60. Catch the Tornado with Kristian Tysander

Krystian Tysander

Brink Commerce


Piotr Karwatka  0:53  

Hello, everyone, I'm super excited to host Kristen descender. In today's episode, Christian is a founder of Brink Commerce, a serverless high performing headless ecommerce platform offering a flat transactional rate. And I'm super happy to discuss why we need another platform today? What makes BC unique, and how they got started. Hi, Chris. Nice to meet you. 

Kristian Tysander: Hi, Peter. Thanks for having me. 

Piotr Karwatka: Thanks for accepting the invitation. So how did you get into this industry?

Kristian Tysander  1:27  

I've been active in the e Commerce Industry since 2007. I started out as a merchant, together with my brother.

Piotr Karwatka  1:22  

Tell us a little bit about your background.

Kristian Tysander  1:36  

We have basically sold everything you can think of from plastic colorful watches to thin silk ties.

Piotr Karwatka  1:43  

So you have had ecommerce shops before.The merchant having the experience from the other side.

Kristian Tysander  1:52  

Exactly. And my brother always took care of the marketing and product design, and I handled the technical stuff. And that's how everything started.

Piotr Karwatka  2:03  

Awesome. That sounds perfect. I read on your LinkedIn that you also had some episodes in the army forces.

Kristian Tysander  2:12  

Yeah. When we started out, we were quite young, I think I was 23 at the time or something like that. So we were still in school or at university. After doing some ventures together as merchants, I studied abroad in Australia for a year. And when I came back, my brother had already started Daniel Wellington, back then it was a very small company running from a basement in Uppsala, with him and one other employee. So I was looking for a job I started out in the Swedish Armed Forces as a systems engineer doing operations stuff

Piotr Karwatka  2:57  

Did it contribute anyhow to your future career?

Kristian Tysander  3:02  

I think I saw it as a start to be part of a bigger context. They have a long historic background from hosting and operating very sensitive systems. So I wanted to build upon what I was actually most interested in like the hosting systems administration parts in the beginning. And off the year, the off the year there. If you know me as a person, I I can get pretty restless. So I started to look for another challenge. And I saw an ad on the local intranet for the Swedish special operations forces. They were looking for technical personnel. So I sent in an application and did some tests and some basic training and then I was accepted into that.

Piotr Karwatka  4:01  

Awesome and how you got from these special forces to find commerce.

Kristian Tysander  4:11  

After a couple of years, in the Swedish soft community, I felt that I was traveling a lot. I spent more nights away than going home with my family eventually. And I felt that you know, I need me and my wife was expecting our first child and I was looking for something more stable. So one day, my brother asked me that Daniel Wellington had started to really take off, he has started to hire some people that have gained some traction. And he asked me, you know, none of these we all have, we only have external suppliers for all of it and they can't keep up with the pace we want to try to hold there. So he was asking me, can you take Look. And I felt that I felt that this could be a good opportunity in order for me to transition over to something new.

Piotr Karwatka  5:09  

So you joined them as a kind of CTO?

Kristian Tysander  5:12  

Yeah, I accepted his invite, I came in first as a consultant, just looking over everything. And I felt that, you know, in order to do what needs to be done, and I was listening to his vision, together with the CEO, his newly appointed CEO, when I felt that, you know, you need to take responsibility for this by yourself, you need to start hiring developers, you can't rely on consulting firms. So I did like a brief overview, and I told them that we can do it like this, I can start hiring developers within my firm. And we can do like an invoice one to one. If it doesn't play out the way you want it to do, you can just distance yourself from it. It's a low risk, where we can start hiring developers. So we did that. And I basically became CTO started to hire developers,

Piotr Karwatka  6:16  

and what you built there for Daniel Wellington became kind of a foundation for a commerce platform.

Kristian Tysander  6:23  

Yeah, I would say back in, I think it was 2014. With like a flash sale campaign in China. We were early into China. And we actually hosted our own Magento platform inside Mainland China. And there was a specific moment where me and my operations guy and another developer, we were just kicking back at the office and the marketing team had prepared a flash campaign where the first 50 customers were to receive a free watch strap. And, and we thought back then that we had set everything up properly, you know, we can handle the pressure. But, you know, in hindsight, we didn't understand Chinese company, shopping culture at all. So when the clock hits midnight, I think it was like six o'clock in Sweden, it took 30 seconds, and then like an alarm started to go off, everything just went black. And we didn't understand what happened, you know, if you lose the connection to Asia, or what's what's going on. But after looking we did like a quick post mortem in the database, and we saw that Magento managed to register 291 orders during the first 30 seconds, and then basically collapsed.

Piotr Karwatka  7:47  

That's really awesome for us because Magento is like it's reached its roof very quickly.

Kristian Tysander  7:54  

Absolutely. And that's that's that's pushing the limits, we think the actual amount of orders were much, much higher. Yeah, this was only the orders that Magento managed to register. So I think that was an important aha moment for me and my team that you know, this. Yeah, exactly. These systems, they're not constructed to cope with this kind of shopping behavior. You know,

Piotr Karwatka  8:17  

So, what makes Brink Commerce different? Because I really liked your value proposition. I suppose it was when you announced your first round, that I read that your value proposition was hyper reliability, meaning a platform that never goes down like it never fails. How do you make it happen?

Kristian Tysander  8:46  

We make it happen by a couple of principles that we try, I want to try to keep it high level. I'm not a developer. But we apply a couple of principles. And that mainly consists of the fact that we're doing everything like microservices architecture, we try to apply a shared nothing approach, where all of these services are independent, they can't share any data store, so anything and then we try to apply or not try to apply but we apply an event driven design. So we try to consume and produce and everything is producing notifications that you can subscribe upon. So the idea here is to create a stable enough system that can push if we take an order, for example, we can push events and orders further downstream. And try to avoid having hard dependencies between services. So you can create an order and then we can push it to the next service and to the next service and to the next service. And by and by doing so. And also making sure that all of the services, all of the third party, all of the managed services for us, everything is run on serverless infrastructure.

Piotr Karwatka  10:11  

Gotcha. Yeah, that's, that's super interesting and human serverless. And on our pre call, when we prepared for this interview, I noted that this is super important for, for the whole thing was why it's so important. I mean, the serverless architecture

Kristian Tysander  10:25  

Serverless infrastructure is a technique where the cloud providers take a bigger responsibility than you're normally used to, when you're, when you're running a virtual service. When you're running a virtual service, they give you the option to manage the actual, the address of the server, they're doing this for you. So the handover is much smoother, it's much more elastic, what we're basically doing is that we we insert code into a cloud provider and say that this code is being supposed to be run on a certain event, for example,

Piotr Karwatka  11:05  

Yes, so you deploy the function, and it should just work and you don't need to take care of logs, rotation, I don't know anything about system maintenance anything.

Kristian Tysander  11:16  

Exactly. And, and the ability for these functions to scale, when the code is like floating, the functions are floating on top of it much, it makes it much more scalable. And

Piotr Karwatka  11:29  

making the things a little bit more complex. I mean, you know, when the commerce platform was more mono political, it was very easy to orchestrate everything. And here you have those functions that need to talk to each other. Via this, even buses, as you said, aren't things even more complex? And I was always thinking like, with complexity games, a little bit less of reliability, because there is more things that can fail isn't like, yeah,

Kristian Tysander  12:00  

I would say, to try to answer your questions, there are some disadvantages, for sure. You have more virtualization layers than then you're used to. So it can add a couple of milliseconds in latency, you will never be as fast as if you run it on your own hardware, you know, here you have a couple of virtualization layers. So in a vertical sense, it's a bit slower. But the big advantage is that you can scale it horizontally, horizontally. So yeah, much, much better. So even though a competitor can be faster, local in one region, because they're closer to the hardware, we can be maybe a bit slower in a vertical sense. But we can be much, much more scalable in a global sense, horizontally.

Piotr Karwatka  12:59  

I think it makes a lot of sense. Because no, those lesson latencies you mentioned probably are not that significant per each individual user. I think it's something like in milliseconds, so you can deal with it anyway. But the advantage is that with this girl distribution also and the whole, you know, cloud provider, managing your infrastructure, because this is what Serverless is actually, with all those, the benefits are still bigger, because your platform is faster the global, the global way, I can say yeah, check, it works for bigger number of users.

Kristian Tysander  13:39  

Yeah. And that's how we try to visualize it is that if you take a traditional platform, and you host it in Europe, you have, let's say, 200 millisecond response time from within Europe. And we have, we might have 300 milliseconds, so we're a bit slower. But if you apply the same thing to a customer that has customers from Australia, we can also be in Australia, we can scale out

Piotr Karwatka  14:04  

to 300 milliseconds on those guys can be one and a half seconds in this case.

Kristian Tysander  14:08  

Exactly. Head on point. That’s, Yeah. And that's exactly why we're trying to do it. And we also think that it's going to be more and more important, in a sense, where the E-commerce market is growing so rapidly. I know a lot of customers or potential customers that had problems when the E commerce penetration was, I don't know 15-20%. During the past 12 to 18 months during Corona in at least in the United States, the penetration has grown from 15 to 30%. What are these companies feeling when we went from 15? They were in pain at 15%. What are they feeling now? What are they going to feel when we reach 50%? You know what I mean? So I think it's getting more and more important to be able to scale

Piotr Karwatka  15:00  

Absolutely, I think that makes perfect sense. I'm curious about the feedback you're getting from the market. It's an early market, isn't it? Right? Yeah, not that long gone. actually went when you started with no selling, there's this platform.

Kristian Tysander  15:16  

Yeah, when we started to sell it as our first customer went live in September, little over a little over a year ago in September of 2020. So we've been live for 13, almost 14 months now. And the feedback has been very positive. Most of our customers aren't that technical. I think they see an upside in it, like, we always identify three things that sooner or later every ecommerce company will realize is that either you get performance issues, you get lock in effects, you can't access your data the way you want to. And everything gets really, really complicated after a while. The time to market is really long, you know. And I think that the biggest benefits our customers see now is that they're much faster, like the transition in time to markets. When the front end is the couple, they can move much, much faster, they can focus on the customer experience. So that's one of the like, the biggest benefits they see in the short term. The next one is as we move or progress behind the fact that they can move at the pace they want to, they start to look at, you know, what, what data are being produced with embracing what can we see what can we do? So they start to open up and think about, you know, how can we use this data to move even faster, take better decisions, and, and that's something they're not used to. So that's something we're helping them with now. We're giving them apps access to a lot of data that has been unavailable before. And I think to be honest, the performance things are benefits that most companies, as long as you don't suffer from it, you just take it for granted. So I think they're really benefiting from moving faster and being able to work more with the data to take better decisions on.

Piotr Karwatka  17:36  

It's actually a good point, you mentioned that the performance is something that makes sense with a scale for customers. So if you are absurd you probably are not thinking about it. And then when the scale grows, you see some needs for performance improvements. My question is, what's your market strategy? How do you aim for those clients that are already on the market? Maybe experienced some problems with performance? Is it the best segment for you? Or maybe you aim differently?

Kristian Tysander  18:15  

Yeah, we started out with like, kinda small emotions, just to get some early adopters, and we're trying to climb, climb the ladder and get higher and higher up. So we're trying to target more like enterprise customers. In that sense, they're not traditionally enterprise, they can be direct consumer brands, but they have big, big volumes. So that's the market like mid to enterprise that we want to try to get into, or we feel that we're already working with customers, like accelerate Gafta, for instance. And one of the value propositions there is that they really feel that as the further they go into their, like, online journey, the more complicated everything gets, it's like for them, it's like walking in quicksand, you know, so when we, when we talked to one of the founders there, and we were talking about like performance, and he just stopped us and said that, you know, this is this is this is not the biggest issue for us right now. The biggest issue is that I wanted them to deploy like a really simple feature, just a general feature that you as a on the commercial side take for granted. My developers have been working on it for like six weeks, six weeks, and that's that that's something that's restricting us from doing or executing our plan. Everything for them is about the customer experience. We want to try to improve the customer's experience everyday in everything from the visual stuff, how they can communicate with their customers and everything. That's what we want to focus on and we can't have a solution where it takes us six weeks to deploy a simple feature. So we're trying to like the value add where they can, we try to simplify everything, they can move faster. Everything takes less time, you're trying to decouple everything and make it much more easy. In that particular case, they want to try to focus on the front end, on the customer experience, and we let them do that, and we take care of the rest. So I think that it loops back to like the time to market faster to market with new features.

Piotr Karwatka  20:40  

Absolutely makes so much sense. My last question is on the product strategy. I mean, which features can we expect in the foreseeable future?

Kristian Tysander  20:50  

Yeah, yeah, we have in the beginning, we were mainly driven by customer feedback. So we have a product roadmap that consists of the most prioritized features, from some of our larger clients. Now we're trying to stay one step ahead, really looking into more advanced things like subscription management. That's something we want to try to focus on. And we also want to focus on building a much more advanced system like price, runes, logic, graphical user interface, where customers can compose campaigns in a better way, in a much in a free way where they're not restricted as they are traditionally. So we want to give them a larger freedom in creating campaigns.

Piotr Karwatka  21:45  

So getting features on the things Yeah,

Kristian Tysander  21:49  

but we stay strict within our scope, in terms of price rules, logic. We've never tried to compete with any type of CRM or email marketing system in that sense. So it's only price with logic, and subscription management, that's, I think that's something that's really going to boom, or is it booming right now. And it's going to get even bigger. Yeah, and I think we see a glitch in the market there. You know, you have your payment providers, a lot of them can do recurring billing, but most of them don't do subscription management at the moment. So you have a lot of players popping up like chargebee, and there's just a bunch of them, like trying to tap a piece of the market there. So I think we're gonna see some competition between subscription management, providers and payment providers. And I also think that more and more platform providers are going to try to get in on that, because that's something that's going to grow and be really, really big.

Piotr Karwatka  23:04  

That's interesting. Okay, that was my last question. Christian, thank you very much for your time and for conversation.

Kristian Tysander  23:11  

Thank you very much for having me. It was a pleasure.

60. Catch the Tornado with Kristian Tysander

Continue Reading