00:05:59 · George Bonaci
[00:05:59] S0: I completely agree. I think where a lot of people get tripped up is they do a lot of thinking about something, invested in it and then they think, like the minimum amount to do x is some big number. And usually that's just a problem with the experimental design. Like you can almost always scope things down enough where it is a minimal budget, whatever that might mean for your company. There's a minimal budget and you'll actually get some signal that this is going to work or not. Fortune cookies is a good example where, okay, like what was the hypothesis there? The hypothesis was that like a lot of business decision makers or finance professionals are eating out at restaurants when they're traveling or or whatever it might be and like reaching them in an unexpected way where maybe their guard is down would help build brand awareness. So that's like very high level, like hypothesis. Maybe we didn't actually need a $100,000 to go and test that hypothesis. Maybe the minimum order size was a $100,000 for this fortune cookie project, but you could probably go test to the same thing with, I don't know, some smaller mom and pop restaurant where, you know, business owners go. You could actually just go interview customers. You could try it in some other way, maybe instead of like fortune cookies where there's a large minimum order size, you could do it with an insert in the bill and you could do a partnership with a couple couple restaurants. So I think it always comes down to if the dollar amount is too big and, like, you're actually thinking about making a very risky big bet. Like, can you scope that down and learn the same thing?
00:07:11 · Guillaume Moubeche
[00:07:11] S1: Yeah. Really interesting. Another point I wanna touch base on in Ramp, I think in 2023 was valued at around like 5 or 6,000,000,000. Now it's like 44,000,000,000 this year. Total payment volume, I think, was still growing 170% year on year in March, despite the business, like, being roughly, like, 20 times larger. So how exactly do you keep growing like a startup when you're definitely, like, no longer a startup?
00:07:38 · George Bonaci
[00:07:38] S0: Yeah. I think most people, including myself, conflate, like, startup in terms of company size or valuation or or or those types of things versus culture. And I think culturally, ramp is 100% still a startup. I remember like the old Paul Graham essays and and one of the main points I always took away from them when I was thinking of being a founder and and I started a couple companies on my own was like the main advantage of startup has is speed. They can just move faster than anyone else. Like, they generally don't have more money. They generally can't attract like the best talent or the hottest talent, I should say, but it's speed. And I think whether you're a... I don't know, just starting out or or your ramp size, that's still really the competitive advantage. And I think you generally start to lose that culture the bigger you get. Again, going back to coordination costs. But I would argue that ramp is still a startup by by at least that measure. We still very much operate where velocity is the most important thing for every function, for every individual at the company. And you see that with like our product velocity, you see that with like our marketing stunts. I'm constantly impressed with what the team and individuals are able to do on their own. And so like, how do you foster and can retain that cultures as you grow? I think it's number one, just hiring the right people. I think it's number two, putting the right processes in place where they can unblock themselves. And then it's number three, I really do believe strongly in in podding teams, like having single threaded, very small teams that can run independently. There are costs to that. It probably means that you might have duplicative functions. There's designers on three different teams or there's growth engineers on four different teams. But that operational cost is well worth the ability to just have more shots on goal, run more experiments, learn new things and just fail faster as well. If you're hiring very smart, ambitious, analytical, like Swiss army knives that can do anything and like you put them on a project and it fails in two months, like that's totally fine. You can move those people a totally different project. There's no shortage of ideas and things to do, whether you're like a, whatever, million dollar company or like $100,000,000,000 company.
00:09:26 · George Bonaci, Guillaume Moubeche
Don't think anyone's ever said, ah, we have enough resources or we have too much. So it's it's really just about, like, moving fast and retaining that. [00:09:33] S1: And how exactly do you, like, incentivize, like, the different pods inside the company or inside even like the growth department? Like do they have KPIs or like how exactly do you like measure success, split them into project, etcetera?
00:09:47 · George Bonaci
[00:09:47] S0: That all has to be defined upfront. I talk about experimental design a lot, but really maybe another way to think about it is just planning ahead, outlining what are like the success metrics upfront, having some estimate on what the resourcing is going to be needed. Understanding like, even if you don't use this terminology, what is the minimum detectable effect to know if something is working? And that's basically it. I think a lot of times what happens is we... A company will say, okay, I don't know. X is a huge strategic initiative. This new product is a huge strategic initiative. It's like, Guillaume, you're in charge. Go make it successful. And then you go and you just start like, you grab some people, you pull them together, and you just start doing stuff. So really, should be like, hey, you're like, I need to understand why is it a strategic initiative? And like, what does success even look like? And are the expectations from the company level aligned with what I think it would be versus what my team thinks it would be? And then you go and you get people that are really bought into that project. I think that another key to being able to move fast and make these pods successful is people wanna be on the pod. If you are convincing someone to be on a pod, I don't know, people do their best work when they're excited to do something, not when they're forced to do something. So I think a lot of the pods also kind of form organically. It's just like, okay, here are the priorities for the company or, oh, this new product is like a really, really important thing. If someone naturally is gravitating to it, great. You're on that pod. You're you're responsible for the success that we are going to define upfront. Now having said all that, I think like the rigor to which that you're doing all of this work upfront depends upon how big the bet is and the level of resourcing. If it's like two people on a relatively small bet, you have the luxury of being a little bit more lax than something that is going to be a bigger bet, which is again why I think it's important to scope things down as as small as possible to learn. Otherwise, the coordination costs increase.
00:11:18 · Guillaume Moubeche
[00:11:18] S1: And do you have any example of one, like, pod that's been created with how exactly, like, what goal was it? Some things that that actually worked out, but I would love, you know, like, to walk through, you know, the the the first thinking of how do we make this like a bold goal up to, like, what do we measure, and is it successful or not?