who is putting out some good cyberpunk fiction about the impending future?
The scary thing is that tech corps figured out how to stay in the game even when they lose.
Look at Microsoft, they did everything wrong and still won big.
They all hacked the innovator's dilemma: they're building bigger and bigger moats. And when they miss on a paradigm shift they jump on the next one. For example, maybe you miss mobile but win cloud etc.
Anyone know of some good authors with a contemporary grasp of where we currently are with technology who is putting out some good cyberpunk fiction about the impending future?
Neal Stephenson has done world building involving megacorporations that devolve into semi-feudal colonies. His style is goofier than Gibson's gritty realism but worth reading.
Logging in to recommend _not_ reading that book, it almost turned me off fiction. It's a mediocre YA sci-fi at best, IMO. Read the 1 & 2 star reviews for lolz, they are spot on: https://www.goodreads.com/book/show/40604388-walkaway
I think playing Devil's advocate would be to say my prediction is wrong. You are discussing the moral and ethical implications of my prediction, which is not a direction I am interesting in exploring now.
The future will have it's own set of advantages and disadvantages. It's important to leverage your own position in the present to give yourself a better position in the future.
If you're in a position to create a $100B company, don't you think you're in a position to help create the kind of future society that will benefit everyone rather than merely trying to look out for oneself in the context of a fait accompli future?
Keep in mind the basic premise of OP's post - that success requires an accurate vision of the future and positioning oneself for it. OP probably can't imagine a plausible future that will benefit everyone, if you can... obviously it would be best to position yourself for it.
I think decentralization will be a bigger theme in the next decade. Crypto nerds got a huge amount of surprise money in 2017 and the smart ones seem to be taking advantage of it. Even if only a small fraction of those are pursuing worthwhile projects I think the stage is set for a wave of "surprise" upsets 5-10 years down the line. The biggest will get bigger in that time as well, but they'll also be reaching the age where they start slowing down and turning consumers off with out-of-touch decisions.
Healthcare/Surgery could be practically free. You slide into a machine and there is an AI running on the cloud which compares your health data with your prior states and also with billions of other records from other human beings and makes the perfect diagnosis. Then it makes the precise cut into your flesh and gives you the perfect amount of anesthesia and inserts the stent and your heart surgery is done.
You don't need a radiologist, a general practitioner, a cardiologist etc.
The only thing that WONT happen in your prediction is the "free" part.
Hell, Intuitive Surgcal (makers of the Davinci) already have plenty of data to run procedures automated - but they wont do that yet...
The data from recorded successful surgical procedures executed by robots but driven by human surgeons is extremely valuable data.
That will be an interesting fronteir in med when you can run diffs across thousands of surgeries to smooth out the execution tailored to each particular subject.
And if anything goes wrong, we just assume it was unavoidable and not, say, a bug. So bugs don't get found and people die and nobody cares. Just spinning another possible scenario.
It could still be a net gain since vastly more people get the care they need.
But are they systematic and across potentially thousands of identically-programmed bots, all updated at the same time? The potential for widespread simultaneous errors is colossal.
As is the potential for instantaneous updating of optimised procedures, which would take years to decades with humans. In the end I'm certain they won't be infallible, but just as with autonomous vehicles they're likely to be many orders of magnitude more reliable than humans.
It's important to have an order of magnitude as a goal. In general the size of a company in dollars correlates with their impact on the world. The best talent is interested in making a big impact on the world.
2x or .5x won't make much of a difference, but if you are thinking 100x or 1,000x smaller then it will have an impact on your business in the early stages. If your goal is to make a $100M company you aren't likely thinking big enough to attract good talent and partners.
I don't buy it. Why would the best talent (according to your criterion) join some nobody who aims for $100B, when they can join ~$100B companies and start making impacts right now.
The video offers a great example of why a talented engineer would choose to join a startup (1990s Google) versus work in a large established company (Intel). It also turned out to be very beneficial for Paul Buchheit to join Google instead of stay at Intel.
There are a lot of talented engineers that prefer working for startups than large corporations. There are not enough talented engineers to staff all of the start ups. It's a competition for companies to attract the best talent.
Google's mission statement: "to organize the world’s information and make it universally accessible and useful"
Microsoft's old mission statement: (Paraphrase) To put a computer on every desk and in every home.
Amazon's mission statement: "...build a place where people can come to find and discover anything they might want to buy online"
These mission statements don't specifically state they want to build a $100B company, but it seems pretty obvious that a company that does the things in their mission statement would be worth $100B.
You’re actually correct. There’s a great story about Larry Page telling John Doerr that he thought it could be a $100 Billion company in their first meeting.
I think it would a huge distraction to think about billions while you still are struggling with millions. Focus on that goal first and then see where to go from there.
I was thinking about this in light of the recent MegaMillions lottery frenzy. Chance of buying a winning ticket is about one in a million times the number of successive draws in which no winner is declared in order for the jackpot to reach a sizeable pot. Ballpark estimate around 1:10M.
Whereas the chance you could actually build a company with billion dollar valuation is probably well under 1:1M. Just from being in the pool of startup school attendees or auditors. There are additionally many other factors which could substantially increase those odds.
Also, would just like to mention the recent podcast with Laura Deming of Longevity Fund was excellent. Incredibly informative and reasonably optimistic about the probability of a breakthrough in our lifetimes.
Of course the odds of building a company to 1B are better, because of the work involved. Buying a lottery ticket takes a couple of minutes of work and $2 at most.
Building a billion dollar company takes generally 10+ years of work (and even if you're YouTube, 1.5 years) and also probably a lot of your own money, at least in opportunity cost if not actual money.
Odds in a lottery are uniform and do not account for personal attributes. Assuming 1 ticket per person, anyone that buys a ticket has an equal chance of winning. A 98 year old grandmother has the same chance of winning as a 21 year old man.
Odds of starting a successful startup do not follow the same rules. There are a small number of people with the skillset/connections/etc to start a successful startup and thus have a very high chance of success, and a large number of people without those attributes, and thus a very low, if not zero percentage chance of success.
One small caveat is if someone is very wealthy, they can spend $0.9Billion buying lottery tickets which increases their odds vs someone that just buys 1
Also, building a company is not all or nothing. You could even shoot smaller and likely get to $1M in revenue for a $5M or so company. The odds are much better.
The odds of winning a lottery (lets say the Mega Millions or the Diversity Visa) are, to an extent, evenly distributed. The odds of succeeding in a social endeavour (lets say a business or a Green Card obtained via application) are not.
While on average the odds of the latter (the business) may be better, for a given person (which is always me and never an average person) the odds of the former (the lottery) may actually be much higher.
On a side note, while it may seem "un-meritocratic" to the HN crowd, having lotteries as gateways to social mobility and resource allocation is a decent idea.
My concern for thinking of lotteries as a societal benefit is that my (in no way perfect) observations is that the lowest in income, and education tend to <i>contribute</i> the most money towards the lottery which may have had a greater net benefit if each ticket buyer had spent their buy in price on themselves.
Also consider that poorer lottery winners are often ill equipped to deal with their winnings and frequently lose the entire sum of money within a few years; they may enjoy spending the money, but it doesn't lift a winners children and community out of a cycle of poverty as I would normally think of "social mobility".
Maybe we should randomly force people to get an MBAs instead.
The insight on Facebook's success and its implicit focus on the network vs Google's focus on product is interesting. Same with their conviction on the newsfeed.
Not sure if fully remote is the best goal. What if a company is fully remote, except for ~10 employees in a city who request an office at the local downtown co-working space? That kind of flexibility would be ideal, IMO.
I think it is important that there is no headquarter. Some people having desks at the same co-working space is fine. All the executives in the same building means everyone else is at a satellite office.
Around 32:00 : I was confused a little when Paul was talking about when 80% of facebook's user revolted against the decision on newsfeed and then facebook decided to stick with the decision which is ultimately successful. Seems to in direct contrast in what he believed when he was building Gmail.
Why would anyone worry about scaling to 100B? Once you are at that level you probably won't watch this video and everybody else is probably more worried about 1 or 10 million.
Yeah, 1-10 million is actionable with a lot of hard work but 100B requires the world organizing around you. Eg having good timing, being fortunate, lucky, what have you. But hey if it happens to you, congrats. It was bound to happen to someone!
I'm more interested in knowing if some companies are inherently designed to be $100B companies, or can any company get there by gradually expanding its operations.
Can the $1M ARR SaaS startup build products that can take it to $1B ARR? Or does that require a different DNA altogether?
Focus on what you can ship today. Find users that love your product and determine what they love about it. Talk to the users that do not love your product, and determine which features you can add to make them fall in love with your product. Go after high impact easy to implement features first.
100 users paying $8.25 a month will equal a $10k/yr MRR business.
I did try it with gridgenerator[0]. A painting application to create geometric patterns and shapes. I tried to have users, people loved it but didn't feel they were the target audience. Tried to get some b2b partnerships afterwards which failed for various reasons. I have a few debts to pay now (because I was using my pension money for it) and a large amount of bugs to fix when my willpower comes back up again.
Today no one uses it. I had to shut down the export server, and got about 40 followers on instagram with some of the works I managed to do with it[1].
Now i work at a web company trying to slave my way to pay the debt and in another 10 years try it again. Does it entitle me to say "been there done that" ?
It was an easy decision, no one was doing any real use with the app. Money was running out and so I tried to contact a couple of portuguese tile brands to sell a more specialized version of it. Nothing good came out of it. We don't have enough market for specialized/customized tiles/ceramics to move the industry players into it. Also the construction world is crippled by a different logic and mechanics. Free market ? yeah right... :/
Still in spite of all of these I do believe that there is a lot more this tool can give. My plans for it in the near future (early 2019) are:
- Present it in a slightly different way (remove landing page and instantly start playing a dumbed down version of it that introduces the current features as a 'initialization/introductory process')
- Improve the mobile experience (by default shape creation must be done by choosing which areas to paint instead of the current approach of connecting points, this is a small fix that makes using it in mobile a much better experience)
- Allow anonymous export and sharing of works. Remove the login until it is absolutely needed.
- Create thematic portfolios with it and share them with current graphical artists that work with grids
- Restart instagram marketing
If you have any other ideas that don't fit the traditional "startup book" approach I will gladly do them or try to incorporate them in the reboot.
I wouldn't take this as the path to success. You are in a very different industry than I have ever worked in. I do have some ideas for you to consider.
- Add a share button. Freeware gets a watermark, paid gets the clean image. Integrate with twitter, facebook, and instagram to make posts.
- Rebrand the application to something more artsy sounding. I now have three tabs open that say "Grid Generator" and it sounds like an engineer tool, while the application runs like an art tool.
- Hows the tablet experience? I would focus on desktop and tablets more than mobile. I think this application would lend itself to larger screens than cell phones have.
- Run a subreddit for people to share their work.
- Server costs for the export engine can be lowered by using AWS lambda to provide on demand computing. Implementing this can range from easy to difficult depending on which language the engine is written in.
If you want to reach out to me more about this subject you can find an email address in my profile. Good luck to you!
I applied to YC for the first time after attending Startup School.
They said they would release the outcomes of who got in or not at 10pm PST. Instead it started to trickle in some 3 hours later. During that period the Slack channels had entrepreneurs panicking about whether they got in or not. Some even talking about suicide. And apparently this happens every, single time.
I have lost any and all respect for YC and the partners who claim they represent the interests of founders.
Founders should focus on making great products not trying to reach some arbitrary number.
This is not YC's fault. I am not sure why you would blame YC for this.
As you said, founders should focus on making great companies (not just great products), so they should not focus on the outcome release time.
If you get into YC that's great. If not, you are still going to build your company anyways. There's not going to be a change in direction of your company either way.
YC is an accelerator. It accelerates your progress up a path, but ultimately you are responsible for setting the path and running up it.
Please don't consider suicide over a YC application. There's going to be a lot more obstacles ahead, that will be more stressful than applying to YC.
This is ridiculous and sad. Getting rejected from YC is not the end of the world or even the end of your startup. And being 3 hours late for releasing acceptance decision does not mean YC unilaterally doesn’t care about founders. Get real
You can’t blame YC. Startups have become another high visibility “sport” that attracts people who are both incapable and unstable. None of those people would have known or applied back in 2005.
Startup School material has been incredibly helpful for us at Biterica.com. Highly recommend checking out all their videos, especially on finding product market fit and growth. We just launched if you want to check us out: https://news.ycombinator.com/edit?id=18293997
The best advice of the video is to have a strong vision of the future. How will the world be fundamentally different in 10 years compared to how it is today. Think about that change, and then build a company that will be well positioned with the new change. The stronger your vision, and the more accurate it is, the better your chance of success is.
In 10 years there will be a dozen 1 trillion dollar companies. The number of $100B companies is going to explode as well. More and more commerce is happening on the internet which scales much easier than brick and motor commerce. More and more goods purchased are entirely virtual, and have huge margins. More and more value is in software, which can be copied and used without practical limits. If your goal is to start a company that grows to $100B there has never been a better time to do that than right now.