The Exchanges

Q So you're Chief Product Officer at Wyze, which I don't know if you knew this, but I'm a very happy, uh, weekly active user of. To give folks a little bit of context on Wyze, could you just explain what does Wyze do and also share maybe a few stats to give people a sense of the scale that Wyze has reached at this point?

A We're looking to solve the problems associated with cross-border money movement. Which is that moving money across borders is pretty slow. It's actually really expensive, and it can be really hard to do. We solve it with three products. Our money transfer product, which is what we started with. Our account, which we like to think of, uh, trying to solve the problems of international banking with our account for people and for businesses. And then finally, we've also got an enterprise product. Where we take the underlying infrastructure that's powered those products that we built and embed them in the banks and products that people use every day, and then zooming into the numbers. So we've got to come up a little way on the journey. So today we're now moving about twelve billion dollars a month, growing at between 30 to 40% year on year. We take about .65% on average across all our routes as price. And we've been profitable for about more than four years now with 20% EBITDA margins, but probably that's that I'm most proud of. And the hardest thing to make happen out of all of that was, uh, we acquired 70% of the users that found out about Wise last month through word of mouth. So, contextually, we have sixteen million customers, and we're acquiring about a million a quarter, um, about ten million actives, and, um, yes, and a million that joined Wise the first time, 700,000 found…

AI assessment note: “We solve it with three products... So today we're now moving about twelve billion dollars”

Q I just watched that too. So good. You can actually stream it now. I don't know when this comes out, but it just, uh, yeah, you can watch it on home. Uh, but it's not cheap. I think it's like 20 bucks in the US. What is a favorite interview question you like to ask candidates when you're interviewing them?

A Yeah, it's only asked to, I got down to asking just two questions over time. The first one's probably my favorite, uh, which is, what is it that most frustrates you about, like, instead of why you're leaving, what frustrates you the most about where you're working right now? And this is as, uh, people, people always tell you why they want to join Waze or join whatever company you're coming to, and that's not that interesting, but what's interesting trying to figure out is what they're running away from. And usually there's something broken there that's really wound them up. But what's more interesting is they've been unable to fix it. And so in asking this question and probing, you kind of get quite good at getting a sense of like, what is the limit? What's the thing they found and what did they get stuck with? And you kind of say, okay, you're gonna run into that here every day, every week, or you should be fine. And that's, uh, that's kind of why I asked that question.

AI assessment note: “what frustrates you the most about where you're working right now?”

Q Um, yeah. I was gonna ask, how did you, how did you do that? That sounds like everyone would want to do that. Yeah.

A So let's try to unpick it a little bit. So first question you'd ask is, I know what you're doing. You're losing money on every transfer. It's like, especially what you do, but we've been profitable for, uh, five years. And then one of the magical things here was, uh, like we're actually profitable on every transaction. So that's probably about four or five years ago. Um, I led this project to kind of start to pull together our pricing. So every month you get bills and they turn up in your P and L, but every single bill we got, we allocated the cost back to the customer or the transaction that generated it. And then we add our margin on top and that's our price. And when you look at this and you analyze it, you'll find obviously there are, uh, 20% of customers generating 80% of the costs. And what you do is you get those 20%, you give them a rise because they should cover their costs and you drop the price for everyone else. And then the team works really hard on reducing these costs down. And then you move into a different segment in the market as the price costs come down. Does that make sense Lenny?

AI assessment note: “every single bill we got, we allocated the cost back to the customer”

Q at my survey results. I'm going to figure out these pillars that are driving word of mouth. I'm going to think about how to make things 10 times better. Just like broadly for someone that's starting to approach this, what would you say to them? How would, should they approach this? Any major learnings at a higher level of just like how to drive word of mouth? For a product.

A I think it just comes back to talking to customers and this question we've kept coming back to, like, what would it take to make it 10 X better? And then you get clear in your head what it would take. And then it's usually the thing that everyone's looked at before and thought it's the thing that's impossible. One more example is like on the partner side, like, so rather than find a cheaper bank for a banking partner, you think, well, the cheapest banking partner is the central bank. Great. And imagine you're a startup. How the hell do you get a bank account at the central bank? But that, that kind of thinking, we, we now have a bank account at the Bank of England, the National Bank of Singapore, Bank of Australia. And each of these was like as hard as, you know, getting that face-to-face verification thing in Singapore. It took like years of lobbying and all kinds of stuff in order to make it happen. Right. But it's setting your goal all the way up there. That's what enables you to build a tenant better product. That's what gets you to the word of mouth. So the first step is getting super clear on What's the problems that my customers are caring about, worrying about? And then once you're clear there, like, as you said, how, how can I solve that completely? And what's the best it could possibly be? And then the hard bit is figuring out how to move that.

AI assessment note: “I think it just comes back to talking to customers and this question”

Q What did you find most helped increase trust in that way? Is it just get more people using it and then they'll share with their friends or is there something you did there too?

A No, it was literally get more people using it and they'll share with their friends. There's obviously a bunch of learnings we've had around what specific trust sentiments matter, especially geographically, but less powerful in the macro than, than get more people to use it. So coming back to that, and I'd like to get your thoughts on this one. Um, so as you said, lots of people go up to NPS and they kind of heard people talk about it, heard people talk about recommendations. So my, my learning on this is you gotta work really hard To, to get recommendation. So to get a nine or a 10. So IPS is, uh, 70%. So it's really, really high. So it's kind of higher than the iPhone and Google, Google search. So like really, really high. And when, when we launched a market or at the beginning, it was like much lower, right? Twenties and thirties. So instead of in context, like banks and financial services, NPS is -30, right? So most people don't recommend their banks, right? So it's like comes, comes from a low base. But what I found is, uh, when you build a product, most founders and most teams kind of stop when it works. As a next step, some people focus on conversion rate. Like I'm going to make this really, really slick. Okay. And that's cool. You get a bit more growth. But to get to recommendation, you're going to blow your user's socks off. And the thing I, the phrase we use is you hav…

AI assessment note: “No, it was literally get more people using it and they'll share with their friends.”

Q Yeah. And along those lines, the other benefit of experimentation is, you know, the impact. And so team members can understand, here's what I did this quarter this year. How do you think about just like, you know, performance reviews and people's impact and that kind of thing?

A Yeah, that's a good one. This was definitely an ongoing debate. So, um, I generally ask teams, what's their impact, right? So every quarter, every team, what's the sound of Crystal will ask, what did you ship? And I generally ask how people used it, what was the impact on volume, et cetera. And we have analyst teams that can answer this either with pre-post analysis, all kinds of techniques, or all through split test. We, we generally have this. The debate is, is where the analysis slows us down, and we wouldn't make a decision off the back of the analysis. And then this generally is, uh, what you said, where the team needs it for validation, mainly for themselves, um, and maybe enough performance, but not, not too much. And so we just, there were ways in which you can maybe get some read on it that isn't quite as strong as a split test, which reducing these things. So it's more just getting some, you can understand that people worried when you do split tests, then slow down the release of something. Uh, but you just, in order to get, get an impact, if you know you're not gonna roll it back, then, okay, you should just roll it out and try to reduce the need for the validation.

AI assessment note: “I generally ask teams, what's their impact, right? So every quarter, every team”