How to Start a Startup – Week 1 (part 2)

Reading 3: Stupid Apps and Changing the World – Sam Altman

  • Many very important things end up looking as if they don’t matter. so it’s a very bad mistake to dismiss everything that looks trivial
  • Why?
    • The value of a network grows as a function of the square of the number of notes. This means the value/importance of the service grows hyper-exponentially.
  • The key takeaway here is this simple rule to know if you are on track:
    • “If some users really love what you’re building, engage with the service or product as an important part of their daily lives and interesting new behaviors keep emerging as you grow, keep working on it
  • Two time-tested strategies to change the world with technology
    • Build something that some people love but most will think is a toy
    • be hyper-ambitious and start an electric car company or a rocket company
  • Also some advice for the people who try:
    • Don’t claim you’re changing the world until you’ve changed it
    • Ignore the haters and work on whatever you find interesting

Reading 4: Do Things That Don’t Scale – Paul Graham

Paul Graham was the author of Hackers and Painters (see my prior post on this), and several other essays you can find on his website. His basic argument here is that you do not need to skip ahead to automating everything and mass marketing in the beginning stages of a startup. You need to focus on two things: Aggressive user acquisition and getting people to love the experience. He argues that startups take off because the founders make them take off.


  • Go out and recruit your users manually – you need to pursue very aggressive early user acquisition techniques
    • One example is what he calls the “Collision installation” where instead of asking people if they will try your beta and sending them a link,  you actually say “give me your laptop” and set them up on the spot
    • The main reason people do not do this is because of shyness and laziness – it is easier to sit home and write code than it is to put it out there.
      • For a startup to succeed, at least one founder will have to spend a lot of time on sales and marketing
  • Don’t underestimate the power of compound growth. Measure progress by a weekly growth rate metric
    • If you do the math – if you have 100 users and grow at 10% per week you will have 14,000 users after 1 year, and 2 million after 2 years! Exponential growth
    • Start by recruiting users manually and gradually switch to less manual methods
      • Airbnb started by going door to door in New York recruiting and helping the existing ones improve their listings


  • Almost all starts are fragile initially
    • Know-it-alls and reporters will dismiss your startup, but they always get things wrong. The big danger is that you’ll dismiss your startup yourself. Even Bill Gates did this when he went back to Harvard after starting Microsoft.
    • One question you should be asking yourself is “How big could this company get if the founders did the right things”
      • Microsoft started out as a couple of guys writing Basic interpreters for a market of a couple thousand hobbyists.
    • How do you find users? You solve your own problems, then you just have to find your peers


  • Take extraordinary measures to both acquire your users AND make them happy
    • You should always be racking your brain for new ways to delight them (example of having a handwritten note from the CEO of Wufoo)
  • Why do we need to teach startups this? 3 reasons:
    • Mainly founded by engineers which are used to building things without much handholding. Customer service is not part of the training of engineers.
    • They are worried it won’t scale – but larval startups have nothing to lose. If they end up having too many customers to make super happy that would be a good problem to have. You will also find that delighting customers scales better than you may expect. It will scale for two reasons:
      • You can usually figure out how to make anything scale more than you would have predicted
      • Delighting your customers by that time would be engrained in your culture
        • I have never seen a startup ‘lured down a blind alley’ by trying too hard to make their users happy”
      • They’ve never experienced such attention themselves
        • Their standards for customer service have mostly been set by the large companies – Tim Cook can’t send you a handwritten note when you buy something, but you can.
        • This is the advantage of being small: you can provide the service no big company can
    • What you need to think about is how far you can go to delight your customers


  • Steve Jobs conveyed how extreme your attention to users should be: “insanely great”
    • The most successful startups YC has funded have been insanely great experiences for the customer
    • In the first couple months of your startup, it is not the product that must be insanely great, but the experience of being your user.
      • You can and SHOULD give users an insanely great experience with an early, incomplete, buggy product if you can make up the difference with attentiveness!
    • Over-engaging with early users is not just a permissible technique for getting the ball rolling, it is often a necessary part of the feedback loop to make your product great
    • Outside of domains with big penalties for mistakes, it is often better not to aim for perfection initially. It works best to get something in front of users as soon as it has some quantum of utility then see what they do with it.
      • “perfectionism is usually an excuse for procrastination”
      • The feedback you get from talking with your initial users will be the best you ever get. When you are big you have to rely on focus groups vs. going over to their homes and watching them use your stuff.


  • Focus on a deliberately narrow market (like Thiel’s Zero to One). Start in a subset of the market, then gain a critical mass of users as quickly as possible.
    • You need to ask yourself if the subset you are in is a market in which you can gain a critical mass of users quickly
  • Among companies, the best early adopters are usually other startups. They are the most open to new things by nature and since they have just started they haven’t made all of their choices yet


  • “pulling a Maraki” – the first several hundred Pebble watches were assembled by hand by the founders since they couldn’t afford mass production. They learned a ton about the process and how important it was to order good screws. Being close to the process allows you to see things you wouldn’t see otherwise.


  • They often suggest B2B startups take over-engagement to the extreme. Use the initial users as the form for your mold- keep tweaking to fit their needs perfectly.
  • If you’ve got a toehold in making something people want, that’s as much as any startup needs initially. Your goal is to find that one user that REALLY needs something and can act on that need.
  • Use the software on their behalf. See where the bugs are, what needs improvements, what is tedious, etc.


  • When you are small you can afford to do things by hand what you want to automate later. You don’t have to do everything from the start.
    • If you make the manual components look like software to your users, it doesn’t matter. One example is that Stripe offered “instant” merchant accounts to its first users because the founders actually signed them upper traditional merchant accounts behind the scenes
  • You can gradually automate the bottlenecks as you go


  • The tactic that usually doesn’t work is ‘the big launch’
    • Startups are not projectiles, you don’t need a huge initial takeoff.
    • The most successful startups are focus on making a few users happy then expanding that base.
  • Founders typically think what they are building is so great that everyone who hears about it will immediately sign up. – DON’T fall into this trap
    • Getting users will always be a gradual process
      • Great things are usually also novel
      • Users have other things to think about
  • Partnerships usually don’t work either for the same reason
  • This is the key:
    • Do something extraordinary + put in an extraordinary effort 


  • There are two ‘vectors’ you should view as pairs when going into growing your startup:
    • What you are going to build
    • Plus the unscalable things you’re going to do initially to get the company going
  • Unscalable things change the company for the better and allow things to happen that would be impossible if you were only focused on scalable things
  • You’ll learn things that you wouldn’t have otherwise
  • If you’re aggressive when your small, you will probably be aggressive as you grow



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s