Ep # 164: Fraud, Signals, And Spend: How Modern Risk Engines Really Work ft. Robert Kraal (CEO & Co-Founder, Silverflow)

Robert Kraal 00:03 In the after-sales process, we have a migration and a setup phase, where we do all the end-to-end testing, making sure that the product works exactly as our customer asks us to deliver it, and once they are live, the most discussions are on the one side about how to ramp up the volume. Taylor Kennerson 00:20 Welcome to the Hyperengage Podcast. We are so happy to have you along our journey. Here, we uncover bits of knowledge from some of the greatest minds in tech. We unearth the hows, whys, and whats that drive the tech of today. Welcome to the movement. Adil Saleh 00:39 Hey, greetings, everyone, from everybody, from North America, from Australia, Canada, UK, Europe, all over Europe. Thank you very much for keep on listening to this podcast for a very long time. It's about 40 episodes, more than 40 episodes we have done this year. And previous year, almost equal or more than the preceding year, like altogether, we've done more than 150 episodes. It's good to talk about different industry problems, and it's good to be a part of this, one of the biggest shifts in technology in the modern era, at least in my lifetime. But with the guests that I have today, it might be different. So today, we're going to be talking about how FinTech industry is evolving, how it has a greater degree of impact that nobody talks about of AI and automation and scale and collecting payments at scale and throttle. This was one of the biggest thing was cyber security and all those components to specifically in the European regions and then its implication towards the North American market. So today we are with Robert Kraal, the CEO of Silverflow, one of the leading payment integration platform out of the Netherlands. Thank you very much, Robert, for taking the time. Robert Kraal 01:55 Thank you for inviting me. Great to be here on your podcast. Love it. Adil Saleh 02:00 Perfect. So I know that, Robert, it is not an easy thing to have it as an inception 15 years back. I know that you've been doing it for seven, eight years now, but somebody working in the same space as an executive, senior executive, it's so hard to think of it. Hey, I'm going to be in the payment platform because it's not easy to be very honest. You can simply, like a lot of these founders said, yeah, I'm going to do something with the AI. You cannot do that for a payment integration platform. So tell us a little bit about your background, what kind of thought process you had, what sort of drove you at that time to make something like, hey, I want to build something and I want to solve this problem and all of this. So that'd be so interesting. Robert Kraal 02:43 Thank you. It's a long story. It goes back quite a while. I started my career in payments in the late 90s at a company that at that moment not even was a payment company, a company called Bibit. And we were with the four of us there and it was at the early stage of internet, definitely early stage of internet payments. And we had a different product which didn't work at all, but we had a little payment module connected to that product. And we had a few customers asking, can we use that particular payment model? We don't want your product itself, just the payment module. And that is sort of where we built out the first payment service provider, which was called Bibit. We sold Bibit in 2004 to Royal Bank of Scotland and RBS at that moment also acquired another company called Worldpay. They merged the two together, used bits and pieces of software from both these companies, but they preferred the brand name Worldpay. And to a degree, that is still the Worldpay that you're there today. And so that was my first encounter within the payment space, really in the early days. Then I joined Google to launch their payment product, Continental Europe. That product essentially was not ready for the market at that moment. It was too early, too much card focused, too much Visa, MasterCard, American Express, too little on all the different domestic payment methods, multi-currency in the market. So I wanted to do something which was more like what I used to. So I became the CEO of a company called Docdata Payments. And besides CEO, I was also there responsible for all the financial partnerships, would be with the acquirer, the card acquirers, the alternative payment methods, but also with Central Bank, Treasury Bank, stuff like that. Did it for a couple of years. And then my old buddies from Bibit said, Ok, we're starting something new. Would you like to join Adyen? So I also joined Adyen in a similar role, so a combination of CEO and responsible for the financial partnerships. And in that time, the regulation changed a little bit in Europe, Payment Service Directive. So we're talking about 2009. And I said to the team, I said, listen, guys, I think on the base of this particular new regulation, we can become an acquirer ourselves. So we want to get our own licenses for Visa, for MasterCard and some other cards instead of working with the banks who traditionally hold these licenses. This was a bit a new process for both the regulators as well as for Visa and MasterCard. They didn't anticipate that based on this new regulation. But we managed to get through with with the help of a lot of people, obviously. And at that moment, we were sort of thinking of how shall we technically connect to the card networks? And there were quite a few players. And since we and the team and we personally had quite some experience with all the different solutions that the banks at that moment were using, we also could see the differences between those different platforms. And we benchmarked all of them and said, OK, listen, we like this feature of that platform, but it's always a little bit late. We like that reporting. We like that. We don't like that. But essentially, we found out that none of the platforms did what we wanted. And that's the reason why we decided to build it ourselves. And that has been become at Adyen a little bit my later stage mission of setting up entities across the globe, getting the acquirer license across the globe, making sure that from a technology point of view, it would work and also the operations around it. So the onboarding. So fast forward to Silverflow. Well, the problem that Adyen faced, we don't have a good acquirer processor and never really was solved. So if today, if you're becoming an acquirer, you still go through the same process of talking to Visa, to MasterCard, to Regulator. And at some stage, you will know that you get the license. You still have to tackle the technology question, essentially. And previously, there was two options. Option one, use legacy platforms. Option two, build it yourself. But both are not easy solutions. To build it yourself is complex, takes a lot of time. And actually, it really is quite complex. And to go with legacy technology is also not really what you want as a fancy next generation payment processor. You want to work with the new technology. And we thought that there was a gap in the markets. And that's what we built. And that is what Silverflow is today. Adil Saleh 06:43 Interesting. I love the way that how you understood the times when industry was not at that point, technology wise, and you learn and took the leap and made sure when the time is right, you definitely break the ice. Now, if you were to launch it in North America, is that going to be similar sort of model, business model, or you were you were ready to do it starting out in North America, or it was just like thinking about Europe at that time and how we can expand there? I know you're also a board member of Volt. One is one of the finest there in Europe as well. So tell me a little bit more about your mindset as a leader at that point of time, thinking about the North American market or just completely leading it and thinking, hey, I just want to focus on Europe. That's where I have the most stronger roots. And I know the regulations and everything. And I've lived closer to the problem all my life. Robert Kraal 07:38 So when we designed the products and definitely in the early days, but of course, it's still ongoing. We said this should be a global product. So it should be available easily to be adopted in pretty much every market. And there are a few differences per market. Sometimes in the regulatory space, sometimes on the functionality space. But definitely when we started this company, we said, OK, should also function in North America. And it does. So we have customers in the U.S. We have transactions in the U.S. with live customers in the U.S. So, yeah, that is that's demonstrated. It's for us, it's still a growing area. So we are a little bit at the early stage of our journey, particularly in the U.S. But we've got quite a few nice names signed there. And the volume is growing rapidly as well. What typical differences are between, for instance, the U.S. and Europe is also the one side of regulation. So in Europe, you have this Payment Service Directive. And now we have PSD2 and PSD3 is being worked on. And essentially how I always explain it to people outside of the fintech world, a sort of bank lite license. So you're allowed to do certain financial products. And today, of course, it also includes some of the crypto stuff. But it's not a full banking license. And the difference between a full banking license and all these different payment licenses in Europe on the base of these PSD licenses, you can become an acquirer yourself. The U.S. doesn't really know that that splits. So you have your full federal banking licenses, but you have nothing comparable to PSD license. Well, maybe a money service business license could be two degree comparable, but it's less crystallized out what you need to do. And the result is that in the U.S. you very rarely see new acquirers popping up. Where in Europe you do see them and in many other countries across the globe, you do see them. So for us, that's a slight difference in how our go-to-market is. In Europe, our initial customers were young, innovative, tech focused, new acquirers. They just got their license and didn't want to work on all technology anymore. They want to do that. And that's an easy way to step into the market and to meet your initial customers. And as a result of having a number of these customers, we were also talking to and able to sign bigger banks, which is of course also for us quite interesting. In the U.S. you don't really have that pattern. So our first customer has to be a bank. It can be a small bank, can also be a big bank, but you don't have a new acquirer, a very tech savvy acquirer. And that makes it a go-to-market slightly different because typically banks are a little bit more risk averse. They have a migration process and set off a greenfield, what they can set up. The bigger the banks usually, the more internal controls. And that's not a problem per se, because we work with banks in other regions across the globe. But it would be probably easier if you would be able to showcase your capabilities domestically on the U.S. market with an easier startup customer instead of that your first customer has to be a bank. Nevertheless, it takes a little bit more time to do your proof pointing, but we're live there as well. So it can still work. Adil Saleh 10:37 It is, it is. I love the way that you're trying to make sure that this technology is evolving always, but it always stays global. And with new challenges coming from different, it's not just like here, like Australia is also a decent enough industry. And the UK is a decent enough industry when it comes to small business, small to mid market business altogether, more than 150 million inclusive of Australia, UK and the U.S. in the small to mid market segment, which is quite large. So now looking into, I'm sorry, I'm making you look into, not to market a lot, because I wanted to make it even more, I want people to more echo with a lot of this conversation. And I'm sure they already know about Silverflow and you're doing an incredible work. So now thinking about these small to mid size businesses, a few options that they have is PayPal, Square for subscription based payment collection. They use like these are not just banks. These are like platforms built with banks like Brex and Mercury has been one of the biggest here in the valley. Also, there's TransferWise, like these kind of platforms, but TransferWise is not for like subscription based payment collection. So what do you recommend? Like how do you relate it to those? And what these people, founders listening, how do you think like what kind of processes Silverflow can actually replace today and can add value into enabling them, empowering them, collect payments and use it as their payment gateway? These are mostly like subscription based products and they're like license based products. Robert Kraal 12:18 So we're talking about if you talk about our products, we're purely focused on cards. So Visa, Mastercard, American Express and a bunch of other cards. And so if you look at recurring businesses, you look at specific cards, transaction types, and obviously we support them. So there are different types of flagging for doing your recurrent transaction, merchant initiators, customer initiators, recurring transactions. And what we see is that many of the platforms that are out today, and then we're talking about acquired platforms, not necessarily the platforms that you mentioned, that they have been designed in the 70s, 80s, sometimes in the 90s. Quite often you will find COBOL code running on IBM mainframe. So at the moment that Visa, Mastercard or any of the other cards launch new functionality around recurring transaction, subscription transactions and in particular the security around making sure that you actually get the money when it's come to card updates or that you always have the latest card, stuff like that. You see that for those old platforms to actually develop that new functionality, it usually is quite complex because they have complex platforms and they have old technology and they have different ways of doing their tests. And quite often the technology that they have is a combination of different platforms that they acquired over the last decade, so to say. We don't have that. So in general, if there's new functionality coming from the cards, we are able to launch that very quick. And the cards teams love that. They love to work with us because they can sort of push products to their customers, to the acquirers. And that allows them to put a little bit more pressure on the market in general, but they can demonstrate that the products that are working. Particularly if you're coming at subscription models, yeah, our set of tools, this is not new in the market. It may be new in some markets. It may be new for companies using older technology, older platforms that those platforms don't support it, but we do support it. So how does it translate into potential customers that would like to use that? Well, on the one side, and we have different variants here, so I'm describing one variant. They could technically connect to us and we would give them from a technical perspective all the functionality that they are looking for when it comes to managing those transactions, all those different transaction types. And we can help them with that. And I can pretty much with a lot of confidence say that we can support all the functionality, all the technology that we have also in the U.S. But we're not an acquirer ourselves and we don't touch the money. So in these type of setups, you always still need to have an acquirer. And most of these companies... Adil Saleh 14:55 Could you also please simplify the acquirer for the audience, please? Robert Kraal 15:00 Sure, sure, sure. An acquirer is in the U.S. mostly a bank who has licenses to accept Visa and MasterCard transactions or American Express or Discover Diners Club transactions on behalf of the retailer. So if you're a retailer and you want to have acceptance of cards in your shop, whether it's an online shop or a physical shop, you typically go to your bank and ask them for a solution. And the bank has the right of a license of Visa and MasterCard to accept you as their customer and handle your payments for them. But that's the legal and the licensing part and the financial part, maybe. We provide the technical part to those acquirers, so we help those acquirers to actually take those transactions from the retailers and we push them to the card network itself. And as mentioned, there are other companies doing that, but our competitors in general are dating back 70s, 80s, sometimes 90s and literally use very old code. Adil Saleh 15:54 So here in the U.S, you talk about like you're going to be working with banks or with these third party products like Brex, like a lot of these companies, like startups and businesses, like SMBs and tech companies, they are basically using these platforms like Mercury, Brex, they're using for subscription, they're using Stripe. Stripe is more like an integrator, more like a payment integrator and very minimal amount of businesses are using Square, more local businesses using Square and all of these like legendary technologies, like legacy technologies from 90s or early 2000s. So how do you like relate, like how Silverflow can fit in and what it can replace? Is that going to work with Brex and all these tooling for payment? Robert Kraal 16:45 Yeah, absolutely. So our customers essentially are multifold. So on the one side, we have the banks, so they typically run all technology. They're looking how to get from this old technology to the new technology. Why would they do that? Well, essentially, they want to have better technology, the more modern features. They also, of course, look at the cost element. So they want to reduce costs, both the direct costs as the indirect costs. Think of all the people involved, because usually quite a lot of many work. So that will be one of our customer types. Another customer type in the US would be the ISOs or the MSPs. So typically, independent companies who may sell you your card present terminal may handle the support and do the installation and stuff like that. And they get a margin for that. Behind those companies, there would be an acquirer, an acquiring bank who actually would fulfill the financial role and also the role of Visa MasterCard. But from a technical point of view, these companies can connect to our system and using the bank's license if the bank agrees with that, of course. So we will put part of the bank's license on our platform. So this ISV can technically connect via us to the card networks. But since it's the bank's license, they still have the same legal, commercial, financial relation with the banks. And that is a model where essentially all the companies that that you mentioned would fall in as long as they have their own bank. And this is particular for the card makers. Adil Saleh 19:05 Yeah. As long as they become their own acquirer, like they become a bank. Robert Kraal 19:13 Or if they have their own preferred bank. So it could be that they already have a good relation with the bank. They don't have to break that relation. They can keep that relation with the bank. But essentially, we go together to the bank and say from a commercial and financial and legal relation, we would like to keep this relation between this gateway and the bank. But what we don't like is the technology that you offer as a bank. We want to replace just the technology with another type of technology. And the typical question of the bank would be, well, do I need to pay for that? Well, in general, no, because that other gateway or that other company would definitely connect to us and we would connect them to the card networks. But we do need to have access to part of the license of the bank. And therefore, we can do these type of integrations for a fraction of the cost. And if the bank themselves would have to build this innovation. Adil Saleh 19:05 Hmm, interesting. So it becomes a right like kind of a bridge for them and a bridge. And also, you guarantee the security because it's a big problem. And that's why a lot of these payment gateways over the past 20 years, they came and failed because they were not secure. So now thinking about, of course, go-to-markets, that's going to be a big talk of the town and also in fintech as well. I know it's sales cycle and everything is pretty on the lowest side. It's so hard to break the ice with all these big banks and companies. And it's a mission critical decision that you got to make it. It's more like seafood conversation, not like low level conversation, more table talk conversation. So how is that going along for Silverflow in Europe and outside of Europe? You know what it's like. I would say what kind of your customer segments that are your sweet spot as you mentioned the banking sector and anything beside that, too. And how is gold growing in general as a founder? Robert Kraal 20:02 Yeah, I think that's your spot on. So to launch a product such as Silverflow is from the go-to-market. It's it's not easy because you're sort of delivering the core of companies to do something with payments and for banks to migrate away from that. Typically, they see they know that what they have today is not ideal. It's a bit outdated. But they also don't really see the path forward to get out of that. And that's a big project with a lot of migration, a lot of risk, a lot of people involved. So that's as difficult as an easy early sell. And this is why we deployed sort of like the strategy of the one side to tap into new acquirers first. Well, you're right. You don't see them that often in the U.S, but you do see them, for instance, in Europe or Southeast Asia or Australia, New Zealand or Africa or Latin. So essentially, the U.S. is a little bit of an exception here where you don't see that many new acquirers, but in pretty much all the other markets we do. And if you are a new acquirer, then usually there's a reason why you're a new acquirer, because you are you think you can do things better, quicker, cheaper, more functionality. Quite often, these are quite tech savvy payment companies who've been very successful, because otherwise you wouldn't be in the discussion with Visa and MasterCard to become your own acquirer. So a logical match when it comes to a technical processing platform what we deliver is with those fast growing, very tech savvy, fast moving payment companies. And they will talk to other of our competitors. So a little bit more the legacy players. So they will talk to them as well. But quite often the sort of the duration of the integration projects. In our case, we've seen integrations going as fast as five, six weeks where market standards average of our competitors. You're looking to 18 to 24 months. So that alone, even aside the technical differences, is an argument for many of those tech savvy companies to say, well, I don't want to do an 18, 24 month integration project. I've spent a lot of time and money to get my license. I want to go live tomorrow. So that in a base is already a differentiator. But there are also many other differentiators. We've set up our system to be data first, to be cloud native, to be very simple and foolproof and stuff like that. But the other systems, they have never been safe. You set up a system in the 70s or 80s, you were concerned about hardware. Everything was expensive. Storage was expensive. RAM memory was expensive. For transactions, you want to have as little processing cycles as you could have CPU cycles. And all those things don't count anymore. Today, you would design different systems. You would keep the data instead of only keep the data that you barely minimally need. And you can keep it for a little bit longer time, where back in the 70s, you would throw it away. In real time, in RAM memory, you can hold much more information, which you can use for Adil Saleh 22:57 Cloud access and all of that. Robert Kraal 22:59 Exactly, exactly. So you design your system as a completely different type of system. And this is essentially what these new players want. For the banks, that's our second go-to-market. They typically are a little bit more risk averse. They're also big organizations. So within this organization, quite often, a lot of people need to decide. And that's why we focus on both. So we focus on the first group, because they're usually quick and have decent volume. We focus on the second group, knowing that they're a bit slower. But quite often, these banks, once you're in, you get access to their total volume, and many banks still have significant volume. What is good on our mind is that all the banks, whether they like it or not, they will have to migrate to newer software between now and, let's say, the next five to 10 years, because the systems that many of them are using today, they're too old, essentially, for them to maintain the business as they have today. So either those processors should upgrade their systems, or the banks should migrate away from them. Adil Saleh 23:59 I think you absolutely made it. Because at the end of the day, all these new-age acquirers and banks that you're talking about, and you're working very closely with, it's always equally important for them to evolve the technology as well, to be able to not be replaced in 10, 15 years, like many are from the legacy systems and all. And it is so hard for them. You talk about these big banks from two, three, four, five decades. It's so hard for them to do the chain management, because regulation, security, all that. Robert Kraal 24:32 What you see is, of course, that where will those newer acquirers take the transactions from? Well, on the one side, from the old banks who do not innovate. They go to the customers of these banks who, for different reasons, do not bring these innovative products, and they go particular to those customers with their new products. And you see that, of course, on the one side, you have quite a bunch of big global companies. We've mentioned a few, Stripe, Adyen, Checkout, and there are a few other pretty large global companies who are firmly beating on the banks. And they're beating on the banks with better technology and more global technology and better pricing quite often and better customer support and better adoption to the latest technology. So that is already a big competition to the banks. But you see that in every market, whether you go to domestic markets in Southeast Asia or in Latam or in Africa or in Europe or, yeah, in my perspective, in the U.S. as well, pretty much all the bigger banks run all technology. And there are a few exceptions. I must say that as well. So there are also a few banks who actually have this quite under control. But in general, you could say that the smaller players move quicker, are able to adopt quicker, are able to provide new functionality quicker. And typically, the fast-moving retailers, they want that. They want to have these high technical functionality products. So they will move away from the banks to these newer players. And what is left with the bank? Well, that's a good question mark. Adil Saleh 25:58 Absolutely, that's a big question mark. Absolutely. Love it. So now keeping on the go-to-market side, I know that you're starting, you're trying to go upmarket from like bottom-up approach, and it's always good. What kind of new innovation are you working closely with these new acquirers and some payment integration giveaways that you mentioned that this is also your sweet spot? Like what kind of new innovations that you're so excited about this coming year in terms of product, integration, new initiatives on the go-to-market, new investments, new partnerships, all of this that are having a direct impact on the commercial side of your business as a founder? Robert Kraal 26:37 So one very important element is geography. So at this moment, we're active across Europe, across Southeast Asia and North America. And as I mentioned, we've designed it to be a global platform. There are also a number of other markets where there is quite a lot of opportunity for us. Typically think of markets where cards is big and also in the markets that we are active, there's still a lot of growth, space for growth. And that means that in some markets, we also want to have specific focus areas. So on specific segments, I think of the hotel, restaurant segments, fuel segments, for instance, travel segments, retail segments in general. Of course, we do card present, Ecom. So a lot of the products that we develop are geared towards acceleration of growth in our core markets. And quite often working with partners in those markets who are quite often have a vertical SaaS approach. So that would be payment partners specifically focused on a particular segment, and we would help them with specific products for their vertical. Adil Saleh 27:51 Interesting. And you also mentioned you're trying to be a data first company, which is what I absolutely love about it. Like how you're investing into measuring success for your customers. I know that these are smaller banks and smaller payment gateway integration retail businesses that are using Silverflow. But again, at the end of the day, you need to make sure you're evolving them and you're growing with them. So it is essentially important for you to measure the success, like how they are winning with your product, like how many number of transactions or goals or any KPIs that you guys are tracking. You must have your post sales or sort of a go-to-market team. Tell us more about that formation. Like you have VPs, you have account executives. What is the formation of your team and what is the core focus towards the customer success side of things? Robert Kraal 28:37 To start with the teams, we have essentially three different type of teams on the commercial side. One focusing on, let's say, high-tech new acquirers. The other one focuses on the banks. And why do we differentiate that? Well, you have to have a different type of approach. Within banks, you need to talk to an enormous amount of different people and really understand the value. So it's longer sales cycles for sure, so also slightly different people. And we have a third segment, which would be the largest retailers, typically processing over, let's say, 1 billion on card volume. So we have those three teams purely on the sales side, but they work very close with our product people because banks typically have a complex setup today. They have complex demands. So the salespeople, yes, they do the commercial contractual negotiation, but we typically, in those sales process, are also consultants on how the migration, for instance, would work or which functionality, how we can plug in the relevant set of reports into the systems that the banks usually want to keep or other customers want to keep. So there are quite a few people involved. In the after-sales process, we have a migration and a setup phase where we do all the end-to-end testing, making sure that the product works exactly as our customer asks us to deliver it. Once they are live, most of the discussions are on the one side about how to ramp up the volume. And there are different ways how we do that, but sometimes it also depends on the contracts that these companies have with their other providers at that moment. So we have technical account managers and commercial account managers. The technical account managers typically help our customers to understand everything about the latest functionality and the technology around it. If there would be implementation consequences, if there would be changes from the card networks that are optional or mandated, and how relevant they are for the customer. So there's really a focus on the product itself and the technology around it. And we also, of course, have our commercial account managers who look at how we can increase the relation with our customers. So essentially how to help them with ramp up the volume to our platform, but also, of course, cross-sell and update with new products. We do that slightly differently in different markets. So, of course, in the U.S. we have a banking team, essentially, which also focuses on ISVs and MSPs, stuff like that, simply because there are very few new acquirers. So that is slightly different. But in general, we take the same approach across the globe. Adil Saleh 31:07 Ok, that's right. That's right. And also, you mentioned, the sales cycle is always going to be longest with acquirers, working with these big banks and acquirers wherever you are. So now thinking about the subscription economy, I know when you talk about subscription economy, it's a huge chunk of tech companies. There are millions of, there's actually a transaction millions per day, on average from seed to series A, based on more than 170, close to 200 interviews that I've done between seed to series A. Annually, they spend between three to five million dollars on a recurring basis. These are just recurring payments I'm talking about, not some investments that they're making on the marketing and spend and all of those internal that their CFO always bang their head around. So now thinking about these tech companies, how do you think Silverflow can play a huge part? Talking only about, let's say, the Netherlands is pretty big, Amsterdam is pretty big in tech too. So how these companies are trying to adapt to these platforms and payment gateways, shifting and moving from these legacy systems. And what is the adoption going on in the tech in Amsterdam specifically? And, you know, these people listening in the U.S. can relate to. Robert Kraal 32:22 Well, when it comes to this type of recurring and subscription type of payment, what's very important, there are a few things very important. First of all, to store your data. So we have PCI, so everybody needs to store them. But you can also look at tokenization and then there are different type of tokens, whichever token suits the best for you, whether they're network tokens or your own proprietary tokens. So there's a whole set around that management of those transactions. Then as part of that, you have cards to expire. So you think of products like Card Updater, which should be included, which you actually want to have included in your tokenization solution in general, that you can use these tokens and that you can guarantee that you always at least have the latest files, latest card details on file. So those things are important. Then when it comes to actually sending the transaction and getting authorization, of course, you want to get the auth rates as high as possible. Also, you want to have the fraud as low as possible. And in both cases, data plays an important role. So make sure that you flag the transactions in such a way that you get actually the highest outcome and that you can also predict that and can analyze that looking backwards. Also, the part of transaction gets declined. There's also an enormous amount of information potentially with those transactions. So what is the reason for that kind of transaction? Is it because the card has been withdrawn from the markets or stolen, destroyed, whatever? Is it insufficient funds? And you can imagine if it, for instance, would be insufficient funds that you might try at another time to still recover that card. And many of those companies that are in the subscription business, recurring business, they actively are doing this, but quite often they don't get the right amount of data to make a good prediction of what would be the best moment to do a retry on this particular card. And quite a few of those type of solutions that all have to do with improving that conversion, making sure that your card details are always accurate, making sure that your details are safe, secure, that you use the right network tokens or other tokens, and that you can actually send them in and that you get enough detail to analyze why do I get this all the way back? And can I do something with the transactions that have not been authorized now or maybe in a slightly later stage? This is all a data game. So currently they would send them into a black box and you can sort of offer a client message back. And the more detail you get back, the better you can analyze if it's correct, what you get back. And you can make small changes to optimize this to get a higher conversion on day one and also to have a better recovery process afterward for your declined transactions. This is different whether it's Amsterdam or the US, by the way, there are companies across the globe doing exactly this. Adil Saleh 35:02 Yeah, I mean, a great story like that happened this last month. One of my friends, he was in D.C., got stabbed with this phone. They figured out they were actually chasing and stalking him for a good 30, 40 minutes and they figured out his passcode of his iPhone and he went into his bank, which was Brex, and they started making transactions and starting pulling out cards and adding cards onto their Apple Pay, a different device. And of course, that is quite easy for Brex to detect: hey, this card is at the middle of the night, got added to another device. So they actually blocked it. That's easier. But from the same phone that they stole, they got inside the bank apps. Somehow he's not sure how they get, maybe they kind of drugged him or something. But they actually got in and they made the transaction. It was like 700 bucks, 80 bucks, Uber Eats and Uber, there was 70 bucks for buying some gift cards. So they made four or five small transactions from within his Brex mobile bank app. And good thing is, of course, his phone is gone, not with him. And they were inside of the banking app within 30 minutes, they made all these transactions. And when he got back home to his computer, he logged into Brex app. And only one transaction that was of 70 bucks actually went through, the rest got flagged and got actually disputed. I'm still figuring out how smart those checks could be. What could be is you tell me, how do you see this? What's your viewpoint? How were they able to flag those transactions? This is so smart, altogether, it was around $1,200 that they wanted to basically transfer from his phone, from his app. And one, of course, Brex won't know, but they actually disputed it on themselves, apart from that 70 dollars. Robert Kraal 36:56 Yeah, there's some smart technology, to a degree, some technology is predictable. So if you look at pretty much any person, they might have a different difference per card. They might have two credit cards, in the U.S, I think, three or three and a half on average. They typically have like three email addresses. One is your business email address. One is your private address, number one. And the other one is what you use for commercial newsletters, stuff like that. Essentially, the stuff that you don't want to have in your main mailbox. People typically have a few devices, a phone and laptop and iPad. They also have a few IP addresses, which would be the one from your phone or the one at home or the one at work. And delivery address, same story. People have a few. So at the moment that you have sort of profile of every customer, you can also see which one is the odd one. So, yes, it may be the same phone and the same payment methods, but the delivery address is different and the IP address is different. And the speed that somebody types in the key code on the phone might be different. Adil Saleh 38:05 So many things Robert Kraal 38:06 So a lot of these things can be measured. And some risk systems, they do measure the angle that you hold your phone, for instance. Typically, those type of things you can measure and you can use to make a prediction whether the person who actually holds this phone is the actual person that is normally holding this phone. And these are small pockets. There are a lot of different smaller things that you can measure on. If many of them are normal and there's one odd one, so everything is normal, but the address where the food gets delivered is different. Well, that's not seen as a high risk because you might be at a friend's and ordering food there. But if too many of these parameters are different from compared to what the algorithm would expect, then it would be flagged as a high risk transaction. And it depends a bit on what type of products you can imagine. If it will be a digital download of a newspaper, the actual risk is not that high because the value typically is low and the cost of the product typically is low. But at the moment that you have differences in the product that actually is being purchased, where it could be the value high, margins low, stuff like that, that you make different decisions on the basis of those risk profiles. But a lot of different smart companies do an enormous amount of interesting work to try to detect if people are actually the people that they say they are. Adil Saleh 38:25 Yeah, that gentleman had a really good meal in the middle of the night, but that transaction got disputed and got refunded by Brex at the end of the day, too. Robert Kraal 39:35 I can imagine why. Adil Saleh 39:36 Love that. Yeah, go ahead. Robert Kraal 39:37 Because a meal, for instance, is typically a low risk product. Yes, there are certain costs involved, but you might have the same risk profile and say, well, but it's only a meal, so let's deliver. Where the other products would actually apply.. Adil Saleh 39:50 And he ordered Uber for $110 as well. And he thought, he's a good guy, financially very well off, he's a founder, but he never booked an Uber for $110, you know, he maximum booked an Uber for 50, 60 bucks at max, but never $110. So it was quite about 90, close to 95 miles from center of D.C. to Ashburn, that's Virginia. So that was a long ride. And yeah, so he had a long way. We spoke with the Uber driver, too. And we actually got the number from the Uber app and we called the driver in the morning. Hey, you dropped this guy from D.C., from this place at 1.56 a.m. in the midnight. Ok, yes, he said yes. And this is what happened. He stole our phone. We suspect and he said, Hey, He was talking by this name. His name was Chris. It was a black guy with the gig and everything. He showed us everything. We've done half his job, like police job. You know, we went to Apple, did the police report and everything. But yeah, Apple security sucks. So that was moral of the story. So it's just kind of an experience that I wouldn't imagine that I would get to have it on a recorded podcast. But it was really nice to pick your brain on this, how they actually analyze this data to navigate through this fraudulent payments, beyond just basic parameters like middle of the night and all of this. So it was good. So now thinking about, as a founder yourself, I know that you've been traveling all across. You mentioned, you know, is that Asia? Is that Central Asia or Southeast Asia? Robert Kraal 41:29 Singapore is now Southeast Asia. Adil Saleh 41:31 Southeast Asia. Ok, so what kind of countries are you more focused on, Singapore and then? Robert Kraal 41:37 We have customers in the Philippines, Malaysia, Thailand, Vietnam, Singapore indeed. Yeah, I don't know. It's quite, quite spread across the region. We have discussions in countries also in India and, of course, in Australia, New Zealand, Indonesia, Japan. So it's pretty scattered all over to where the innovation happens at this moment. Adil Saleh 42:05 Love it. Love it. And just as you are as a founder, that is tech first, data first and trying to ride this wave of AI. What is it next week that you're thinking about involving AI into it? I know that AI is not just about building AI wrappers in fintech, but there are a lot of platforms building agents or vertical agents within their fintech platforms working with banks or a lot of these financial platforms like financial analyst, these QuickBooks and these platforms, they are using a lot internally and then making it available for the customers to optimize their operations and efficiencies and accidents. Robert Kraal 42:45 Yeah, so we use that as well. So on the back end, we use AI to essentially, most of our customers, they don't want to assign a lot of people on their side to do daily operational stuff. So we use a lot of AI there to maximize, to optimize so that you need as few people as possible to do your operational stuff. You still want to have a few people actually controlling that everything is right, but you don't want to have batteries and holes of people doing dispute management and stuff like that. So we use AI for, let's say, those back end operational processes. What is very, very in the spotlight at this moment is everything has to do with agentic commerce. So although, of course, we are not the end retailer and we are not building agents to your personal shopping agent, for instance, we're not building them. But of course, we have to make sure that our platform is ready for that so that the messages, that the payment messages are generated by these agentic payment agents, that we can work with those messages, that we can put the right flagging to it, the right level of security, that we can pass them on to the corporate person the right way. So there's quite a lot of development on our side as well on that element, which is quite interesting. There are not that many real life good examples, but there are an enormous amount of companies building on this technology. So this is quite cool. Of course, we see this difference with market by market, things around stable coins. So we quite often get the question, what is your view on stable coins? We are not the acquirer ourselves. We don't touch the money, so we don't touch the stable coins. But you can imagine that just as any other currency, the card networks might want to settle you in stable coins or that they might want to accept stable coins if you have a card, which is a stable card. But it also has impact on what we are supposed to, what we are required to do in terms of managing that and sending it the right way. So those are the things that are not used that much today, but where there's a lot of development in general in those markets. And we have to make sure that our platform is ready for that. So the companies that want to work with that at least have a processing platform that can work with all these type of tools. Adil Saleh 45:04 Interesting. Now, the one last question that came from the team, how much kind of initiatives, investments are you making towards the training side of your people? I know that there's so much AI and coaching platforms for sales and enablement, success, customer support, customer service. So what kind of initiatives are you guys taking to train people, any tooling, any consultants, anything that you're doing at this moment? Because this has been the biggest talk of the town, how to train your people over time, have them empower their skills, especially soft skills, because hard skills are now pretty easy. You know, you can get a chat between OpenAI courses and all of this is on the Internet. But when it comes to soft skills, having the right motivation and alignment with a career, attitude, drive on a daily basis or bottom of the vision and being a team player, all of this. So how you're instilling this, what kind of training initiatives are you guys taking? Robert Kraal 46:05 Well, we indeed we cherish the specific culture that we have, which is a very international culture. So we're with 84 people. And I think we have like 35 different nationalities across the globe. Main language is in English, but typically we celebrate a lot of different independent holidays from all the different countries, whether it's with sweets or candy or other festivities. Culture is extremely important. Our business is very knowledge intensive. So it takes quite a while before people really understand what we're doing, why we're doing it. So it's also important for us to keep our people as long as possible. It's too expensive to train them and see them go after a couple of months since they will only be productive after six, seven, eight months, so to say. So we do spend a lot of time in building corporate culture. We like to work indeed with, well, but probably every company wants it, with smart people, with entrepreneurial people, with friendly people. We have a no hire, no nasty people policy. And they get a lot of freedom to work on complex project with colleagues. But at the same time, we help them train as well. And we give them an insight of other parts of the business as well. So if you're an engineer, you're very welcome to join in commercial discussions or in which is a large part of my discussion with the investors or potential investors. Just sit in and see how an investor call works. And we have an open floor space. So that makes it always easy to ask questions and what I'm telling to pretty much all the people and say, Ok, you're working here now and you're probably not going to work here for the whole of your life. And that's Ok. I want you to work here as long as possible. And how will you stay as long as possible as long as you keep on learning? And there will be a moment that I cannot teach you anything anymore. And then I hope that the biggest, biggest thank you that you can give me if you start up your own company. So try to do everything in this company so that you feel comfortable enough to start up your own company afterwards. And if they're lucky and if we're lucky, they all have their option shares or whatever, that they get a little bit of money with it as well. So that would be the greatest thing that you can see if your team stays. And at a certain moment says, well, I loved working for you, but now I want to set up my own company. I'd rather have that than that they say I'm going to move to the competitor. Adil Saleh 48:30 Absolutely. Yeah. My last job was back in 2017. That was a tech company. I helped them build the customer success and their sales organization from scratch in three years. And then I quit. These were my friends back then as well. And now they're all friends. Like we are doing everything together. Like we had an event together in SF just last month. And we're doing investment, investing in the same things together. And it's such a great feeling that you get to experience this entrepreneurial journey with the people that you have worked for in the past and share the intuition, share the knowledge and love it. The last thing that you mentioned is knowledge sharing is knowledge heavy kind of training. So is there any tooling that you're using, any AI part that you're using, any sort of training centers that you have to enable these people to get that knowledge with the right context in the quickest time? And then they have some sort of maybe assessments to validate that knowledge in the quickest time. So they get top notch before going to the customer, to go to the product team, to go to the field operations or whatever functions you have. Robert Kraal 49:31 We do that for some specific sets of knowledge, but not for all the knowledge. So some of the knowledge is so specific that it's easier to train them, particularly within their team and stuff like that. But for some of the more generic blocks of knowledge, we use those type of tools, yes. Adil Saleh 49:50 Interesting. Now, Robert, it was no less than a pleasure to meet you and getting to know you in the industry and how I got to learn a lot, by the way, and thank you very much for taking these 45 minutes for this episode. I'm sure a lot of people listening to this will definitely have a huge curve of knowledge to seek in the payment gate, present a new generational payment integration processes that are making a big shift. So appreciate your time. Robert Kraal 50:15 Well, thank you as well. It was a nice, nice discussion. Adil Saleh 50:18 Thank you so very much for staying with us on the episode. Please share your feedback at Adil at Hyperengage.io. We definitely need it. We will see you next time with another guest on the stage with some concrete tips on how to operate better as a customer success leader and how you can empower building some meaningful relationships. We qualify people for the episode just to make sure we bring the value to the listeners. Do reach us out if you want to refer any CS leader. Until next time, goodbye and have a good rest of the day.