"The very best startup ideas tend to have three things in common: they're something the founders themselves want, that they themselves can build, and that few others realize are worth doing." Paul Graham on how to get startup ideas:
@ycombinator Decentralized compute is one of those ideas, obvious only in hindsight.
@ycombinator don't forget that the best startups should also have a seamless customer support experience ;)) from beginning to end. way too many chatgpt wrappers nowadays with a broken product + terrible support. ideas are important but so is execution.
@ycombinator Building something you actually want keeps you going through the inevitable rough patches. External motivation can be hard to find when things get hard.
@ycombinator Is it about fulfilling a personal need, or anticipating an overlooked collective one? Perhaps both are essential for true impact.
@ycombinator Step 1: Have a problem Step 2: Solve it Step 3: Convince the world it’s a problem too Steve Jobs ideology ☝️
@ycombinator Most technical founders have the same problems, so markets get saturated by them following this advice. Plenty of people in other fields aren’t learning to code, the smarter move is to partner with them and solve their problems, imo
@ycombinator Check, check, and check. Wish I could update my entire YC app because I pivoted.
Yeah or just make cool stuff. Calculate the market size by identifying the ideal customer, multiply it by what you'll be charging. Figure out the risks of doing it, which is usually what drives people away, but you're passionate so those are challenges to you. Then, just post everywhere. Make cool stuff. Do cool stuff. Don't do it for anybody else but yourself. Use investment as an accelerator, not a crutch.
This is exactly why I want to help others get better at decision-making. I spent a decade learning how to decide under uncertainty — now I’m building tools to help people do it faster and more systematically. If you struggle with decisions (or making them fast), try this free template I built to cut the mental clutter of irrelevant choices or problems: notion.so/ClarityCue-Pro…
@ycombinator As a founder learning to understand what drives me sometimes, at @ycombinator learning just what you are saying here has inspired me to keep pressing on. Also, for those founders who have not taken the @startupschool you are missing out on some incredible insight and tools!🧰
@ycombinator The first part of this has always felt so obvious
@ycombinator Honestly it feels extremely motivating to solve your own problems with the product you are building. That is 100% why I am building EdenLM edenlm.com It would have solved so many pains at my previous business.
@ycombinator A lot of people don’t have the patience to build the tools they want.
@ycombinator This is literally the best advice anyone can give and anyone can receive in the startup space.
@ycombinator Yes yes and yes! And we’re doing it! 🙌
@ycombinator To get into YC faster, should I ship one banger or 100 mid projects?
@ycombinator relating to this so much lately fr
@ycombinator Every founder should ask themselves if what they are building checks these boxes…
@ycombinator Great criteria for building something awesome.
@ycombinator The best startup ideas do reject consensus early on. Funny how many icons built out of irritation, not inspiration. Building what matters to you first feels less risky than most would like to admit.
@ycombinator My best products came from scratching my own itch. Solving your own problems means you already have at least one customer.
@ycombinator There is also a metric of why they are the best people to do it
@ycombinator Hey, are you having trouble finding ideas that fit this mold? If so, reply to this message. I’ll get you one back.
@ycombinator Pivoting can be very different. I'm sure pg will agreee
@ycombinator "Build something you want." Meanwhile, half of AI startups are building chatbots for industries they've never worked in, solving "problems" they read about in TechCrunch 💀
@ycombinator But doesn’t this contradict your advice that b2c is really really hard?
@ycombinator Looking for problems that other people have is also not a bad idea 💡