TripAdvisor co-founder Steve Kaufer explains how he built a speed-obsessed culture at the company. He kept a sign on his door reading "speed wins" and hired only A-players while enforcing a "no assholes" policy to maintain fast decision-making and parallel work.
I actually want to jump in here and ask a question for a familiar face or name, actually, Ravi, that added a question to the Google Doc in advance, Steve, which is around, you know, TripAdvisor's value of speed wins. And he noted that, of course, you know, most companies tend to slow down as they grow. So, you know, how did you create and maintain a speed-oriented culture at TripAdvisor? What advantages do you think, you know, TripAdvisor experienced as a result of being focused on speed?
So I have to Break to share an amusing anecdote. I thought I would last about a year after TripAdvisor was acquired because like I can't work in a big company. Like I'm like, and I was, yeah, we were certainly planning to grow and like the administrative, not my favorite part of the product engineering guy. And a friend of mine said, Well, if your complaint is that the company's gonna slow down and you're gonna be unhappy, didn't you say you were the CEO? I'm like, yeah. Like, so that's kind of your choice as to whether the company slows down or not, right? Like, oh, but there's so many people. Like, take it as your challenge, keep the company going fast. And so I like, hey, that's a pretty fair perspective. So I put a sign on the door to my office that said speed wins, because anyone coming in to talk to me about anything would have to see that sign. If I was in a remote culture, it would just be on my Zoom background. It was speed wins. Like, I want to drive home with every single person in the company that we value going fast. And the flip side is I'm not gonna slap you. If you miss a deadline, because I used to be a— like most of the deadlines missed are engineering, right? I used to be an engineering manager for many, many years. When I would miss a deadline, I'd get— I'd be penalized. It would be a bad thing. So I got used to padding my deadlines so I wouldn't miss them. Then I was a hero. Everyone optimizes for being a hero. So therefore, like, oh, can't you pull that schedule in? No, because that would risk being a hero. Whereas if it's a speed wins, and like postscript to that experience at the prior company was like, and rarely did anything ship earlier than expected by more than a week or two. So we had optimized at that company accidentally for security, safety. Things took a long time to get done, but they were done on time. There was a big penalty for missing the date. But TripAdvisor, I'm like, no, there's not gonna be any penalty for missing the date by a week. We didn't have external, like we had features we needed to get out. Now, work hard, work fast, make decisions that you're proud of in terms of the code. But don't tell me it's going to take 3 months to do something that— sniff, sniff— I think you can do in a month. Like, I will— like, it'll, it'll take 2 weeks to do the spec. No, it'll actually take 2 days to do the spec if you put the right person on it and go, oh, well, what about usability? Great. They should hop on the train and quickly do some usability while you're speccing the rest of it out, while you've started your architectural implementation. What are you going to parallelize? Don't tell me engineering has to halt while usability goes and does some stuff. That's bullshit. So again, pressing every moment on how fast can you go, trying not to err too much on the Shipping stuff that has a bunch of bugs in it, which TripAdvisor didn't, didn't do in general.
I—
other people have a valid point of like, hey, the go fast at the expense of deciding what the right thing to do is has its risks. I agree, it does. Try to decide what the right things to do are, but when you decide what the right things to do are, just go blazingly fast. Optimize for speed, reward speed, thank people for getting it done quickly. If it's late, hey, we ran into this problem. Sorry, we didn't see it coming. It's gonna take an extra week. My answer, hey, no problem. Thanks for keeping me up to date. Like, new people, they were shocked that I wasn't upset. Like, I asked for a super tight— I asked for not the 80% confidence. I asked for the 50% confidence date. You're not sure if you can make it. That means sometimes it's gonna be later and I'm going to reward you for giving an optimistic date, not penalizing you if I think you are truly working hard at it, which Cryptowisers, they always were.
I'm glad he asked that question. I didn't know about the, you know, the sign with the speed win. So I had to make one and have it up back here.
So the CEO can only deliver a certain number of messages when your company is over 100 people. Like, What is, what are they known for? What do they care the most about? Are they always out networking, never walking around the building, proverbial building? Are they caring about how you are, how to make you more effective? Do they wanna learn? Are they stuck in their ways? Are they stubborn? Are they open? Is your culture one of creative disagreement? Or yes, sir. Yes, ma'am. Yes, sir. Like you get choices and I didn't pay a lot of attention initially to spreading the culture, maybe as much as I should have, but a couple of things stuck. One was if it's worth doing, it's worth measuring. The other is speed wins. And the third is— Have impact. Like anywhere you're in the company, like is what you're doing having an impact on the business? Ask yourself those questions. And then to you guys at this stage, I stayed involved as the final round interviewing every single person that came into the company until I couldn't do it anymore. And that was about 250 people. Like, do it. Kind of like the, the talent level in the company always seems to decrease. And the phrase I learned a long, long time ago is, you know, A-caliber individuals hire A-caliber and B-caliber. B-caliber individuals hire B-caliber and C-caliber. And so you just end up with this Nice people. TripAdvisor had a no assholes policy and we lived up to that. We had nice people. Some were just not very good at doing something, at doing what they were doing. And as CEO, your job, number one job is to not let those people in the company. Your number one job is not having a great performance system that gets the people out. Your number, that's your number 2 job. The number one job is Make sure they don't get in the company in the first place. It just breeds the mediocrity. It breeds, you know, executive tolerance for someone. People could say who's doing the job and who's not. And when the person who's just kind of not or isn't very good at it is around for several years, it sets the tone. And when you're small, you want the A-team. It doesn't mean, in my opinion, when you realize you've brought on a B or a C player that you fire them immediately. But at least you can look yourself in the mirror and say, I need this person for a bit longer. I'm gonna look at replacing them. I'm gonna try to mentor them up to where they need to be. But if that's not realistic, I know I don't want to have this person on my team forever. And I'm like, mentally you put a plan around what that timeframe is and then you move on. Hard, but like that, a key lesson in growing a company.
