(0:00) Patrick co-founded Stripe in 2010 with his brother John. It took 2 years from the first line of code to the official launch, although he doesn’t recommend that. They however started getting users every month even before launching, since 3 months after starting work on the product. By the time they launched they had a hundred users. Today Stripe has 1,300 employees, 9 offices worldwide, and is worth $20B.
(1:37) The two brothers used to go around Palo Alto to companies’ offices to install their API into the web apps, which was a great way to do user research and get UX feedback. (3:00) They did that themselves to avoid having users continuously postpone the integration of their API.
(7:10) If a company has 3 co-founders and no clear CEO, and there’s no efficient mechanism to reach consensus of decisions, that’s a recipe for failure. Patrick and John are kind of responsible for different areas, and they trust each other in giving a direction for decisions in the respective areas. If there’s disagreement on important decisions, they follow the one who is more passionate about his opinion.
(11:00) A startup has 2 stories, one before market fit and one after, and for each the CEO has different roles.
(14:00) At the beginning they spent time to understand in detail what their users were doing, for example providing customer support on a public chat, or inspecting each and every API call that they were getting. And if an API call would fail, they’d get a high-priority email to look into it. Try to understand as much as possible about every user until there are too many.
(17:45) They had a contact form with pre-made sentences with placeholders, like “The thing I love the most about Stripe is ___” and “The thing I hate about Stripe is ___” to incentivize useful feedback.
(22:00) If you’re working on a your startup and what you’re doing makes you unhappy or it doesn’t seem promising, quit it. There must be some sort of happiness or fulfillment in what you do. Being determined at all costs is not the right answer.
(24:55) When asked about what he wasn’t good at and when it was time to hire someone for those things, Patrick says he wasn’t good at doing the partnerships with banks, and he was told to hire a guy for that. At the time Stripe was all made up by engineers, and they think exactly know what a non-engineer person would do. But they did hire him, and it was an immediate change in their trajectory.
(30:00) If he could go back, he the would be more disciplined and self-aware. A good thing that repeat founders do, post market fit, is in fact building the organisation ahead of where things are today. Once you know you have a good product with market fit, build the organisation that can serve the entire market. This is at least for B2B…in B2C customers don’t know what they want and it’s hard to estimate the market size. Businesses are mostly rational and it’s easier to predict what they want.
(34:13) Before product market fit don’t worry about culture, team etc., just think about getting there and care about speed of iteration according to your customer feedback. So in terms of hiring, get people that can help you build the product faster, but not too many otherwise the team will be too big to manager and you’ll get slower. Something between 2 and 10 is a good size, but it can be anything that minimizes the time between observing a need or a problem, and the execution or the fix for it.
(37:55) Every new person takes time to hire, onboard, coordinate etc. So think hard if there’s a benefit from the new person.
(39:00) The key aspects he looks for in people is being intellectually honest, caring a lot and loving getting things done. It’s hard to fake being smart or intellectually honest or nice. At Stripe they also want people that are pleasant and warm and that make you happy as a result of their presence.
(43:15) Try to balance between speed and agility, which requires some hierarchical efficient structure, and having people that are independent and with an ownership mentality that can identify problems and inject new ideas.
(47:53) Even now that it’s big, Stripe wants to be fast and agile, and they try to have a “yes culture”. Every year they send around a “hackpad document”, and everyone in the company can add ideas to it. They should be “bad ideas”, because if they were good that wouldn’t be risky. People don’t want to look stupid for coming up with bad ideas and often censor themselves, but Stripe tries to incentivize them to suggest whatever idea they have. A number of things Stripe has were born in this way, for example Stellar (a decentralized payment network to which they provided seed funding). This is similar to what Google did for Gmail or Maps for example.
(51:25) Stripe was mostly centralized in SF until quite recently, but now they opened their 4th hub. These are no satellite offices (operations, sales etc.), but offices that actually develop new products. This is because talent is geographically dispersed, the Bay Area is becoming more and more expensive, but also because Stripe wants to be global and work as well in Asia an in other markets as it does in North America.
