September 04, 2025
“Scaling a startup is not replication, it’s reinvention,” says Hayagreeva “Huggy” Rao, a professor of organizational behavior at Stanford Graduate School of Business. He recalls a conversation with one of Uber’s first executives that proves the point. “How long have you worked at Uber?” Rao asked. “Four years,” the CTO replied. “But it feels as though I worked for 16 different companies.”
This illustrates one of the key takeaways from Rao’s course From Startup to Scaleup, where he teaches students how to navigate the complex journey of scaling a startup. Here, he shares five key takeaways from the course.
Full Transcript
Note: Transcripts are generated by machine and lightly edited by humans. They may contain errors.
Huggy Rao: Hello, I’m Huggy Rao. I’m a professor at the GSB at Stanford University, and I’ve been so for more than two decades. I have the pleasure today of sharing some ideas about a course I teach called From Startup to Scaleup. I do that with two co-conspirators, both of whom have been founders and DCs themselves, John Lilly now a director at Figma, and Sujay Jaswa, who runs his own firm called Winder and Company, where they invest in a variety of startups. I love having these two co-conspirators because it helps me marry clean models from research with dirty hands from the world of practice. And as all of you know, the startup to scaleup journey is one where practice can lead theory and not just the other way around. I’m delighted to share with you five takeaways from this course.
Takeaway number one is that scaling a startup is not replication, that is, doing more of the same. It’s reinvention. A quick example, we wrote a case study about Uber in its early days around its CTO, and when we asked the CTO, “How long have you worked at Uber,” he said, “Four years, but it feels as though I worked for 16 different companies.” Uber changed itself 16 times, once every quarter, during his tenure there.
The second implication is that when we scale a company, we are often leading when we don’t know enough. There’s a big difference between leading when we know and leading when we don’t. When we lead when we know, we’re like hippopotamuses. We have a big mouth, very small ears, very small eye. When we lead when we don’t know enough, we have to be like elephants, with big ears, with a big trunk, and with big eyes. So scaling listening is super key to being a founder orchestrating a startup.
The third takeaway is that the job of a leader is to make the right things easy to do in a startup. What are the right things? Curiosity and generosity. How do we make them easier? By removing band friction, that is, obstacles that exhaust and confuse people.
The fourth takeaway is, the job of a leader is to make the wrong things hard to do. And what are the wrong things? Overconfidence and myopia. And how do we make them hard? By putting in good friction that is obstacles that slow down decision-making and foster reflection. A quick example, one of the venture capital companies here in the Bay Area, they have an investment management committee with five people. If all five people agree that a startup is a great place to invest, they’ll never put money in that. They only put money when the vote is four to one or three to two, and the reason is, there is controversy, due diligence, and discussion.
And takeaway number five is, when you scale a startup, the leader is responsible or the leader has to balance two things. One is poetry and the other is plumbing. Poetry has to do with purpose, aspirations, the story of the startup. Plumbing is about process, efficiency, and the like. Our job always is to multiply poetry with plumbing. It’s very tempting for all of us to think there are two types of people, plumbers and poets, but what we hope to convey to you in this course is that all of us have two selves, that of a poet and that of a plumber, and the key thing is to learn how to activate both of those selves when you’re in the process of scaling a startup.
We’re all excited about the course, and I hope I’ve conveyed some of our enthusiasm for the course and hope you share it.
I must tell you, it’s great fun being with my co-conspirators in every class, and the reason it’s a lot of fun is, we often disagree. As you can imagine, if all three of us agree, we are all superfluous, right? Only one of us is necessary. But because all three of us disagree, the students also know that they can disagree, too.
For media inquiries, visit the Newsroom.