The Future of FinTech
What does the financial technology of tomorrow look like? We get the inside story from the bosses of Stripe and TransferWise
Second Home’s Rohan Silva meets 26-year-old Stripe co-founder John Collison and Taavet Hinrikus, CEO of international money transfer platform TransferWise, to discuss the future of FinTech and London’s role in shaping the global internet economy.
Rohan Silva: Why did the two of you choose to do the things you do, what is the mission that underpins your companies?
John Collison: One of my favourite things about Silicon Valley, and to some extent, tech, is I feel like people can go back and rewrite history and they talk about what evolved to be their mission, and after being really close to the space over many years, they pretend that’s what they started out with to begin with.
So once they’ve successfully managed to build a company and communication, they’re like, ‘We set out to redefine communication’, it’s like, is that really what you set out to do? [laughs]. I think it’s very natural and we shouldn’t shy away from the fact that you start out kind of poking at a space, and it seems like there’s something here, and you’re kind of discovering the edges, and you grow into a mission over time.
So in our case, we started out wanting to make it easier to accept payments online. This was a problem we arrived at because we just felt it first-hand — we had built start-ups previously, we had built iPhone apps, stuff like this, and it was really striking to us that even though the barriers to building online products were continuing to tumble — it’s like that New Yorker cartoon back in the day on the internet, ‘No one knows you’re a dog’, you know, these two random kids in Ireland could make products that people all over the world found useful, and yet going and turning those products into a business was remarkably hard. In particular, it felt like all the infrastructure we were using — credit card processing and the merchant bank and things like this — the internet had not arrived, it hadn’t been updated since the 1970s. So that’s the problem we set out to solve.
I think what we found over time was that it wasn’t just the companies starting out, it wasn’t just the people in the very beginning, that were having issues, these issues were manifest everywhere. So we saw that as companies grew, everything they tried to do incrementally — whether it was starting a new line of business, or adapt their business for changing environments, or wanting to sell internationally — none of these were possible.
So now, loosely the way we talk about our mission, is to increase the GDP of the internet, which is a sort of pithy way to talk about providing the tools and providing the infrastructure that make it easier for people to trade online, that make it more effective and more efficient, and hopefully we can have knock-on effects on internet commerce as a whole as a result of that.
One of my favourite things is how you have to sort of pretend that you have that insight at the beginning as opposed to being able to develop it.
Rohan Silva: Yeah, I’m all about the post-hoc rationalisation of what you’re going to do anyway [laughs]
Taavet Hinrikus: I like how John was putting it — talking from a personal angle — and I think a personal angle is always a true one, and if you find a company which has a personal touch then it’s a much better story than one of post-rationalisation after three pivots. For me, I was lucky enough to meet Niklas Zennström and Janus Friis when they had started and sold Kazaa and were looking at next things to do. I had a choice of working with them or going to work for an IT company in Estonia, and I was smart enough to choose Niklas and Janus.
So I had, I call it, the pleasure, luck and the torture of working with them and building what became known as Skype and spending about seven years of my life doing that. That was a pretty profound experience, really seeing how sitting in a post-Soviet building on the outskirts of Tallinn, you can build something which really changes the way we all communicate.
“I had the pleasure, luck and the torture of building what became known as Skype and spending about seven years of my life doing that. That was a pretty profound experience, sitting in a post-Soviet building on the outskirts of Tallinn building something that really changed the way we all communicate.” Taavet Hinrikus
It’s kind of hard to get away from this later on — even so, now in retrospect Skype should’ve done many more things, but I do believe that Skype changed the way we communicate. Maybe we would’ve gotten the same point through some other company or other people, but I think we kind of pushed what was thought to be possible. We made it possible to use a computer for making phone calls and later on for high-definition videos.
I left Skype at one point, figuring enough is enough — I was 27 years old and I figured there’s a whole world out there. But after such an experience, it’s pretty hard to go and get a job. You go into work from 8–5 for what? To come home with a salary at the end of the month? It really feels pretty dull. So I’d say I got a little bit inflicted — I got the virus of doing something which has an impact.
So I figured what I want to do next is I want to have an impact — it could be a social project, to do something which has an impact, which makes the world a tiny bit better. So when we started playing around with turning the way we’d been exchanging money between ourselves into an idea of how it bothered people, it kind of rings a bell. You could see it can have an impact on the world. If we can make a call from London to Singapore or London to Sydney without it costing a dime and being much better quality that what [we] previously thought was possible, then why still can’t we send money around the world?
So if with Skype we solved the problem of communication, I want to make it as easy to move money around the world. But obviously at first we knew we can help ourselves move our money and then we thought we can help many other people. So definitely the vision has expanded, but it starts from the personal problem and figuring out that I’d like to do something which is slightly more meaningful than having a job and going to an office.
Rohan Silva: John, Taavet talked about Skype and that being a formative experience, you started a company — unbelievably, given how young you are, but successfully started a company — before Stripe, tell people here about that experience and what that company did.
John Collison: It was a year, which is kind of compressed as far as start-up experiences go because a lot happens in a year, so I’ll try to compress it down to the essence for folks here.
We had noticed that buying and selling online was pretty hard — in particular, eBay didn’t really solve the problems for a lot of regular people buying and selling — so we set out to build a better kind of peer-to-peer trading marketplace.
We ended up pivoting that shortly afterwards and did seller tools for pro-sellers selling online, because we discovered that it’s hard to get a selling marketplace, solve the chicken and egg problem, so we started solving one side of the chicken and egg problem. And then reasonably soon into the life of the company it was acquired by a Canadian company where some of the team went to work for the acquirers.
There’s obviously lots of useful lessons that you draw from the experience, I think, in particular for us what we took away from it was 1) Having seen the problem in payments and getting to that next start-up idea, but 2) I think a lot of people are — certainly Silicon Valley and in technology cultures — transfixed by the exit, and even the name kind of sounds funny in that it suggests it’s the end of the quarter.
Having kind of been through that and seen what it does for the arc of the product — and in the case of our product it didn’t continue — it instils in you, and I know a lot of other people who have been through similar, people who have been through an acquisition tend to think about it quite differently the second time. Acquisitions can work, and they can work quite well, but I think there are lots of examples of acquisitions that don’t and there’s a lot that goes into making them work.
“Acquisitions can work, and they can work quite well, but I think there are lots of examples of acquisitions that don’t and there’s a lot that goes into making them work.” John Collison
So I think both for Stripe, it motivated us to 1) Be very focused on building a long-term independent business, but 2) Kind of getting ourselves to the position where we could have that luxury in constructing Stripe in such a way where it wasn’t too much of this ‘figure it out as you go’ where we could have that choice. I think the other thing it educates you on is, as an acquirer, we haven’t made any big acquisitions — we’ve made one or two small acquisitions — I think it starts to teach you a little bit as an acquirer how to think about successful acquisitions.
Rohan Silva: Obviously, given that you’re both working in the same broad field, there’s real similarities in terms of this not being your first business, and a common mission, but there’s real differences as well. One difference that’s pretty obvious is that John, you chose to move to the Valley to build your companies — at least build Stripe — but Taavet, you could’ve easily, after Skype, gone to California, why did you choose to stay here, what was the thinking?
Taavet Hinrikus: There are a lot of layers to this — you could go to California and be a small fish in a big pond — I’d say for me to go to California and start recruiting people there and building a team, I have no advantage to someone who’s been there, it’s much more competitive. There’s more risk, there’s more talent, but every schoolkid you hire to be an iOS developer asks for [£500,000] of salary and so on.
But also it comes down to thinking, ‘What’s the problem you’re solving and who are the customers?’. I’ve lived in the US during high school, I go to the US a lot, but I somehow like Europe more as a place to live, I like Europe more as a place to raise my children.
So kind of minor reasons, but when starting TransferWise we had the choice of choosing where to do it, but we saw it as the problem is equally big in Europe— that was way before the UK became the FinTech capital, and to be honest when starting TransferWise I didn’t know what FinTech meant [laughs] — but the UK is a great place to be [the] headquarters for a global company. From here I can hop on a plane and be in Singapore in 10 hours, or California in 10 hours — anyone who’s tried flying from California to Singapore, it’s very painful.
The funny thing is, when we were raising our A round, we accidentally bumped into Andreessen Horowitz. It was a time when they were publicly known as they don’t invest in companies outside of Silicon Valley, I think they had stumbled on a couple of companies in New York.
So we met them and it was a pretty good meeting. It was set up at very short notice, and I thought we were going to go in and meet whoever is left in the office at 6pm on Monday. But we had the whole partnership there, and we told [our] story and they quite liked it. Larry Summers was special advisor, so they call Larry and Larry says, ‘Yes, it’s all cool’, and then they said, ‘Guys, we’ll write you a nice cheque, but there’s one condition: you’ve got to move to California’. We were like, ‘Sorry guys, it doesn’t make any sense for us’. We were optimistic that we would also find the money not having to move to California, which we did.
Rohan Silva: John, London’s tech scene is interesting in that we are at times very confident, and other times you have to really look to the States. But when it comes to FinTech, we think of this as an epicentre. Do you think you could’ve built Stripe in London?
John Collison: In general I think this whole narrative of ‘you have to start a technology company in Silicon Valley’ is hogwash, and people who believe it sure have a lot of proofs by existence to explain — whether it’s Skype in a former Soviet building on the outskirts of Tallinn, or Spotify, all these huge successes keep on cropping up, so we don’t worry too much about that.
We see it in the Stripe data, if we didn’t believe that successful companies weren’t going to come from all over the world, then we’d not be investing as much as we are in lining up that infrastructure. That said, three considerations for us were, 1) We were in the US anyway to begin with. Patrick, my co-founder, and I were in college in the US, and so it wasn’t so much moving there as being there already.
2) For Stripe in particular, I think being close to your early customers is quite valuable. It continues to be valuable in any stage of the business, but in particular the early stages, so Europe has a very big remittance market, and so it makes total sense for us, given we’re selling to technology companies, there was a real critical mass there that we would go down and sit next to and watch them integrate Stripe.
The feedback initially was very tight. I think you see this all over, where the tech companies serving fashion — a lot of them are in New York — that dynamic makes sense to me of being close to where your customers are, because that feedback cycle is pretty important. Certainly I think being off-time zone from your companies would be quite hard.
The third thing that I think is quite cool that you get in Silicon Valley, that you don’t get as much in other places, is — it’s quite unnatural in a way — Stripe is now 320 people and we were two people five years ago. We doubled headcount in the past year, we’re going to hopefully double headcount if we can in the next year, and so you’re doing this unnatural thing where you’re kind of accumulating this team to do this highly specialised thing much faster technology most companies grow.
“The particular advantage you have in Silicon Valley that a lot of people don’t call out, is that you’re not just hiring those individual people, but you’re hiring them and all the experience that they have.” John Collison
The particular advantage you have in Silicon Valley that a lot of people don’t call out, is that you’re not just hiring those individual people, but you’re hiring them and all the experience that they have. So as we go to build our business development team, if we’re building a business development team that’s going to be dealing with other technology companies, we can hire people who’ve been doing business development for 10 years and have been dealing with technology companies, same with all the different rules and specialities.
There is quite a critical mass in Silicon Valley of people who have that accumulated set of specialised skills and that is very valuable. Silicon Valley is by no means the only place that has that, but it’s a huge kind of [18.23] market in that regard.
So I think those are the factors that go into Stripe in particular, but those don’t necessarily generalise to all companies. So your company [18.33] who do you want to hire? What rate do you want to hire at? Who are your potential customers? All these kind of things.
Taavet Hinrikus: Customers are absolutely mobile, but you have to be close to them. That was the reason we decided to stay in London, because we had early attraction here, we figured if we move to the US it will be a complete waste. I also agree on talent, especially when it comes to people who know what they’re doing, there aren’t that many in Europe [laughs]. Especially if you go into more kind of leadership roles, that’s really hard to fill.
But it’s a function of — there was a number of successful tech companies in Europe — it’s purely a numbers game. If you haven’t had companies that have grown up in Europe, you just can’t have these people. If you look around for developers, for designers, even for product managers, I actually think there is a decent amount of them here, and it might be easier to hire them here — maybe pay them less, but they also won’t jump the ship and two weeks later if someone pays them a tiny bit more.
I’ve yet to hear anyone say hiring is easy where I’m based. I don’t think, especially if you’re European, moving to California would make hiring easier for you, it would be much more expensive and these people will ditch you much quicker. So I think Europe is actually in a better position, except the leadership talent.
Rohan Silva: Another difference, maybe, between the two of you, is maybe the approach to talk to and about the banks. Taavet, you’ve got billboards across London with, ‘That moment when you realise your bank is ripping you off’
Taavet Hinrikus: But it’s true [laughs] I don’t like lying.
John, you work a lot with banks, with Visa etc, you’re working with the banks to kind of make them better, is that fair? You’re not kind of taking them on as such?
John Collison: We kind of are and we aren’t. I think the general arc of this has been a lot of payments, finance, baking, financial infrastructure, is starting to become about software. Most banks — and a lot of companies even outside of banking, but in particular in banking — have not adapted quite well to this change.
I think what happened was that as the internet emerged, now the primary way in which your customers wanted to reach you — and people were not actually reaching their customers for a long time — was online and over the internet. Two things happened: 1) People didn’t adapt particularly well to this world and seeing customers complain, ‘Why can’t I use online banking? Why can’t I get a decent mobile app?’, and five years after the introduction to the iPhone maybe people introduce an app and it’s out of date by — and the buttons are all low-resolution and stuff. It was very clearly from a different culture.
But the second thing that happened was that benefit of banks having a monopoly on all the nice buildings all across the land, that started to become less important. So the natural bundling of all these different services became less valuable.
So I think what’s happened — and rightly in a pro-consumer manner — is that now, all these financial services people are saying, ‘How can we do better loans? How can we do, in Stripe’s case, payments for businesses? How can we do better retail payments? How can we do better foreign exchange?’, the examples just keep coming out. And you don’t have to buy these services from the same bank in the same way that you did in the 1970s.
“How can we do better retail payments? How can we do better foreign exchange?’, the examples just keep coming out. And you don’t have to buy these services from the same bank in the same way that you did in the 1970s.” John Collison
So we’ve seen this giant shift — and it’s not at all over, I think it’s actually just starting, there’s still plenty of opportunities for interesting companies to be started as a result of this trend — which is that a lot of financial services as becoming kind of software services and are becoming about software.
In Stripe’s case, if you want to accept payments on behalf of businesses, if you want to let people run an online business, you can’t do it without working within the existing banking system because that’s just what exists and that’s what people have.
So in Stripe’s case what that means in Europe is providing a really easy way for people to accept credit cards and direct debit payments, and that’s what we do. We’re also playing with Bitcoin stuff, but let’s be real, the consumer adoption of that is still extremely low. So we’re, in a way, relatively agnostic to the infrastructure in which things run — we would like to see it modernised and get better still.
I find it so annoying that in the US it takes three days for a bank transfer to happen because they’re still running the same mainframe computer from the 1960s. But, to be useful to people, you need to work with what the consumers are coming to you with. So I think that’s why we’re in the positions that we’re in, where previously you had to go to a Bank of America or Chase or someone to be able to accept online payments. Now that you can go to Stripe to get that, but there will still be lots of banking involved, and there will be for the foreseeable future.
Rohan Silva: Taavet, you want to blow these fuckers up, right? [Laughs]
Taavet Hinrikus: I don’t think there’s anything special about banks. What’s happening is very similar to what we did with Skype. Skype took one feature, which was long-distance calling, we made that better, we made it available for everyone for free. The same thing happened with music and Spotify, same thing happened in the publishing industry and so on.
After banking we’re seeing this happen in insurance, we’re seeing this happen in transportation, we will see it happen in education, so technology will disrupt all industries and it’ll make them immensely better than they were before. But when it comes to banks, I don’t have nothing against the banks, I still keep my money in a bank. If I hated banks I wouldn’t keep my money in them, right? But what I get furious about is companies lying to their customers.
“I don’t have nothing against the banks, I still keep my money in a bank. If I hated banks I wouldn’t keep my money in them, right? But what I get furious about is companies lying to their customers.” Taavet Hinrikus
So if a bank tells me — or go to the Post Office in London — and they tell you, ‘We’ll do your money transfer for free’, and then they actually charge me 5%, that kind of makes me angry, and that’s why I’m pretty happy to point out when banks are being unfair to their customers. So fairness and transparency is pretty important to me, and that’s what I scream out about. If there was an honest banker around, I’d have to hang out with the guy.
John Collison: That is something that I think has a real effect, which is, you see this in mature industries that don’t necessarily have a lot of competition or as much as they should, is this user-hostile behaviour, and I think hidden fees are one example of this. This is something we saw in payments before Stripe — you have no idea what you’re going to pay, and there’s all the fees, there’s just layers and layers of obfuscation and trying to deliberately have you misunderstand what you’re going to pay.
It’s fascinating that, when you think about it, the stuff that Stripe came out with when we launched, in a way it was so basic, like, ‘There is an internet website where you can sign up, get an account, and we will tell you what you’ll be charged’, and people are like, ‘Ah, that’s revolutionary!’.
Rohan Silva: You’re working with Visa, say, which is a great partnership, was there a fork in the road where you could’ve said, ‘You know what? There’s Visa and MasterCard, we’re going to try and replace or create competition for them and create a new kind of rail instead of either of those two?’
John Collison: I think we are building a lot of new infrastructure, but what do you have in your wallet? It’s probably either Visa or MasterCard, and similarly for most people out there. So I think you do see businesses taking principle stands and, ‘We’re going to do this all over bank transfers’ and stuff like that, but that’s not fundamentally what businesses want. So if I’m out there starting a start-up tomorrow, I want to be able to accept money from my customers in the thing they want to pay with, and our principle stand won’t help them actually achieve that.
Taavet, both of you are talking about disintermediation and this great change in the financial services sector, and it strikes me that whether it’s loans, foreign transfer, payments and so on, this great disruption happening. The core banking product itself has remained relatively resistant to disruption and technology. Do you think there’s an opportunity there, starting new banks using technology to transform that experience?
Taavet Hinrikus: What are banks still good for? As a consumer, you take loan, and if you look at lendings, there are lots of good, and maybe some less good, companies which have done stuff to lending. All the peer-to-peer lendings and there is Wonga and other forms of credit, various degrees of parameters around them, but around credits there is lots of new stuff come out which makes it more accessible, makes it cheaper and a combination of this.
Look at money transfer and clearly TransferWise has taken that one. You go on and look at wealth management, we have Nutmeg. Every form of banking that we do, there is a start-up or two that are doing it much better. We can argue now, is banking going to be split up completely? I think we’ll see various forms of this together.
“Look at money transfer and TransferWise has taken that one. You go on and look at wealth management, we have Nutmeg. Every form of banking that we do, there is a start-up or two that are doing it much better. We can argue now, is banking going to be split up completely?” Taavet Hinrikus
I think, going back to my Skype experience, Skype today, being a 12-year-old company, owns 40% of long-distance calling. I think similarly we’ll see the FinTech companies own 40% of financial services in 10 years’ time. That’s going to be the Stripes and TransferWises and probably the Googles and Facebooks of the world, and especially new companies who are doing something in a much better way similar to the way we’ve started our companies.
Banks are still good for one thing today, which is keeping your money. I don’t think anyone has cracked that angle to do it in a better way than banks do it today. There are things around — there’s the trust they’ve got by building lots of bricks on top of each other and building these pillars — there is deposit insurance and different reasons, but I think that’s what banks are still good for, and I’m not seeing anyone come up with anything better.
I’ve given up hope on Bitcoin a while ago, I don’t think that’s going to solve it, so I think we’re yet to see who can keep our money better than a bank can.
Why have you given up on Bitcoin?
Taavet Hinrikus: Because it’s not good for anything. If I want to give you a Bitcoin now, how should I do it? You can give me your Bitcoin ID — 37-odd characters… The problem with Bitcoin is that nobody’s figured out a good use case for paying with Bitcoin. The problem is, getting Bitcoin is hard, and getting rid of Bitcoin — turning it back into real currency — is hard. Moving Bitcoin is easy — if everyone in this room had a Bitcoin wallet, life would be wonderful. It seems to me that people have been buying Bitcoins because they think it’s an asset which will only go up, like most assets in the world, but I haven’t seen this practical stuff you can do with Bitcoin. I’d love to see it, but I’m kind of starting to give up on that.
John Collison: All technologies which are some kind of network or two-sided, multi-sided networks, need that killer app to get them going. In the case of Bitcoin we’re still searching for it. I also think that in a way, success is boring — when something is getting off the ground it’s a movement, when they talk about with the Apple I, with Steve Jobs and Steve Wozniak bringing it to the Homebrew Computer Club. Because back then there were clubs for enthusiasts with computers and talking about them. What makes the iPhone possible is all the underlying components like container shipping, global supply chains that are well-managed and things like this. When you see an iPhone you don’t go, ‘Ah some fine supply chain’.
Taavet Hinrikus: Supply chain doesn’t make it possible
John Collison:: It does
Taavet Hinrikus: I disagree; they tweak a supply chain to make more money
John Collison: My point with Bitcoin is, in a way Bitcoin could find a use as a better underlying protocol for lots of financial, or non-financial, transactions. I think that is quite real, but in a way, that may lead to fewer Bitcoin enthusiast because it just becomes a distributed consensus protocol that’s really good.
We found on the current accounts thing, I find this a really interesting question, because I think so clearly current accounts could be so much better, and they haven’t yet emerged as such. I think it comes back to partly what I was saying, which is the bundling of financial services that used to take place.
So when I was a teenager growing up in Ireland, the local banks would come by and they would try to get you to open a bank account and ensnare the students for that particular bank — not because there’s such great margins to be made in teenagers with their pocket money and the current account, it’s the value of getting that lifetime customer, that’s the really profitable stuff.
It’s well-known in banking that current accounts are just not that profitable — you’re servicing these customers and providing customer support, and earning, especially these days, fairly minimal interest. So the question is, in an unbundled world, current accounts are a little bit the red-headed stepchild because they’re one of the less profitable parts, and traditionally they’ve just driven the adoption for all the other financial products, it’s an opportunity to upsell people on loans.
“Current accounts are a little bit the red-headed stepchild because they’re one of the less profitable parts, and traditionally they’ve just driven the adoption for all the other financial products, it’s an opportunity to upsell people on loans” John Collison
So I don’t know how we’ll solve this — you obviously have this start-up, Simple, in the United States, that set out to reinvent current accounts, that I don’t think really did.
It didn’t own the underlying banking infrastructure
John Collison:: That was a big part of what made it hard to make a really good customer experience, but I think this is obviously still a clear — you go out and measure the VNPS scores of people and their current accounts, and I think it’s clearly a problem that still needs solving.
Royal Bank of Scotland in this country has a VNPS score of -6
Taavet Hinrikus: Most banks have a NPS score which is negative. I’d love to see banks that have a positive NPS score.
I think it’s also, in some ways, I’ve yet to see someone who makes a current account so much better — anyone that makes current account five times better than it is today. It’s still, ‘you keep my money, you give me a card to access it, yes, my mobile app might have a few more bells and whistles’, but I think that’s part of the challenge — it’s hard to make it so much better. To get people to pay for it if they’re used to having a subsidised current account, is a kind of challenge.
Simple in the US was great, but let’s be honest, it will fail. They didn’t start a business and sell it for £10m to a bank, so in that sense it failed. Now, being owned by a bank, it’s doing super well. Did it need the bank to do well? No, it can develop.
We’re seeing there’s a lot of kind of mobile current account start-ups being announced around the world. I think, sadly, the vast majority of them will fail as well because they aren’t going to have a product which is vastly better than what banks offer, and they won’t be able to acquire customers anyway because there’s no — there are also no great social loops. It’s a hard problem, I don’t know how it’s going to get solved.
Eventually, as a techno-optimist, I believe it’ll be solved, somebody will crack it, I believe… but most won’t.
Both of you have mentioned that there are really interesting opportunities for new companies in the FinTech space. So if your companies were bought for £200bn by Apple and you were persuaded somehow to let go of your babies, in this FinTech space what opportunities do you think are really still there for all these hungry young people in the room to have a crack at?
John Collison: I think definitely the most interesting problem I see right now is consumer banking. Business banking is another one that’s relatively unsolved in a bunch of adjacent problems like payroll and all this kind of stuff, but I think consumer banking is so fascinating because literally everyone in this room has problems that they encounter with their bank.
Taavet Hinrikus: I’m kind of blindfolded by doing what we’re doing, in FinTech… Hard to say. With TransferWise we have 4% market share in the UK, which is a pretty cool starting point, but we’re far from being content with what we’ve done in the UK, and 0% globally. If there’s anybody here starting a TransferWise in Mongolia or Brazil, do it and be quicker than we can be, because we’re coming to get you but it’ll take us time [laughs].
The bigger opportunities, I think there’s a lot of other industries which I’m personally hoping will get disrupted. There’s so much in health — MedTech or whatever — but if you think about merging the interactive things and predictive health economy, I think the world is going to look very different. Similarly [with] education. Transportation I think is probably starting to be done at this point, I don’t imagine our children will need a driving licence because there are plenty of self-driving cars being invented. It might be a bit late for drones or self-driving cars, but education and health there’s still a lot to do. In FinTech, insurance, I think, is the next one.
“It might be a bit late for drones or self-driving cars, but education and health there’s still a lot to do. In FinTech, insurance, I think, is the next one.” Taavet Hinrikus
John Collison: It’s quite exciting how it feels like the level of execution and quality that you need to lead as a company, is continuing to go up, I would argue. A lot of the very old incumbents being replaced, the quality of service is quite poor and they’ve been around for a long time. Whereas if you look at some recent companies — Nokia was making very good phones up to kind of 2007 when everyone’s perception of what a very good phone was kind of shifted more or less overnight, and since then times have been really, really tough for them.
I think, similarly, there’s no reason to believe that what we have today in consumer devices, payments, or anything like that, is necessarily the apex. Similarly, there are so many examples of technology companies that were the leading and bright star of their day.
So I think what is also is exciting is that Stripe, as we look 1–5 years out, we absolutely have to earn it, and there’s plenty of opportunities for other companies to come along, that if we’re not the best way to start a business online, someone else will come along who will [be]. It’s very good and healthy that it works that way, but we’re under no illusions that that’s how it works.
John are you investing in machine-learning?
John Collison: We do have a lot of data at Stripe, and so we have a data team and a machine-learning team. So one obvious use case for us is fraud and combating fraud, money laundering and stuff like this. It is something that I think — it sounds very internal and operational, whereas in fact you can use [it] to make the product better.
So if you look at online payments, the reason it was so bad for so long, was partly because you needed to sign up and fill out all this paperwork, then it got sent off to a human who looked at it, stamped it and approved it after everything was good — and they weren’t even doing that good a job at analysing the data, because I think we’ve found that humans make fairly error-prone judgements. So it was a bad experience for the consumer and it didn’t particularly accomplish the industry’s goals.So in our case, part of what lets us make the sign-up for Stripe quick and easy, is the fact that we’ve now gotten fairly precise about the data collect and how we use it to determine that you are in fact a legitimate user of the service. So it’s only through being able to rapidly sift apart the legitimate and illegitimate users of Stripe, that we can offer a really good on-boarding experience.
We don’t talk publicly about all the stuff we do, but it’s quite an interesting challenge: can you tell when someone provides you a few basic details about their business, can you tell whether or not it’s a legitimate business? When you start to think about it, you’re like, ‘Okay, legitimate businesses tend to be unique, whereas scams often go and copy a bunch of text from someone else’s website’, so by checking the website for uniqueness and using the plagiarism detectors they have in university, that actually ends up being a pretty interesting signal.
One of the pretty interesting ones for us is making better product experiences by kind of taking things that were human and manual and slow, and making them instant.
I don’t quite understand all [of artificial intelligence], I’ve been reading a few papers recently, but all the recent stuff in deep learning and all the progress we’re making there, is kind of mind-blowing and I’m still trying to piece it all together.