Nilan Peiris: What Growth Hacks Worked and What Did Not, from Wise CPO | E1182

Nilan Peiris 0:00 I define product marketing as this, closing this delta between the perceived value a customer has of the products and the actual value that you deliver them. You have to be an order of magnitude better than the alternative to get word of mouth. You need to give customers an experience they did not know previously existed. (camera shutter clicks)

Harry Stebbings 0:20 Ready to go? (instrumental music plays) (mouse clicking) Milan, this is such a joy to do. I've been really looking forward to this, so thank you so much for joining me.

Nilan Peiris 0:39 Thanks for the invite, Harry. (laughs)

Harry Stebbings 0:40 Not at all. But I would love to start with some

Harry Stebbings 0:42 chronology.

Nilan Peiris 0:42 Mm-hmm.

Harry Stebbings 0:43 So how did you first make your way into the world of growth and product, and what was that real entry point for you?

Nilan Peiris 0:48 Actually, going all the way back, I, uh, I did maths at university, which I advise everybody to do out there. (laughs) Uh, but it was quite formative. I love maths, mainly as, uh ... And like, uh, training for university-level maths was about looking at problems and trying to find a completely different way to look at the problem that no one else had looked at before that enables you to solve it in a way no one had done it before.

Harry Stebbings 1:12 Sorry, I, I just have to ask that. Y- you said about kind of doing things differently and in ways that other people hadn't done before, and why, why you like maths in that way. I always wonder, are playbooks good or bad?

Nilan Peiris 1:22 In product and in growth where this reaches its fortés, when there is a, like a reproducible model, uh, when the playbook becomes something that feels like a reproducible model, and that f- is in two forms. Uh, usually there's a kind of a market expansion playbook, so we know we ... "This works. I'm gonna keep doing this-"

Harry Stebbings 1:39 Mm-hmm.

Nilan Peiris 1:39 "... again and again and again." And in my experience, it's usually disappointingly not much. Like, there's, there's very few, like, truly globally scalable products that you build once and they just kind of go everywhere. Uh, every, every country has their nuance usually then. That could be regulatory, could be financial systems, it could be cultural, and these are the things that get stuck in the way of the playbook, uh, (laughs)

Harry Stebbings 2:03 Terribly annoying.

Nilan Peiris 2:04 of expense.

Harry Stebbings 2:04 Much nicer if it was just repeatable.

Nilan Peiris 2:06 (laughs)

Harry Stebbings 2:06 I actually find playbooks can be more applicable on channels.

Nilan Peiris 2:09 Ah.

Harry Stebbings 2:09 Like there's a playbook to creating good Twitter content. There's a playbook to creating short-form video hooks.

Nilan Peiris 2:15 (laughs)

Harry Stebbings 2:15 Do you see what I mean?

Nilan Peiris 2:15 Yeah, and that type of, like, growth-hacking marketing side, absolutely. I'm sure you've, you see this, and you're f- quite successful with The Podcast Playbook. But the bit that the edge is, is usually in the creativity.

Harry Stebbings 2:29 Yeah.

Nilan Peiris 2:30 So when ... And also-

Harry Stebbings 2:31 Telling the story.

Nilan Peiris 2:32 Oh, yeah.

Harry Stebbings 2:32 I completely agree.

Nilan Peiris 2:33 Or when someone figures a different way, a different form of content, that suddenly gets cut through and suddenly everyone's doing, everyone will start copying that. But these kinds of bursts usually have a outsized outcome than the-

Harry Stebbings 2:45 Right.

Nilan Peiris 2:45 ... than the generic version.

Harry Stebbings 2:46 I agree. I always think actually to Seth Godin and the Purple Cow-

Nilan Peiris 2:49 Ah.

Harry Stebbings 2:49 ... which is kind of the power of being different-

Nilan Peiris 2:51 Yeah.

Harry Stebbings 2:51 ... not better. And I think we get to that point when we look at, like, content landscapes today-

Nilan Peiris 2:55 Yeah.

Harry Stebbings 2:55 ... for example. I, i- so, okay,

Harry Stebbings 2:57 we have that. How did you come to Wise then? 'Cause you were an advisor at Wise first, innit?

Nilan Peiris 3:00 Yeah, yeah. So just going back through the, through the long story. So I was a consultant for a while, and then, uh, uh, dot-com boom two happened and I started working with, uh, with companies getting them online. Uh, most of the shops on Oxford Street didn't have online shops, so I helped, helped build and launch their shops. And they asked me to run their shops. And once I ran a, a revenue line, I realized that's, that's what I wanted to do for, for the rest of my life, and I've been doing that for the last, um, more or less 15, 15 to 20 years (laughs) , I think even 20 years now. It's 'cause it's so hard. Like, figuring out how you get people to turn up on your site or download your product and use it. That's like, you need to do psychology, marketing, mathematics, like, technology, everything. And that's, uh, that's what gets me out of bed in the morning, and that's what I've been doing ever since. So starting off, uh, with very large companies, and then over time working with earlier and earlier stage, 'til I ran into Christo and Talbot, um, probably 12 years ago now.

Harry Stebbings 3:57 Yeah, it was 12 years ago.

Nilan Peiris 3:58 Yeah.

Harry Stebbings 3:58 I stalked the shit out of you before this.

Nilan Peiris 4:00 Yeah. (laughs)

Harry Stebbings 4:00 So, uh, I think it was 12 and a half years ago. Where was TransferWise then? Or it was-

Nilan Peiris 4:04 So, yeah.

Harry Stebbings 4:04 ... TransferWise, then, not Wise.

Nilan Peiris 4:04 So I met them ... (laughs) 'Cause you, you'll enjoy this. I got introduced to them by an angel, uh, who I had helped previously, and he said, "You've got to meet these guys. There are just two of them. They have a great product. They have no customers, and, uh, they definitely need someone like you." And so I, I met them, and, uh, they had a great product, they had no customers, and I started working with them, helping l- launch their first marketing campaigns, helping start to build a product part-time, and then, uh, after a couple of years, went full time and started building the

Nilan Peiris 4:36 team.

Harry Stebbings 4:36 Before we dive into kind of the marketing channels that we leveraged in the-

Nilan Peiris 4:39 Mm-hmm.

Harry Stebbings 4:39 ... early days, what worked, what didn't-

Nilan Peiris 4:40 Mm.

Harry Stebbings 4:40 ... just to set the foundations, like, what do Wise do, what are the products?

Nilan Peiris 4:43 Yeah.

Harry Stebbings 4:44 And set those.

Nilan Peiris 4:45 Wise is focused just on one thing, which is solving the problems with cross-border money transfer. So the one market we're focused on is the cross-border money movement market, which is huge. All right? So it's, like, about a trillion in terms of cross-border flows for consumer, about ten trillion for SMB. The problems with cross-border money transfer, like, it's slow, it's expensive, it's hard to do, and it, and it's not very transparent,

Nilan Peiris 5:10 frankly. Yeah.

Harry Stebbings 5:11 Okay, so we have this very strong belief that actually these additional fees are wrong, we should make it more transparent, and we should actually remove them in most cases-

Nilan Peiris 5:17 Yeah.

Harry Stebbings 5:17 ... rural cases. So, what is that entry then, like what were the first channels you did when you came in? How did you get the first customers?

Nilan Peiris 5:24 Our first customers came from, uh ... And the day we, we launched Wise was when the, the TechCrunch blog post went up. (laughs) And our TechCrunch, first TechCrunch interview happened, and, uh, I remember it, and then we ... The, the post went out, and then you saw the first customer that we, that we didn't know, that we hadn't found ourselves, kind of started trickling in through that. Um, post that, it was all very ...... hackey word of mouthy, right? So trying to find communities of people that were sending money internationally and starting to talk about Wise with them.

Harry Stebbings 5:59 Does that mean like seeding Facebook communities?

Nilan Peiris 6:01 When I talk through the growth model now, it'll sound very logical. But in the early days, like with every startup, you just try everything. Right? So we tried ev- you know, absolutely every single marketing channel in a very systemic way, as well as trying hard to get word of mouth to work.

Harry Stebbings 6:15 What worked?

Nilan Peiris 6:16 Everything did work, but to differing extents. So, um, we had, um, paid marketing working well. We still do. It's about 30% of the 1.5 million customers we acquired last quarter came in through paid, about 70% come- came in through word of mouth.

Harry Stebbings 6:34 Mm-hmm.

Nilan Peiris 6:34 And if you went back, uh, 10, 12 years, it was about 30% through paid, 70% through word of mouth. And that paid's been really important, and it works really hard. Getting underneath the skin of it, what- what's really working there was social for us, which started off being Facebook, now has gone increasingly towards YouTube as well.

Harry Stebbings 6:53 Really?

Nilan Peiris 6:53 Yeah. And on-

Harry Stebbings 6:55 What, w- w- what, talk to me, like what have been lessons in making YouTube work? That's interesting.

Nilan Peiris 6:58 Uh, so it's taking a step back, why social worked for us is it, is ultimately word of mouth work. So people telling their friends about Wise and this great product is, and driving word of mouth is what's been the big driver of Wise's growth. And when the marketing amplifies that, that's when it works best. So, um, Facebook supports marketing to your friends. So if you have a- a usage of Wise, then it's likely that your friendship group may need to send money internationally. Think French students in London know French students in London. Uh, small business owners who are exporting will know other small business owners that are exporting. And when you advertise to and target to- to people who are similar to your existing customer base, then this becomes pretty effective. So that- that's where the Facebook thing started. On YouTube, this more came out of our SEO playbook around creating really useful content for end users.

Harry Stebbings 7:53 What does that mean, sorry, creating useful content for end users?

Nilan Peiris 7:56 The way- way I think about content, um, starting with SEO and then going to YouTube, is, uh, there's two broad kinds of content that I see people push out there, uh, and I encourage people building- building startups to do. One is what I call on-topic content. And you should really think through how you can own all the on-topic content in your, um, chosen vertical. And then there's the- the off-topic adjacent. So I'll talk you through with examples. So on the on-topic side, um, for Wise, if it's sending money internationally, you wanna build the best page on the internet that tells people how to send money to India, for example. Right? But then you can go a step further. You know your customers that are with Barclays. So if you build a better page than Barclays has on how to use Barclays to send money internationally, then people will link to this page, and it'll naturally go up the rankings, and you'll start getting traffic on Barclays, on Western Union, et cetera. If you start helping people understand their fees and the exchange rate, this page becomes even more useful. What else do you need? When you send money internationally, you need those like pesky BIC and IBAN codes.

Harry Stebbings 9:02 Mm-hmm.

Nilan Peiris 9:02 So obviously if you search for a BIC or an IBAN code, you'll find Wise there. It's 'cause we've built pages on every BIC and IBAN code. We bought and built our own, uh, BIC, uh, calculator and IBAN calculator and converters so that you can look up your own bank's BIC and IBAN, which is surprisingly hard to do with some banks. (laughs) Um, so yeah, making really useful on-topic content for people trying to- to move money internationally is- is cool.

Harry Stebbings 9:27 (sighs) It reminds me of actually Zoopla's strategy of having, where you put in your postcode and it'll do like a guesstimate of your house price.

Nilan Peiris 9:32 Yeah.

Harry Stebbings 9:32 And it's a very useful tool, which is actually an incredible lead gen tactic.

Nilan Peiris 9:37 Yeah. Absolutely. So that's the- the on-topic one. And the off-topic is where you go slightly- slightly further. So maybe you're looking at, uh, content that's useful for students living in London. And yeah, and then that naturally turns into YouTube content, if that makes sense. Um, increasingly we even have some of our own, uh, sub-brands running this stuff focused on different- different demographics and different markets.

Harry Stebbings 9:59 What didn't work?

Nilan Peiris 10:00 So what's really hard to do is above-the-line, and getting an above-the-line to work, and I think we've run, we've run above-the-line TV campaigns-

Harry Stebbings 10:09 For those that don't understand, what is above-the-line?

Nilan Peiris 10:10 Oh, sorry. Above-the-line is- is brand advertising.

Harry Stebbings 10:12 Yeah.

Nilan Peiris 10:12 So you have like gen- generalizing terribly, there's two types of marketing, performance marketing where you spend some money and you try to measure a user turning up.

Harry Stebbings 10:21 Yeah.

Nilan Peiris 10:21 And then brand where you- you spend money, and you can do it relatively strategically, you could buil- you could spend the money to solve perception problem in the market, uh, to raise awareness of your product. Uh, and then you have to m- measure in like maybe a much more difficult way, uh, what the impact of this is. And the things that, uh, struggled to work were getting above-the-line to work. When you look at our marketing mix, it looks light on that. And- and we still wouldn't say have cracked it. It's hard, right? So our customers aren't, uh, everybody in the UK, for example, or everybody in Australia. See-

Harry Stebbings 10:53 Specifically, why is it hard to get above-the-line working, and why do you think you've struggled?

Nilan Peiris 10:57 So I think there's two parts that make above- that will make above-the-line easier. If it's, if you have a mass market product that's relevant to everyone, then generally, mass market channels will work. And then the second part is who else is competing for that eyeball at that point in time? Can they monetize that user better than we can, and therefore they can spend more on acquiring the customer? So these are the- the economical side of this, and then, uh, the kind of getting a- acquiring a targeted audience. These are the- the parameters that make it a little

Nilan Peiris 11:26 challenging.

Harry Stebbings 11:26 Can I be blunt?

Nilan Peiris 11:27 Yeah, go for it.

Harry Stebbings 11:28 K- I don't know how spicy I can be.

Nilan Peiris 11:29 No, let's branch off.

Harry Stebbings 11:30 Revolut have so many different products today.

Nilan Peiris 11:32 Yeah, they are.

Harry Stebbings 11:33 I'm sure they can monetize across those different products if they wanted to more efficiently on a per customer basis. Like they could extract more value from them.

Nilan Peiris 11:41 That's a good question. We should definitely go deep into this. So leaving the marketing to aside, um, I think let's talk about FinTech and neobank business models, right, in general.So our overall business strategy is that, um, we drive our cost down every quarter. And then every quarter, after we reduce our costs, we try to-- we add our margin on top, and that's our price. Okay? And the kind of th- strategic thesis at the heart of Wise is whoever has the, uh, lowest cost, highest quality product will take the whole cross-border market. The challenge to this is if somebody kind of cross-subsidizes this cross-border stream with another stream where they're earning a lot of income on. So neobanks have two other principal streams that are really valuable, uh, holding deposits and what's called interest income from their customers, and the second is lending. Let's, let's come back to lending. So the, without, without talking about Revolut specifically, but just talking in general, like, any, any neobank could, or any bank, full stop, could take its interest income and use it to offer cross-border transfers for free, right, example. Yeah. There's, uh, so there's two challenges with this. One is, like, it's actually very expensive moving money internationally, so you couldn't, you couldn't do it for free, right? But you could kind of maybe subsidize the cost a little bit. So working it backwards, like, um, if your customers are moving 10 billion and it costs you 3% to move that, that's an awful lot of your interest income that will get, go in, in moving this. So it's, it's, uh, incredibly expensive covering the cost. The second challenge, though, is what happens if customers are, are, uh, expect to get all of their interest back? So... And that's the trend that we've begun to see and begun to drive in the market around this. So historically, banks offer low retail deposit rates and then lend out at much higher amounts, and that difference is called the bank's net interest margin, is how they make m- money. Through, over the last two, three year, two years, as interest rates have gone up, what we've seen is a number of innovative products entering the market, where consumers can now get access to central, effectively central bank rates, which is what the bank gets, which is much higher, about between 3 and 5% at the moment. Wise, we offer this to our own customers as well. So in Wise, you can, in the Wise account, you can push one button and hold all of your cash in government bonds, and that means you'll get a 3 to 5% interest rate. When you spend with your card or you send money out, we aggregate up and sell those bonds. When you put more money in, we put more in. But what this means is we're in the market with an interest rate of 4%. So if a neobank starts, like, uh, using that money to subsidize another pro- product, it becomes unfeasible because then people would churn. So I think these two bits, right? Running the service, the cross-border services l- for as low cost as we can. Others have a much higher transaction cost of running it. Two, putting pressure on the other income streams that other neobanks would have around net interest margin, I think end up with a long term better outcome for customers.

Harry Stebbings 15:00 Back to above the line being so challenging.

Nilan Peiris 15:02 Mm-hmm.

Harry Stebbings 15:03 I'm just intrigued. Why bother doing it, then? If it's so difficult to do, and you can actually get a relatively guaranteed performance marketing strategy-

Nilan Peiris 15:11 (smacks lips) Yeah.

Harry Stebbings 15:12 ... why bother?

Nilan Peiris 15:13 Just you try everything. So, like, and I'm sure e- when you're, when you're starting out and even still today, you still keep trying everything to test to see whether the assumptions you had previously are still true, whether someone's come up with a, a new creative, uh, or a new creative way of using the format that gets cut through and works better.

Harry Stebbings 15:30 Can I ask, did stay the same over time? Do they get cheaper? I always think this with companies as an ambassador.

Nilan Peiris 15:36 Mm-hmm.

Harry Stebbings 15:36 They are like, "They should get cheaper over time." Because you build brand. Word of mouth into use more. You become more well-known. But then part of me thinks they get more expensive because you saturate your core market, and you're reaching kind of further and further edge cases in terms of customers.

Nilan Peiris 15:51 I-

Harry Stebbings 15:51 Which one is it?

Nilan Peiris 15:52 I think that's pretty smart, what you just said there. (laughs) So I think it is, uh, I think both those things happen literally at the same time. So as you start out, usually there isn't much competition, especially maybe on the very targeted audiences that you're going after at the beginning. Uh, and then over time, as the competition enters, it kind of goes up in the market, and then slowly, slowly over time comes down. For us, uh, we had this really fun challenge. There were small, there's still a, a bunch of the market, not much, so I'd say about 10, 15% of the market that will Google, uh, before they send money. So they'll Google, "Send money, uh, cheapest way to send money to India." Right? And, uh, people were doing this before Wise existed and custom- and companies like Western Union existed there. Western Union w- is, like, uh, seven to eight times more expensive than Wise. So they could spend seven to eight times more than us on marketing. And therefore, there's no way we would ever be able to compete on those, on those SCM key head terms. So, that was a point at which when we started realizing the economics of this, we pushed so hard towards word of mouth. And getting to, moving on to word of mouth, what it became was understanding, what would it take to build a product that's so good that people would talk about it? How do you instrument that product? And how do you organize a, a company and systemize it to enable and drive that word of mouth

Nilan Peiris 17:21 growth?

Harry Stebbings 17:21 This is too good. So, like, I always have founders who are very product-led and they're like, "You know, I built amazing product and customers will come." And I go, "No. That distribution is a product in itself."

Nilan Peiris 17:33 Yeah, totally.

Harry Stebbings 17:34 "And we must create an archetype that..." Do you agree with me? Or is that what you're saying, in terms of create such a great product that people talk about it?

Nilan Peiris 17:42 Totally. But, uh, so just stepping it through and making that more practical. Um, f- with us, customers talked about price, talked about speed, talked about ease of use. You can measure price. You can measure speed. You can measure ease of use.

Harry Stebbings 17:57 Mm-hmm.

Nilan Peiris 17:58 When we enter a market...... we find that if we are, um, slightly cheaper than the incumbents, then we don't really get hockey stick-like growth.

Harry Stebbings 18:09 Mm-hmm.

Nilan Peiris 18:10 Banks are at 6%. We're at .3 in most markets. When we launch in some markets, say, we were at 5.8. When you're at 5.8, you don't get any growth. So we let... You have to be an order of magnitude better than the alternative to get word of mouth. You need to give customers an experience they did not know previously existed. So when founders say, "I built a great product," it needs to be that great. People didn't know you could move money instantly, right? So we're moving about 60 billion pounds instantly from one side of the world to another, and that is the kind of experience to get people talking about it.

Harry Stebbings 18:46 You do, but actually, I think you can be taken for granted unless you message it well. My question to you is, like, unless you're very explicit about those, like, benefits, it's gonna be very challenging for consumers to understand them. If you told me, "Harry, naught point three percent transfer fee," I'd be like, "Cool." But it wouldn't strike me that, shit, that's a twentieth of the cost of the competitor. So how do you think about product marketing and messaging-

Nilan Peiris 19:14 (laughs) Of course.

Harry Stebbings 19:14 ... the measured data of how you're better?

Nilan Peiris 19:17 I think you more or less got to my definition of product marketing. So actually, it was my dad that taught me this (laughs) . Um, so my dad used Wise, and then he'd use Barclays. And then he'd tell me, like, he'd see, he'd see Barclays would say it's three pounds, and he'd see with Wise it would cost 2.99, and he'd miss the whole hidden fee in the exchange rate. And so he'd tell me, "Yeah, I like your products. I mean, it saves me, like, 10, 20 P." And I realized that even my dad didn't understand how much he was saving. So we spent a lot of time trying to create a, a UX that explains to customers really clearly, like a picture, that says, "This is how much you've, uh, how much this, this transfer costs you, and this is how much you would've spent with your bank," by showing the rate and how much was hidden in the rate. And we iterated, and iterated, and iterated, and iterated this till the point that when we show this to people, they're like, "Oh my God, I didn't realize it was that amount." And then we learned, I learned one of the biggest areas of growth that we had was when we took that picture and we put that on the success screen. So once you've done a transfer and it completed, it's like, "You saved 25 pounds." They believe they saved some money, but they didn't believe the 25 pounds. But then we put this, what we call a comparison graph, on that page that really, like, explains to you where exactly this number comes from. People were like, "Oh my God." And this thing started getting screenshots and shared on Twitter. We saw, like, I think it was a three to 4X increase in our recommendation rates off the back of that. And what's changed here is we haven't changed anything. Customers are still saving the same amount, but now they know what the value we've delivered is. And so I define product marketing as, um, this... Closing this delta between the perceived value a customer has of the products, and the actual value that you've delivered them, which is usually could be higher and is in our, in our case. And our... The, the way I look at the return on our product marketing is, how much have we closed that? And I know that if our customers, uh, see, believe they're h- having more value from Wise, they'll recommend more. So that's kind of how it, how it all works through.

Harry Stebbings 21:25 I totally get you. Do you know what I think now the best thing is, like, how do you create UGC on the savings gap that you create, which is, like, how do you create... Congratulations, you saved $25 or 25 pounds. With that, you could have bought 33,000 straws.

Nilan Peiris 21:39 You've got it. (laughs) Yeah, all that. All that kind of stuff, yeah, yeah, yeah.

Harry Stebbings 21:40 Do you know what I mean? Like, like, a shareable thing, where it's, like, "Thirty-three thousand straws, huh?"

Nilan Peiris 21:45 Yeah.

Harry Stebbings 21:45 "That's fucking random."

Nilan Peiris 21:46 The other one that we had this on was on speed, as, like, 60% of our transfers are, uh, are instant. And customers would go through the transfer and get to the page, and we'll say, "The money's there." And then I remember a research team, like, interviewing these customers and asking customers, "Where do you think the money is?" And they go, "I don't know. It's with Wise. It's in the account." You realize customers didn't believe the money was actually gone from the UK to Australia and was available for somebody to spend in their bank account in under 20 seconds. That's how we define instant. And so we built a little whizzy animation that kind of, like, shows the money going and then landing there. And then, and then when we ask people, "Where's the money?" they go, "It's in the other person's account." That's amazing. They can go spend it. So again, like, this UGC really bringing to life that. And we could have a lot of fun with the... That went faster there than the... Like, sometimes, like, not as fast as speed of light, but pretty, pretty damn close sometimes.

Nilan Peiris 22:40 (laughs)

Harry Stebbings 22:40 How quickly do you know when a product is working, a new product is working?

Nilan Peiris 22:44 Yeah, it's pretty fast when you see the kind of, like, hockey stick happen. A good example is Brazil. When we launched the Wise account in Brazil, it, it went like that. It's still, it's still going like that in Brazil. And that's mainly because we, we found a way to help Brazilians move money internationally really cheaply, especially when traveling.

Harry Stebbings 23:05 In a way that Nubank haven't?

Nilan Peiris 23:07 In a way that Nubank haven't, to the point of this is what the... our partnership with Nubank is. So Nubank have, uh, offered the Wise account now to their customers, and we, we have Nubank and Wise co-branded cards to their ultra violator, their most, most valuable customers. And we continue to work with Nubank to kind of improve and build out that proposition.

Harry Stebbings 23:26 How do you know when to kill a product?

Nilan Peiris 23:28 We know when to kill a product usually when it's not working. When it's not, uh, when it's not paying for itself, it's a good sign. Um-

Harry Stebbings 23:35 My question is it obviously doesn't pay for itself, and I, I think about content a lot in this. Why, why have we been successful? Honestly, it's not my dulcet tones or my charm, clearly. It's... I just grinded longer than anyone else. And for three years, I didn't make any money, and we didn't have a thousand plays on the show.

Nilan Peiris 23:51 Yeah.

Harry Stebbings 23:51 And so, like, the payback period is so long.

Nilan Peiris 23:54 Yes.

Harry Stebbings 23:54 So... And we... It wasn't obviously working. And so I find often it doesn't pay for itself until it does.

Nilan Peiris 24:01 That's so entirely fair.

Harry Stebbings 24:01 And then it really does.

Nilan Peiris 24:03 ... ya, the- that's (laughs) a really good question. So one is when it doesn't pay back for itself and that- and you're right, that isn't a problem, but it does become a problem if, imagine the costs of running it become really high.

Harry Stebbings 24:15 Mm.

Nilan Peiris 24:16 So when the costs are marginal, you can kinda keep something ticking over. When the costs become really high, you have to pay for that, and that means you've gotta charge this customer over here to keep this going, and then suddenly this becomes less competitive, so we- we've learned to be s- very disciplined on this front. The second point is, it's not the financial cost that's the real inhibitor. It's the people cost. It's your time. So how much time have you got (laughs) to keep running a product that you hope will take off? And what else could you be... what's the opportunity cost, and what- what else could you be doing with that time? And there are some things where teams have grinded away and grinded away and grinded away at Wise and then things have taken off. And then there are other ones where they've grinds away, grinded away, and then there's this- been this other thing that started moving really fast that they've jumped on top of and killed this one instead. And it makes sense.

Harry Stebbings 25:05 I guess, yeah, I guess there's also like this outcome scenario planning which is important as well, which is like, to what extent is the upside capped, unlimited, a needle mover? And if it's like a massive needle mover for the company, which could be unlocking South America-

Nilan Peiris 25:20 Yeah.

Harry Stebbings 25:20 ... massive, fuck, we could probably hold out longer to see. But if it's like, "Mm, actually SMBs in south of England need this," or it's a much more, whatever that is, that outcome is smaller, fucking cut it quicker.

Nilan Peiris 25:33 So it's just, uh, I mean, you- you figure it all out. So (laughs) that's absolutely right. So what we do is we, uh, we have in our heads, like, what is... And Google, Google have this famously, like has to be a billion-dollar business. So we have, what is the minimum hurdle rate that something needs to hit? And you're right, we don't time bound it, but we shouldn't be working on things that can never be huge.

Harry Stebbings 25:53 Yeah.

Nilan Peiris 25:53 So everything has to be huge, the, every- every team at Wise is clear what huge means in their context.

Harry Stebbings 25:57 What is the minimum hurdle rate for you?

Nilan Peiris 25:59 Um, very big. Unfortunately as a public company (laughs) I don't think I can share the specifics but it's- it's very big, and the bit I'll share back is, generally, yeah, it should cover the... We look for payback generally within 12 to 24 months.

Harry Stebbings 26:13 Mm.

Nilan Peiris 26:13 So you look at the costs, how much would it cost, and then you can work backwards to what that figure is.

Harry Stebbings 26:18 Does that frustrate you, being a public company? Because you have less flexibility on extended paybacks, because public markets are essentially saying, "Hey..." I mean, they're- they're a short-term voting machine in many respects, not a long-term weighing machine -

Nilan Peiris 26:31 Yeah, yeah, you're correct, yeah.

Harry Stebbings 26:32 ... if we were to go back to that. Yeah, yeah. Um, and so respectfully, they don't always give you the luxury of 24 to 36 months for a more outlandish bet that could pay back. They want the much more tangible short term. Is that frustrating for you?

Nilan Peiris 26:45 No, no, no, I think, uh, I think we've... So we haven't changed since we went public, and a, a really good example of, uh, a very long-term bet that is- is weighed in the, in the public market very visibly is our investment in price. So we, we dropped price, uh, we've dropped price quite a lot this year so far. So say we, say we drop price by 10%. There's not 10% of, uh, marginal customers hitting our product that will then switch to Wise at that point in time. Does that make sense?

Harry Stebbings 27:19 Yeah.

Nilan Peiris 27:19 So if- so you drop price by 3%, it's not like magically there's a 3% who weren't gonna use Wise and now are gonna use Wise, so incrementally we got more revenue.

Harry Stebbings 27:28 Yeah.

Nilan Peiris 27:28 So this is much more a long-term bet, kind of like Amazons that you keep chipping away at driving down your cost and your price, and eventually the market will switch to you. And so we have... we very publicly keep... and so the people looking for us, uh, investors, uh, looking for us to kind of like optimize the company for maximizing growth in the next three months, like, this doesn't make any sense. Maximizing growth in six months doesn't make s- any sense. But maximizing growth over a five-year horizon or three to five-year horizon, where, uh, this makes a lot of sense. So the public markets, yep, they're there, and they will, uh, there are, you know, you can't s-... people will invest in- in the stock, people will look at the short-term results and make their- make their own judgments about it. But there- there's enough, uh, enough investors out there that buy into our longer-term vision, uh, on price, which means, uh, we can get a- we can keep doing this. On the challenging side, though, there will always be- there'll always be noise from- from people that don't buy into your strategy, uh, but as long as you, as long as you can find people that buy into your, into your stock, you're- you're fine.

Harry Stebbings 28:34 Do you worry with the outcome scenario plan, like estimating upside?

Nilan Peiris 28:38 Mm.

Harry Stebbings 28:38 The challenge is some of the biggest outcomes come from toys, or come from non-obvious ideas that started off as side projects. You know, one of the engineers coming in -

Nilan Peiris 28:48 Yeah, yeah, yeah, totally.

Harry Stebbings 28:49 ... tinkering about an... and actually it wasn't a big ambitious play. It was kind of something that someone just kind of tried, and whoa, it took off.

Nilan Peiris 28:57 Yes. That, this is true. This has happened. So we- we encourage this. So a fun bit we (laughs) we haven't got to is how we run. So historically, we ran in autonomous independent teams, and that literally is what it- it sounds like. So when I, uh, joined, so about 10, 10-odd years ago, um, every quarter we'd get together and every team would talk about their plan. They'd get feedback from everyone around, including customer support to Christo and Tavet, and then they'd adjust their plans, and then they'd execute, and then they'd share back how- how they did. And this kind of made logical sense. We had teams around each of the big problems you need to solve to get to low-cost instant transfers. We had a team on driving down the costs of KYCing a customer, a team on making it easier to onboard, a team on, um, making the pay-ins and the payouts really cheap and quick.

Harry Stebbings 29:54 So just so we understand these verticalized teams are by product or by function?

Nilan Peiris 29:58 They're cross-functional teams with everything you need in order to move those KPIs.

Harry Stebbings 30:02 Gosh, so they're like independent companies?

Nilan Peiris 30:04 Yeah, but with a matrix structure where all the engineers report up to CTO, all the product managers up to- to me, and designers up to the VP of design, et cetera. So it's-

Harry Stebbings 30:13 And they're all fighting for resources?

Nilan Peiris 30:14 No, they're all able to... if- if a team's doing well, then... and it's able to invest, they can ask for more resources and able to- to take off, off the-

Harry Stebbings 30:23 Was that a good structure?

Nilan Peiris 30:25 That worked. Uh, and it worked...... up to a point. Now we have teams into squads, and squads into tribes, and, and these guys and, and ladies hold the teams accountable in very much the same way. So, the strategy still does come bottom-up. Does that make sense? Because the teams are still incentivized through this, but the accountability is there around their local leaders who, who make sure, make sure that it's working.

Harry Stebbings 30:51 I don't understand, sorry, the transition from these int- independent autonomous units to squads, and teams, and tribes.

Nilan Peiris 30:57 Yeah, I'll step you through. So, um, a good way of explaining this is like, we, as I shared at the beginning, every quarter they would stand up and talk through their plans. This works fine when it's six teams. When it got to 30 or 40, it stopped working. So we started grouping those teams into units.

Harry Stebbings 31:16 Mm-hmm.

Nilan Peiris 31:16 Call them squads.

Harry Stebbings 31:17 Yep.

Nilan Peiris 31:18 And then you kind of hired somebody to run the squad (laughs) and tell them, "You're accountable for the outcomes of the squad." And so that same structure that we were running when we were a startup is kind of happening within the squads. Does that make much more sense?

Harry Stebbings 31:31 Totally does.

Nilan Peiris 31:32 Yeah.

Harry Stebbings 31:32 Where does that break down?

Nilan Peiris 31:34 So we went from squads to tribes. It, it keeps laddering up, so it works. It even works when you're public, because I, when I stand up and talk through our product strategy, I've kind of like a whole, uh, engine that synthesizes all of these up and helps me talk through to the market, "Where are we going? What are the big levers of growth in Wise?" And like, like I talked you through at the beginning, where we have a money transfer business, we have a account business, we have an enterprise business, and could, yeah, explain what our strategy is in each of these areas. But they all come bottom-up rather than me saying, "I think our strategy in X should be, should

Nilan Peiris 32:07 be this."

Harry Stebbings 32:07 You, you mentioned the different business units.

Nilan Peiris 32:09 Yeah.

Harry Stebbings 32:09 And there's quite a few. You mentioned like, like 30, moving from six to, like, 30. But how do you think about effective product marketing when you have so many different ones? It's so easy when it's like, "Hey," like, fast and, you know, quick. Done. (snaps fingers) But when it's 30, and you have, like, multicurrency accounts, and cards, and partnership members, uh, how do you do it?

Nilan Peiris 32:31 Yeah, it's not easy, and I wouldn't say we've done it particularly well, still work in progress. But the answer is you devolve it as much as possible. Great example is Wise platform, uh, where we're taking Wise, our infrastructure out, and we're selling it to banks. So most of the neobanks in the world use Wise. Google Wallet is powered by Wise. Uh, Stripe use us. Uh, so Platforms use us too. But probably more excitingly, uh, some of the largest banks in the world are starting to use Wise. It's done very well in APAC, so some of, uh, the largest bank in Indonesia, um, second-largest in Japan, third-largest in South Korea.

Harry Stebbings 33:08 So you run it as a separate business, really?

Nilan Peiris 33:10 Within Wise. And so to get to your product marketing question, this team, uh, and this squad within Wise has to do its own marketing. So we run, I, I kickstarted it, Wise Connect, our enterprise conference. We run it in, in London, in Singapore, and in San Francisco, and we've gone from a standing start to doing that in, like, uh, I think we're now in our second year of running it three times. And this time round in London, we managed to fill a room with 200 bankers from Tier 1 banks in Europe. So we do have a centralized marketing function, but really, it's devolving and embedding marketing within those teams.

Harry Stebbings 33:47 How do you prevent brand confusion when you have successful product marketing in one category, when it's like, "Hey, we've sold so well in terms of product marketing for simple and cheap..." And then the bank is like, "Really?"

Nilan Peiris 33:58 Yeah, yeah.

Harry Stebbings 33:58 "We're gonna use Wise

Nilan Peiris 33:59 in B2B?" Or going, or, or going further, like, uh, consumer marketing can feel like it's, it's bashing banks, like, how do we, how do we tailor that? Two bits here. One is, w- we do have a very strong brand marketing team that puts the architecture in place around all of this and thinks through, like, what is, what is the Wise brand, what is the Platform brand, et cetera. And then from a creative strategy, everyone understands that we're gonna end up becoming an API business at Wise. Like, um, in a five to ten-year period, the majority of our customers won't be using our apps. They'll be in banks. And therefore, we need to evolve our marketing strategy. We can't be having, having stuff here that, that makes it harder for that to happen. So everyone's-

Harry Stebbings 34:42 Do you need a brand if you're gonna be an API business in ten years?

Nilan Peiris 34:45 We absolutely need a brand, (laughs) if we're gonna be an API. This, it just, it's just a different kind of brand. It's a B2B brand.

Harry Stebbings 34:51 You don't need a consumer brand, you need a B2B brand.

Nilan Peiris 34:53 So there's a very fun part around how the consumer business and the B2B business work with each other. No bank would ever, would ever integrate Wise, um, w- would ever use our infrastructure, if our consumer business didn't exist. Our consumer business sets the expectations in the market of, "It's gotta be cheap. It's gotta be easy to use." Our consumer business has taken something like between 10 to 20% of some of the major banks in the UK's volume.

Harry Stebbings 35:23 I remember when we, I interviewed, uh, Terry Angelos from Mastercard-

Nilan Peiris 35:27 Yeah.

Harry Stebbings 35:27 ... when he was running the M&A team, and he t- said to me, "You ever wonder why these massive card, you know, transaction machines sponsor football tournaments?" And it's because consumer trust is everything. And so when you go into a store in Colombia, and you see Visa or Mastercard, as a consumer, you've seen it on the side of the Bernabeu, and you're like, "Ah, trust. I, I know that brand." And that is the power of brand for us and for Visa.

Nilan Peiris 35:53 So, uh, we stumbled on this earlier at Wise. So obviously, ev- anyone out there starting a fintech business, when you talk to your customers and find out why they aren't using you, it's because they don't trust you. And the first thing anyone will tell you is, "The way to build trust is brand marketing." Or, and actually it's really hard, actually, to build trust, because you're, like, my dad was used to using his bank, and, like, you can go into a bank, you can touch a bank, you can feel a bank. And suddenly, you see an ad on the internet that says, "Cheap money transfer," and you're gonna click it, and you trust it? You have to spend a ton of money on marketing...... to get somebody to trust you with their money. And the pivot we made was realizing that it's much easier to get people to trust their friends. They already trust their friends, and if you focus on giving them a great service, then that covers the trust thing for us.

Harry Stebbings 36:43 So how does that change your strategy then?

Nilan Peiris 36:45 No, not really. So what we learned was like, we do have all the usual things you'd expect us to invest in from a product marketing perspective, to tell customers, "This product is safe." However, when we're a 70% word of mouth business, I'm under no illusion that most of the work on getting people to trust our product is from our customers telling their friends, "It's safe. My money turned up at the other end."

Harry Stebbings 37:07 But, so like, you know, I had Val Schultz, who was the first head of growth at Revolut on the show.

Nilan Peiris 37:11 Mm-hmm.

Harry Stebbings 37:11 And he was really interesting. He was like, "We changed product strategy to really focus on group payments, because fundamentally social pressure drives adoption, and actually if you have nine people playing football and there's 11 in the team, they'll say, 'Fuck's sake, Melant, like get a, get a Revolut card so we can actually pay for the pitch together.'" And it changed product strategy by realizing that friends drive adoption.

Nilan Peiris 37:36 That's very smart.

Harry Stebbings 37:37 I thought it was quite a good one.

Nilan Peiris 37:38 Yeah, it's very, very smart. I think like lots of the neobanks, when you look at Monzo, you look at Chime, they find a, a hack that really drives adoption. I think Chime with get, get paid a day early because of the quirks of the US payment system, you can get paid a day early if you switch to using Chime. People switched to using Chime. And then similar this k- type of social pressure and some of the features Revolut built out in the early days definitely helped drive their adoption.

Harry Stebbings 38:02 What hack you did worked best?

Nilan Peiris 38:04 I think it's stuff I've covered already, just sharing with people how much they've actually saved is probably the single biggest. The other part I didn't share. So within this 70% that's, uh, word of mouth about, um, say 20 to 30 is refer a friend and the rest is pure word of mouth. And the refer a friend, we've tried everything under the sun. So we've tried sending chocolate. (laughs) Like, you refer someone to Wise, you get money, like, I get money, you get money, we both get money. We've played with different amounts of money.

Harry Stebbings 38:33 From an economic psychology perspective, does the amount of money that you offer as an incentive drive a correlatory return?

Nilan Peiris 38:40 So what happens is, it drives users, but it doesn't drive what I call customers. (laughs) So, a customer d- if you define as a profitable user, like it's very easy to drive users, right? Because if you give a incentive like that, but when you s- when you look at the growth off the back of these, like, um, like giving a, a very heavy incentive, you'll see something that looks like a hockey stick. If you look at revenue or like profit, it'll look m- much, much more tame, right? And then when you try to understand the marginal cost of the extra incentive, it, it generally will fall off.

Harry Stebbings 39:09 How do you think about knowing the CAC budget that you have? 'Cause like LTV is varying and I think actually for banks today it's

Harry Stebbings 39:18 becoming increasingly challenging. Before banks had FinTech products, you had incredible stickiness.

Nilan Peiris 39:23 Mm-hmm.

Harry Stebbings 39:23 And your LTVs were long actually. Now I think you have this increasing transience between different services.

Nilan Peiris 39:29 Mm-hmm.

Harry Stebbings 39:29 Most Monzo users are also Revolut users.

Nilan Peiris 39:31 Yeah.

Harry Stebbings 39:31 They're also probably a traditional bank user as well from their teenage years, they kind of have... And so the LTVs are more variable, which changes your CAC budget. How do you think about that?

Nilan Peiris 39:43 So I think the best marketeers, uh, performance marketeers that I've worked with over the last, uh, 10 to 15 years, they know this. So in businesses I've built previously and just some of the people say that built Booking.com or Zalando or some of the early rocket guys, it was always around, um, when you are advertising, say you are advertising, um, on SEM, on search engine marketing, and you get better and better at breaking down into smaller and smaller groups your campaigns. For that campaign you understand really well the cost, and for that campaign you understand really well the LTV.

Harry Stebbings 40:23 Mm-hmm.

Nilan Peiris 40:23 And every campaign you run will have a different LTV and you need to kind of aggregate up campaigns to a level that you can get a decent read on the LTV. And the other way I think is it's, at the end of the day what you're doing is making a bet. So I'm gonna spend 50 pounds on acquiring, on my CAC 'cause I think I'm gonna earn 50 pounds back in a year. And the other way

Narrator 40:49 (laughs)

Nilan Peiris 40:49 ... of our marketing teams, so when I've spent that 50 pounds, I am now in the red, (laughs) right, in acquiring the customer, and I'm hoping it's gonna track back to this. And the discipline of tracking how much are we down (laughs) from all the bets that we've spent and what's our floor in terms of how much re- you're prepared to spend and how much you're prepared to lose, because it only needs to take... Like, you're not in control of whether these users are gonna keep coming back and using your product. There's hundreds of other variables in there. And that's the fastest way in which businesses go bust, right? When they spend a bunch of money acquiring customers that, that turn up, but the LTV doesn't turn up in the way that you expected.

Harry Stebbings 41:30 When have you got it wrong and what did you learn from that?

Nilan Peiris 41:32 Probably the best example of this was how we became profitable. So we became profitable about four to five years ago, and every year we would go raise money prior to this point. And every year, you probably know this from your portfolio, every year when we raised money we told ourselves, "This is the last time we're gonna raise money." We do a-

Harry Stebbings 41:51 Always, always the last time. (laughs)

Nilan Peiris 41:52 Yeah. And we, we'd do, we'd do a little business plan that said we're gonna be profitable in like 12 to 24 months, and then every year (laughs) we get to this point like 12 to 24 months later and we've come up with all these really good ideas that we should invest in, uh, which could be engineering ideas, they could be marketing ideas. And we, we'd come up with these ideas, we thought they could work and then every year we'd need to go raise again. And then we got to this point where it was about four to five years ago and winter was beginning to set in, um, and we could raise but we decided it would actually be far better to become masters of our own destiny and become profitable.We, we became profitable in about, uh, six to nine months. And there was only one cost we changed to get there, which was our marketing spend. And I think I've shared this publicly before, and this is about four years ago. We halved our marketing spend. We dropped 10% on customers and 1% on revenue. And when you're in a business that's going grow, grow, grow, grow, grow, and then you turn around and you tell the marketing team and you work through the detail of it, "We now need to get to profitability," this type of trick is hiding in there. And I've learned now, one, after I did that once it was like, this kind of inefficiency will build up all the time in marketing. So about two or three years later, we kind of ran the same drill without a need to do it, say, "We're going to halve the marketing spend. Tell me how much we're going to drop revenue." And we barely dropped revenue, but made the product, made the company

Harry Stebbings 43:21 Why have a bother increasing it?

Nilan Peiris 43:22 ... work. Exact- But the point is, it always, it always generates a bit more revenue as you can see, (laughs) but there's a point at which it becomes less and less and less efficient. And unless you go back and do this, you, uh, you kind of miss this massive opportunity (laughs) that's hiding in there.

Harry Stebbings 43:36 But for the 1%, is it worth it?

Nilan Peiris 43:38 No, it's obvi- obviously not, but it doesn't start off with the 1%. It starts off with the, "Let's increase it by 5% and you get like 10," and this, the... And then suddenly at the end of it, the next increment is... giant increment isn't generating that much. (laughs)

Harry Stebbings 43:49 What have been the biggest hiring mistakes you've made?

Nilan Peiris 43:52 I measure, uh, how many times do I say yes

Nilan Peiris 43:59 and it ends up not working out, or how many times do I say no? So generally, I, I rarely am, uh, hiring into my reports. I'll do a final round for a, a leader into one of my teams. So I'll say no, but I'll always tell my team, "You can, you can ignore my no." (laughs) "But (laughs) this is what I expect that person to have done in six months time, and you will... If that doesn't work out, you'll need to let them go." And, uh, sometimes they, they say, "Yeah, this person can totally do that," and I'm proved wrong. So on both accounts, I get it wrong about 20 to 30% of the time. And so what that told me, (laughs) and this is after 10 years of trying, and I don't... Having talked to others and watched others on this, I've learned that interviews aren't a substitute for the real thing, and you just have to hire quickly, try and support people as much as you can. And just sometimes it just won't work out.

Harry Stebbings 44:52 Do you know what I love to do? I love to find a way to get them in beforehand, even for a day.

Nilan Peiris 44:57 Yeah. Tha-

Harry Stebbings 44:57 Just in a day, I can figure out, especially on content, it's a really quick feedback loop.

Nilan Peiris 45:02 Yeah. Yeah, yeah. Yeah.

Harry Stebbings 45:02 If you're a good video editor.

Nilan Peiris 45:03 Yeah. That...

Harry Stebbings 45:04 I know how you work.

Nilan Peiris 45:05 That can work.

Harry Stebbings 45:06 Yeah.

Nilan Peiris 45:06 But yeah, back to my interview questions. Love your feedback on this. So the two ones I ask are obviously a product one of, um, talk me through an example of where you've been hands-on in building something. And then I always explain, when I say hands-on, I mean you ended up having to code it, (laughs) you had to design it, you had to write the copy. But it is that, like, where do you go the extra mile and which extra mile do you choose to go? And then it, it's quite quickly you find the people who've never done that, right? And always worked through someone, and also just don't sweat the details. So that's... At all levels we ask this question, even if you're coming in as effectively a VP of product for one of our some areas. And then the, the second one is, everyone has a reason for leaving where they're, where they are. And they're reasonably open about it, um, and they're joining Wise or wherever because they think that reason doesn't exist in the place they're going to. Right? Does that make sense?

Harry Stebbings 46:05 Yeah.

Nilan Peiris 46:05 (inhales) So I, uh, try to get this out of them, and the way I get that usually, uh, is by saying, "Talk, talk me through what really frustrates you about where you are." And then you kind of get into, "Where does that really frustrate you?" And then it'll always come down to the same things, like, "My boss isn't behaving rationally," right? (laughs) "And they aren't optimizing for revenue. They're optimizing for X or they're being too short term," or this, or there's some toxic XYZ in there that I can't... And then the question is like, "Well, why, why couldn't you fix this?" Right? "So what, what was the hard barrier that hit, you hit that meant..." 'Cause if you're joining Wise, you're gonna find these types of things. You need to be able to work through them, right?

Harry Stebbings 46:44 Sure.

Nilan Peiris 46:44 "So why, why couldn't you fix this?" And in that, you find two things. One is whether they're able to kind of self-reflect and say, "Actually, I understand why the CEO or the founder was behaving that way, and if I was in that role, I would have done that." And that's, uh, usually a big tick. Or sometimes they're like, "Nope, don't, don't get it." But ultimately, through this conversation, they're already talking about their limit, because they hit their limit somewhere, and you can kind of then evaluate, will they run into that every day, every week, every month, or every year at Wise, and therefore, would it work?

Harry Stebbings 47:18 How quickly do you know when you've made a bad hire?

Nilan Peiris 47:21 Constantly impressed with people's ability to learn and develop, um, and I've learned not to trust my judgment on this, uh, if that makes sense as well, 'cause I'm constantly... There are people that I, there are people that I said no to that have done amazingly, amazingly well, and I learn from every day. And so I've learned not to... Yeah, I don't believe that I uniquely, in an hour-and-a-half interview... (laughs)

Harry Stebbings 47:45 When you reflect on those, what did you get wrong about that person estimation?

Nilan Peiris 47:48 I think at the end of the day, we have our own preconceived ideas of how we solve problems. Like, you can see I'm very opinionated on how I solve problems, and there are that many other ways in which they can be solved. And all I really care about is the problem being solved, right? (laughs) And the outcome happening, I don't really care how it, how it happens. And, uh, sometimes the ways in which people work, I would look at it and go, "I would never, ever do it that way," and therefore I think you're gonna fail, but sometimes they, they succeed and you learn something from it.

Harry Stebbings 48:16 We're gonna do a quick fire, okay? So I'm gonna say a short statement, you're gonna give me your immediate thoughts.

Nilan Peiris 48:21 Okay, go for it.

Harry Stebbings 48:21 So what's the most common, expensive, deadly

Harry Stebbings 48:25 mistake you see founders making?

Nilan Peiris 48:27 Marketing. So thinking that you can just grow your startup, uh-... by doing some magic on Facebook. (laughs)

Harry Stebbings 48:34 I'm, I'm so with you, or thinking that hiring someone for growth-

Nilan Peiris 48:37 Oh, yeah. (laughs) Got it.

Harry Stebbings 48:37 ... will suddenly find product/market fit.

Nilan Peiris 48:39 Yeah. Yeah, yeah. Got it. (laughs)

Harry Stebbings 48:39 What's the most dangerous myth floating around specifically about startup growth, do you think?

Nilan Peiris 48:44 Marketing. (laughs)

Harry Stebbings 48:45 (laughs)

Nilan Peiris 48:45 So I think it's the same thing around this, um, that there is some magical playbook. The bit I keep telling founders is, every minute you put into product/market fit or understanding what is it gonna take to make a great product has such a huge return versus, "I'm going to find somebody who's gonna solve the Facebook algorithm." Like, this isn't really what you should be spending your time on for the first three years.

Harry Stebbings 49:09 I think it's also, like, there's so much focus on, like, channel diversification.

Nilan Peiris 49:12 Mm-hmm.

Harry Stebbings 49:12 And like, it is so hard to get a channel working efficiently.

Nilan Peiris 49:15 Yeah.

Harry Stebbings 49:15 I remember Kit Bodnar at, uh, HubSpot said, you know, takes one channel working well to get to 50 million an year, or two to get to 100. And I think a lot of founders I meet kind of really focus on diversification before they've hit any meaningful revenue.

Nilan Peiris 49:27 Totally.

Harry Stebbings 49:28 It's like, fuck, if you get it working, double down. How does growth change in a world of AI? Or how does growth and product change in a world of AI?

Nilan Peiris 49:34 For me at Wise, it is, how does it make payments faster? How does it make payments cheaper? So for us, it's kind of obvious, like, we have, uh, big operational teams. We already use LLMs to augment those teams. Some places, LLMs, uh, automate 100% some of those tasks. Uh, with regards to content, like, you, uh, I mean, you, you, you've seen what people can do with content with these- (laughs) with these wonderful tools. We're obviously experimenting with them as well. I think that's where the exciting growth-

Harry Stebbings 50:02 What growth tactics stayed the same?

Nilan Peiris 50:04 ... I still think there's this role for creativity. So you can run an- you can run two ad campaigns. You can, uh, run them with the same growth tactics, uh, same, uh, channels, uh, same CPAs, same targeting, but one has a creative that is, like, got incredible cut-through, incredibly talkable, incredibly viral, and one doesn't. And this formula, it's not formulaic yet, and there's still this, uh, opportunity for creativity that will always, always have an outsized return.

Harry Stebbings 50:38 What have you changed your mind on in the last 12 months?

Nilan Peiris 50:41 Uh, I think I'm beginning to believe the hype behind LLMs in places now, and seeing them being practically useful at scale.

Harry Stebbings 50:49 Where are you seeing them practically useful at scale that you didn't maybe believe before?

Nilan Peiris 50:53 So you obviously play around with them. You kind of can find the edges of where they start hallucinating, (laughs) and they're less, less useful. And you have to engineer like hell to get them there. Uh, where w- I found it useful is when they're augmenting a person, if that makes sense. And then that, that- and there's still somebody driving a little bit, uh, haven't yet, but you can kind of see how it gets to the point that that, that human, that person will also not be needed in a while as well. So, um, yeah, that's the bit I'm ... On LLMs and AI in general, I think, um, I think we're all (laughs) getting it wrong in terms of where that- where it will end up.

Harry Stebbings 51:27 Tell me, final one, what was the most recent product or growth strategy, not including OpenAI, uh, that you've been most impressed by?

Nilan Peiris 51:36 NewBank and JP Morgan, uh, for two different reasons. So NewBank, uh, very similar playbook to us, customer-led business, uh, incredibly high word of mouth, incredibly high MPS. Lo- when you, when you open their analyst investor presentations, it's how they brought down prices, how they made the payments faster, et cetera. So- but when I look at their culture, it's really, really strong, and we have a, a lot to learn from them. They're obviously ahead of us in terms of scale. Great business. And at the other end of the scale is JP Morgan. So when I look at the- I mean, there have been large banks offering cross-border services for years, uh, biggest, like, Citi, HSBC, Barclays to some extent, NatWest, JP Morgan. JP Morgan have a pretty good infrastructure. They built it out really well. They continue to innovate at scale and at the s- size they're at, and I- you know, Jamie Dimon went pretty hard after payments. Payments have grown to be a material part of their business, and that is a real differentiator for them when they- in their corporate banking and their- they bank banks when they bank banks as well. So still very impressed, uh, and inspired by JP Morgan.

Harry Stebbings 52:42 I love that, and, uh, I'm also pleased with those suggestions. I haven't had either of them before on the show.

Nilan Peiris 52:46 No.

Harry Stebbings 52:46 So that's a good one. Thank you so much for putting up with my prying questions.

Nilan Peiris 52:50 Mm-hmm.

Harry Stebbings 52:50 You've been fantastic.

Nilan Peiris 52:51 No worries. Uh, thanks for having me.