If you are an entrepreneur or are planning to build a startup, do yourself a favor and watch the “How to Start a Startup” lessons that Y Combinator did on Stanford a while ago. You can watch all lessons you YouTube.
Below you will find my notes for lesson 1. I am writing them mostly to make sure I will absorb the concepts.
4 Areas You Need to Cover to Succeed:
Obviously there is a big luck factor involved too, but that is outside of our control, so nothing to do about it.
- Yes, execution is 10x more important than the idea itself, but that does not mean that you should build startups on top of bad ideas. Great execution of a bad idea will take you no where, and while pivoting is fine, it should not be your goal.
- Sam Altman says that pivots usually don’t become very big companies. I am not sure about that (will research for a future post).
- If it works, you will be working for 10+ years on your startups, so you need to evaluate if your idea really has potential to grow and is defensible. Defensible means that your business will eventually be very hard to replicate (i.e., Facebook today).
- Don’t start a startup if you are not passionate about the idea, and believe that it has real potential and will be defensible. In that case it is better to simply wait for another one.
- The best companies are always mission oriented. Have a larger mission.
- Despite the appearance of being overnight success stories (i.e. Airbnb, Uber), usually it takes 10+ years to build a great startup. Again, if you are not passionate, you will quit too early in the game.
- Think about the market. It is better to be in a small market that is going to grow a lot and fast, than to be in a big market that is growing slow, if at all.
- Why now? Why this idea/startup could not have been done 2 years ago, and why 2 years from now it will be too late? You need a good answer for that.
- Good ideas are simple and easy to explain. People might not agree they are good ideas, but they understand what the idea is about easily.
- Great idea -> Great product -> Great company
- Without a great product, nothing else matters
- In the early days of a startup, only two activities matter: building the product and talking to customers. All else is wasted effort and time, and if you are doing much of it, be worried.
- Eat, sleep, exercise, build the product and talk to customers. These are the only activities you should be doing as a startup founder in the early days.
- You need a product that users love. Not something they like. Something they LOVE!
- It is better to build something that few users love, than something that many users like somewhat. Why? Cause from the former position it is easier to go to a product that many users love.
- Building something that many users love would be ideal, but there are few opportunities like that available, cause big companies all over them.
- My Own Note: building a product that people love is akin to finding product market fit. One way to measure that is to evaluate how many of your users would HATE if they had to stop using your product. A rule of thumb is that is 40% of users are in that position, you are have product market fit (in other words, your customers love your product).
- Another way of measuring how much people love your product, or how well is your product market fit, is by the amount of new customers you will get by recommendations and word of mouth. If people love your product, they will talk about it to their friends and colleagues.
- Start with something simple. It is much easier to build a great product if you do something that is simple. Most great companies started with something very simple (Facebook, Google, WhatsApp, etc).
- If you get a great product that users love, you can get many other factors wrong and still succeed. If you get this wrong, the other factors wont matter.
- You don’t need that many users to get started, so recruit them by hand if you need. AdWords and Facebook Ads might be overkill at this point. The manual/personal recruitment of customers might be a good idea because it will allow you to get feedback easier (i.e., by calling or walking to them), and this is the key part.
Create an engine like this:
work on product -> show users -> collect feedback -> repeat
I will post the summary for building a team and execution on a future post. After all, building a great product is all you got care about for now!