Korva is a scroll of the internet’s best ideas.
Ship a first version you're embarrassed of
Launching ChatGPT early
Y Combinator pushed the startup ecosystem into this habit. Before launching ChatGPT, Sam Altman never asked Paul Graham whether to launch it, because he knew what the answer would be. The product was still early and still embarrassing, and the right thing was to get it in front of people.
Office hours
Graham's office hours had their own meme: he would shake a finger and say, you know what you should do? Sometimes what came next was great and sometimes it was terrible, but it came from a creative, open landscape of trying a lot of things.
Speed and success
Looking back over data points from YC founders over the years, how closely moving fast and iterating correlates with success is astonishing. It has been one of the most valuable pieces of tactical YC advice.
Why does early customer feedback beat polish?
Eric Ries described this approach in his 2011 book The Lean Startup, built around a loop of building, measuring, and learning. A minimum viable product is the simplest version that lets a team start learning from real users. Reid Hoffman, the LinkedIn founder, put the same idea in a line often repeated in Silicon Valley: if you are not embarrassed by the first version of your product, you launched too late. Real usage reveals which features matter, which no amount of internal planning can show.
How did Y Combinator shape startup habits?
Paul Graham, Jessica Livingston, Robert Morris, and Trevor Blackwell founded Y Combinator in 2005. It funded startups in batches and gave founders a short stretch of intense advising that ended in a Demo Day for investors. The model rewarded building something quickly and putting it in front of users. Many accelerators around the world later copied the format.
How did Dropbox test demand before building?
In 2007, Drew Houston, then in Y Combinator, posted a short demo video of Dropbox on the site Digg. The video showed how the file syncing would work before the product was finished. According to Houston, the waiting list jumped from 5,000 to 75,000 people overnight. That response told the team the idea had demand before they built the full product.
More from David Senra
