Prior to ArK, Henrik was a partner at EQT Ventures - the Swedish venture capital fund that announced its first 566m Euro fund in 2016. At EQT Henrik was also leading EQT’s Motherbrain division. Motherbrain is the data platform that supports EQT to be truly data-driven in finding and assessing the best tech start-ups to invest in.
Before EQT, Henrik worked as the VP of analytics at Spotify during 2010-2016 when Spotify was one of the highest-growing scale-up in Europe. He is the mastermind behind Spotify’s award-winning Analytics team building it from the ground up and contributing to many major strategic initiatives to build the world’s largest music subscription business.
ArK recently closed a 165M Euro seed round of Equity and Debt capital - round led by
LocalGlobe with participation from with participation from CREANDUM and some prominent angels including Supercell CEO Ilkka Paananen, Zettle by PayPal founder Jacob de Geer, Embark Studios founder Patrick Söderlund, King founder Sebastian Knutsson, and EQT Ventures founding partner Hjalmar Winbladh.
In the words of Ted Persson from EQT Ventures - Henrik has made “analytics” his trademark, from Spotify to EQT Ventures and now Ark.
https://lnkd.in/dkdXCiVB 💃 Sometimes the data gives you the “correct answer” for the question you are looking for but it is not always the best solution for the users.
👁️ To implement successfully a data-driven culture you need to have the right i) mindset, ii) tools and iii) frameworks and continuously iterate.
Big thank you to Ted Persson that provided valuable insights into the characteristics that make Henrik an exceptional data-driven problem solver!
Spotify —> https://lnkd.in/dqnF5TUB
Apple —> https://lnkd.in/dcArQDbr
❤️ This was an excellent discussion on implementing a data-driven culture and the DIBBs method!
Transcript - not edited
data, spotify, data driven culture, build, founders, metrics, bets, henrik, decision, thought, company, framework, felt, create, people, insights, big, team, invest, products
Henrik Langren, Calin Fabri
Calin Fabri 00:05
Hey, guys, I'm just so excited about this episode. My guest today is Henrik Landgren, co founder and CPO of ArK capital, a precision financing company that empowers technology businesses to grow faster with non dilutive capital. They just closed a 165 million euro seed round of equity and debt capital round led by local globe crown Doom and some prominent angels. Prior to Ark Henrik was a partner at equity ventures the Swedish venture capital fund that announced its first 566 million euro fund back in 2016. At equity Henrik was also leading EQ DS murder brain division, murder brain being the data platform that supports EQ d to be truly data driven and finding and assessing the best tech startups to invest in. Prior to joining ATT Henrik worked as the VP of analytics at Spotify during 2010 2016 When Spotify was one of the highest grossing scale apps in Europe. He is the mastermind behind Spotify award winning analytics team from the ground up and contributed to many major strategic initiatives to build the world's largest music subscription business. In the words of third person from BKT. Henrik has made analytics his trademark from Spotify to equity ventures and now arc. During this episode, we discussed about data driven decision making and the dibbs framework, how to build a data driven culture, and why Henrik is called the nine minutes, man. Without further ado, please enjoy this excellent conversation with Henrik Lungren, co founder and CTO of our capital. 1 million to a million.
Mic check. Henrik, let's start with the story of DIBS. What it is, and how were you guys using it at Spotify.
Henrik Landgren 02:28
Alright, dibs. data insights believes bets. This is a framework that was developed at Spotify at the later part when I was there, which I think was great. But since this is now more than six years ago, I'm not sure exactly. I guess it has evolved quite a lot since then. But I think that the theory and concept around it is extremely useful. Even if you don't use the those exact words, I think the the gist of it is great to keep in mind. So data, insights, beliefs and bets. That's how you should think about any kind of decision making stuff that you want to do when you are evaluating what should we try to do here based on your prioritising different alternatives versus each others. And in that, the idea is that you will try to ground all your decisions into some kind of data at the very beginning. So data is facts is observations, it's you can debate whether they are true or not. It's just this these are facts. This is what data says. And then from there, you derive your insights of something conclusions from what you can see from that data. From those insights, you can form some kind of belief of something that you want to do something you want to improve, or change based on those insights from those beliefs, you then form and lots of different bets on what could we do here? What are actions that we could potentially do to act on those beliefs. I think this is a very useful framework to make to enforce yourself to not just make decisions based on your like gut instinct or something that you believe yourself, it's good to challenge yourself to make sure that you have grounded your IDs, and based on some kind of data and Intel. And if you can't answer that, then you should probably do more research to be able to prioritise your ideas in a data driven fashion. Because I think what we learned in Spotify during those years, when we started to get into this data driven culture was that most people really decisive good leaders, they're like, they have lots of ideas that they just want to push through because they believe so much in them. But when you start to challenge them and put numbers into, like, backwards engineer, why do they want to do that? And what is this based on? What are the insights in the data we can actually see many times that helped us to actually not do things that we first thought we wouldn't want to do? I'd say that we learned the hard way, how we got there. And I think that's why we created that kind of framework.
Calin Fabri 04:42
Limit. And do you have some examples when you kind of use steps so you got a sample of facts and then you create a some insights and a conclusion, and then you want it to improve or change something? And then you're like, Okay, what can we do? Which bet can we make,
Henrik Landgren 04:59
I think Given my examples are from the face, which then led to this framework being built, for example, the, it's this famous moment where we could notice in data that there was a lot of movement in terms of new users coming in that came. In the early days, they came in on the website using their laptops, and then they sign up through the service. And then they started use Spotify, eventually, they converted them become a fair user. And at that time, that was when you could use the mobile app. But more and more gradually, we saw in the data that a higher percentage of new users came in to that flow using their mobile phone. And what happened then was that we could find that we didn't have a product offering for them. Because the whole business model on Spotify was a freemium concept where you could use it on the adapter for free, but you had to pay to use the mobile version. But in data, we saw that that's now shifting, and we started to play like, how should we actually do this? Now, our whole business idea and how we price our products is based on the you pay for mobile. But now we see in data that the news are coming in direct straight to mobile, what should we do with this? So that's when we found this insights that Yeah, great. There's a high influx of or it's a shift on going here to mobile. First, it was a macro, Chanda we then we could very clearly see in our data, we then came to the belief here that we had to create a new offering on mobile, that could make our a new business model to work in this new mobile world, too. And then how should we capture that opportunity that will if that was what we didn't know. So this is where the concept of bets come in? Because I think we didn't know what should work here. We knew from data that this is happening. And we were pretty firm on this belief that there is going to be like a new world where mobile takes over. But how should this product be? We had lots of different ideas, we did a lot of research to figure out a set of things or products that we actually wanted to try. And then we decided that we need to bet on one of these different ideas. So we came up with a list of bets that we actually tried out first in the research phase, like more traditional research to figure out the space of what are different things that we could try. And then we actually did go into user testing on that. So we actually built several different IDs that we wanted to try. And then we got data on how those different ideas performed. And then we came up with, okay, so this is the data now that these different types of products are for mobile products, this is the data of those different alternatives, let's now make a bet on what we want to go with. So that was like a very data driven approach to like we observed something, we figured out that we need to change something, we didn't know exactly how. And then we try different alternatives. And we placed our bets on on a specific product, which then ended up being the shuffled product in the beginning. So you could have the first mobile product was that you could sign up through your phone, and then get a shuffled version of Spotify, where you can listen to shuffle music for free. But as soon as you wanted to play a specific track, you had to start paying. So that's kind of where we put that table.
Calin Fabri 07:55
Right. So if I understand correctly, also lay in the bed level at the battle level, you will still you will try to gather some sort of data to try to predict how successful those small ideas in the bet that you will try to do. Right,
Henrik Landgren 08:09
exactly. And that's kind of maybe one of the big things I take with me from thinking about that is that you shouldn't just call them like the sessions, we do this now. Instead, I think you should always honour that we don't know if this will work. It's a bet. So don't over invest your whole soul and energy and oil company on something because you don't really know yet if it's something completely new. That was really true for us. So we've backed on that product, and then we continue to gather data on how it works. And then we try to learn does it work? Or should we tweak it? Or should we change? It's a bet. And I think that's so rewarding, because that means that your look more sober at the analysis of the performance of that that decision that you made? Yes, I think when you don't do it like that, if you make more a decision like we do this now, the only thing you want to do afterwards is to show data to show that how good it is. But when you phrase it as a bet, I think everyone is more like Okay, did it work or not? We don't know, because it's a bet it's maybe it sounds silly, but it's has a big impact, I think if you form your decision making as bets
Calin Fabri 09:08
Kodesh. And how long does it take to run this process? So from data insights believes that
Henrik Landgren 09:15
there is no probably not a like one answer to that. Because if you're good at collecting data, and you have a data driven culture, you will all the time in your daily work, read new data points and you will find new insights or you refine your insights, you will figure out that some of the insights are not true anymore, because you have found other insights and then you need to revisit your beliefs and bets. But if you're really good at that, then you tried to build this whole cycle into some kind of strategic planning process for your company. We actually do review this and the team is like do actually explicitly write this is our strategy. This is based on these beliefs. At recurring frequency, you actually do revisit the data points to make sure that this belief still holds true that thus those bets still are valid. And then while you revisit everything, you will see that while these bets, they don't hold anymore, because this isn't work, and then you need to make new priorities based on that. So, you know, you do that in as part of your strategic cycle, which could be like it, you know, quarterly or half year or something like that, where you do revisit this, but feeding to this whole cycle is your everyday work of everyone to like, make sure that all the time learn from what you do after like the micro and strategy level and feed that into that Australian process. That's very interesting.
Calin Fabri 10:27
And you guys were doing that, like in 2010. So it was quite early oily. Now you hear a lot of Intrapreneurs being very data driven, or they said that they like to be data driven. And you mentioned data driven culture a couple of times. So I was wondering, like, how do you implement a data driven culture within a team, right, because you have at a strategic level, oh, we want to measure everything. And we want to use this framework to be better to allocate resources, on our beliefs, but then at the same time to measure it, and to review it after an interval of time to see if our beliefs still hold true. So how do you want implement a data driven culture? And then how do you go from the strategy level to the tactical level?
Henrik Landgren 11:08
Yeah, good. So I think you can only do with a certain country if you have your tooling rights. So the first thing is, of course, that you have to have the tools needed the infrastructure, so you have to collect the data in the best possible way. And then from that, you also have to have the information layer, right. So you can have your own information model about your company, whatever is important for you, that has to be defined somewhere. So you can talk about the same what you have one common language for how you look at your company. And then you use that to describe what is it that is important for your company, right now, what is the vision, you use the data in your information model to describe that, because then that helps to align the company and then like to focus on the same things using the same kind of definition, then that is really efficient to have, like, define that together, because then people don't have to argue all the time about all these different definitions. So all of that is the kind of the starting the fundamental building blocks are what you need. And then you need to build some kind of process where you use all of this to define that, that strategy. And then at any given cadence or process, you refine what you work on, what are you striving to achieve? What are these metrics? And how, what's the goal of these metrics? What do you bet on now that you want to improve, and then you connect those priorities to certain specific actions that you decide on. So we bet now that the or I believe is that the most important thing that we need to improve now is our onboarding funnel, big for mobile. And we have the belief that that is the shift that we're going to need to transform the company now into, and then we make these bets of these different products, and then the teams are, then they get empowered to do exactly that. But then they figure out themselves and try out different actions that they believe on their, within their respective scope of their teams, what they should do on their level to improve and to get to that to improve that metric. That's kind of how you link your overall strategic goals down to the day to day tasks of every single team, by empowering them not telling them what to do, but like the directions or where they should go towards a key part then from the data driven culture areas that done all the time everyone speaks the same language, you look at, you evaluate the right metrics, you know, what the goals are, what you should, what the target is to improve. And then each team can then try their own experiments, measure what I just did, did that improve or change this target metric or not. And then in the process, you will learn a lot. So in that you will learn like all the stuff that I thought, you know, I tried these different things, some of them did improve the metrics, most of the stuff you think will improve the metrics will not unfortunately, usually, sometimes, and most of the times, nothing really happens, you have you think that you know what you will do to improve the metric. But when you then read into it, you see that nothing really seemed to have changed here. So that's part of that learning process, then to actually then learn from that analysis to figure out, so when I did that, the metrics change like this. And that's how you learn. And then the key in the data driven culture is to then share that knowledge so that not everyone has to do the same thing. So part of the whole data driven culture is done to make sure that you have a good way of communicating and sharing what works and what didn't work. Because if you do that, well, your like knowledge or like innovation speed will go faster and faster, because then you can learn from what others have done. And then the whole key about all of this is to do this, this whole cycle quickly. And you know, the Agile thinking about invest as little as possible to get the feedback you need to know if this is working or not. Because if you can try many different things quickly, then, without spending so much resources or investments, then you will find what works much faster than if you like invest for during a longer period of time big projects and things like that. That's the combination of tools and These processes from strategic to the tactics. That's that's really how you are how you get to data driven organisation.
Calin Fabri 15:07
Got it? And also you, you mentioned tooling you started very, very early on as a VP of analytics at Spotify. How did the tooling change during the years? Like how did the toolbox change actually to use this framework for a data driven company? A lot,
Henrik Landgren 15:22
because when we started, we were in the very early days of this big data movements, it wasn't called big data at that time. But we were early on to have the technology that allowed us to store every click of every user, which then became very quickly millions and billions of rows of data. And the tools like really the data tools out there, were not capable of doing any any user friendly usage of that data. So it was very raw level. So we had to figure out what are the tools that we need, just first to get the whole thing working, but then later to actually get that into the decision process, like we just talked about. So we had to build a lot of things there ourselves, which was good, it was cumbersome, because it took a lot of time. And you would hope that you could get that from like an off the shelf. But that didn't exist at the time. So we had to build a lot of things ourselves. And but the good thing about that was that we could choose and do it in the way that we wanted it to be. But at the end of the day, we kind of created the whole data tools platform was the whole pipeline from, you know, building the data pipelines. So you could actually define how metrics should be transformed from raw clicks to metrics that are really useful at the end of it through lots of different steps, it was like a long tree of a very complex tree of data pipelines to get from raw level to useful metrics. And then that ended up in being exposed into different dashboards that the all the different teams could use to create their own specific dashboards using the common definitions or metrics and then use what is the most important thing for that specific team. So they could like tap into the pipeline or metrics and then create their own versions of that. So together, we like, we built the the metrics framework for real into the hands of every single team. But this was a long, long, long way. Because it's to get there, it was the combination of the technical evolution, but also the data driven process that we just mentioned, in the beginning, no one was used to this, of course, how we should use all this data, and the people are sceptical, also. But later, that's what kind of where we are at. And I've, I've heard now the tools are, of course, much better. And the this is, what the many of the providers out there in the echo system today, they have a, when you look at the tools that are now available, it's very much looks like what we created back in the days, but now they're available for everyone, which is great, right? And the I think Spotify is the biggest user of Google's Cloud Platform. Now, what we did then has helped a lot. It was not only us, of course, it's also the other, the other leaders and during from that time, a Google and Facebook or LinkedIn.
Calin Fabri 18:03
Yeah, it kind of seems that you it's not enough only to have the data but also the mindset first, and then the right frameworks to be able to communicate in the same way and then only like later comes to data, which you need to have a tonne of it. And you need to learn to try to make sense of it, and kind of connected with distraction and tactics. So you spent six years at Spotify during your high growth period, what are some of the learnings that stuck with you or your favourite projects there?
Henrik Landgren 18:29
It's a lot right. It was a super fun journey so much that I we learned and that I learned in terms of all the data stuff, but also like building teams recruiting, just being part of a journey to see like how fast things can grow. If you really focus in the way we did. It was a rare workplace at the time to be working at such a high fast growing company in Europe at the time. So lots of learnings that I've been, you know, keeping with me for a long time super valuable to have seen that. I think that is what really creates the there's waves of you know, the next generation of startups that comes from the first unicorn and then that's the cascading effects of that. And but favourite projects, i don't know i There's so many right. I think one of the fun that came to mind was the first time I personally had a code changed into the Spotify clients that I remember from like, I think it was like my first month or something. There was a What's New page on on Spotify at the time. When you open up the clients. It was like one stock page showing this is the new music that you probably want to listen to. It was called what's new. The problem was that all the time in, in the music industry, there is releases of old music that are re released because they change owner they like music rights are being traded, so they move between labels. And so from the technical perspective, it was a new track, but actually it was a really old track. So most of the music that was new was like all the classic music, more starters stuff like that. So either I mean, the start page isn't really appealing if it's just old, bad music on the What's New page, and we dubbed it as what's that the page? Because it was the most of the music from that artists. But I got why that happens. So, you know, if you looked at the release dates from every single track, it was new, even though the track itself was old. But then this is an interesting thing that what so what I suggested was that, let's use some kind of filter to that to say, That's show on the new track that we know has been played recently. So like the combination of popular tracks and the new tracks, that will be it make it more more interesting. And fifth draft the dead stuff that no one's listened to, even though they're immune? And then we came to an interesting debate, right? Because when I went to the developers and wanted to implement this, they were like, well, you know, I don't I'm not sure if I agree here, because this is a new track. And what you're suggesting here will introduce a lag. And that's true, right? My algorithm here would just say that we need to wait a couple of days to see what is being played. And then you lose a two, three days window of newness. We thought about that balance to say, Should we wait three days to see what is being played in favour for the What's New page to be actually interesting, versus having the real new stuff, which usually is there more than that artists mistake? And we did go for my solution there. So that's, I took the What's that page to become a real What's New page, even though there was a few days lag. But that was a fun first project that I was involved in.
Calin Fabri 21:25
sounds super interesting. And so I would love to actually see the debate between you ended developers, given that you guys both probably are very data driven. So you would use a lot of data argument, you know, like in the favour of actually having a really useful new page, then actually, to prioritise the newness, when actually that's no new.
Henrik Landgren 21:46
Exactly. It is quite, you know, a relevant topic that something that is like, in theory, correct, is not always what the users really want. And that debate I've had so many times throughout my career, I think, like, optimising machine learning models, yeah, this is a really good model. Now, it's much better. In what sense? Well, if you look at these metrics, and then we usually use some kind of a theoretical metric from the academia for how to improve a model, but at the end of it doesn't really help the user. But then maybe so if you look at the more user centric metric, you might optimise for the wrong thing. So think it's a very interesting point that I discovered, like my first days there.
Calin Fabri 22:23
So after Spotify, you went to ATT, you mentioned in a previous podcast that you had eight offers before you decided so you were very diligent in what you want to do next. What about the role at EQ? T got you excited? And made? You say, yes,
Henrik Landgren 22:39
yeah, I thought that was a really nice process to do that, even though it took a lot of time. But I think it was worth it. And I'm actually recommending everyone that I'm interviewing this thing that if you're about to change role, but forget about my role right now that if you think about changing role, take a step back, think about what you really want if you change roles, because it's always good to have that reflection, and not just jump straight into something new. Because you can. So that's what I did, because I thought that I'd learned so much from the Spotify experience. So a lot could be useful for many other industries or companies. And I thought, Should I now go for a big corporation, and like, try to use my newly and new skills from in analytics and the Spotify high growth pace to change and transform a big corporation. That was something that sounded interesting, or, like got around the risk factor was to go to start something new, or jump on a very early stage startup and build this, like it did right from the beginning, or go to a consultancy? Because that's where you go back to that because I've been at McKinsey in the past, right? And I thought with my skills that could be really good consultant, right. So that's kind of the spectra that I used. And then I act to really know what it would be like, I actually did go and do the interviews all the way to them saying, yeah, here's an offer for you. Because only then I could really get to the point I thought the place to figure out what is it that how will this be like so the big Corp track, for example, it sounded really good when I talked to the head, the person that was supposed to be becoming my boss, and in that case, but then when I started to interview with the next level of people that would work with, I felt immediately that even though I will know what I would want to do here to transform this company, or this corporation, I'm just never gonna get that through because they are so much in their heads in their own way of working. So I will just be super frustrated. This is just going to take a lot of energy from myself, dropped out workstream to go to the transform a big corporation, and then a consultancy I'd done before but I felt like it's too far I want to hands on is what I realised when I went down that track. I want to create something myself, start something my own. I tried a few different but then I felt it has to be a really grand vision. I want to do something really big. I came from Spotify high on my, you know, take over the world's ambitions. So I wanted to have something in which had really high ambitions were really like world level and then I couldn't really find a company with that level of ambitions at the time. So it kind of is Stop that works through to. And then I ended up with meeting the people who wanted to create eco adventures. Yamuna and the others, they wanted to create a new, like world class to run funds. And I really liked that. So that there I felt the ambition level is really top notch, I think I really believed I'm going with also what I could contribute with my data thinking was that yes, we can do something different. Here, I've met all the investors on the other side of the table at Spotify, I know that I can do a better investor, more informed a new tech driven in new ways. So I really felt like I can do this. And we can do something that is world class better than what is out there. And then I also got the combination of I could be like a coach or coach, others, but I'm also invested myself. So it's, it's not like a consultant where my business model is to like build slides, and slides and slides, and then you know, they won't make it. So I have a new job coming up next month, and I can do more flights and earn more money. So instead, actually, my incentives are invest in and read to make sure that they do well. So I don't need to do in flight tickets, just make sure to help them. And I kind of like that. And I also like the creation of I'm being part of creating a new from from the beginning, I can read a bit part of shipping it from the very starts, and of course, the data aspect of building a data platform, which then became Mother Brain. That's how I ended up there.
Calin Fabri 26:18
And tell us a little bit more about Mother Brain, like what is it? How did you go about actually solving the problem? In the beginning, building the MVP? What are the variables that actually you decided to have in the beginning, and I would assume that maybe in the end, it was human and machine collaboration, where it's like you would aid the machine to do all the stuff that actually it can do much better than a human. But then also you still have the human touch?
Henrik Landgren 26:42
Exactly, no, you're totally right on that. So in the beginning, it was the Mother Brain concept was already part of the pitch. When I interviewed for the job. The first version was built by a third party freelancing consultancy, which was cool. I thought when they when we talked about that that is probably just spreadsheets, you know, but it wasn't we will ask it was actually already when I started a like a real web app. So that was cool. And but in the beginning, the first versions of it was more like a research platform where you could like look up any company and see what data we had, which was very helpful. And like one place where we have have all data. But then, after the research phase, we started to use it more and more and start to think about what is it that we look for in every investment decision? And then all the time trying to see how can we make sure we have data or what we're looking for in the investment decisions into the platform? And then how can we make those aided by algorithms. So one by one, we try to get more information about that team, like what's the team behind every company, their growth rates to investors, I think a lot, it's very important to build a platform like that. We're very close between the people who are building it, and the people are making the actual decisions out of it. It was a partner but making actual investment decisions and also a product owner of the platform, which helps a lot because then we can focus all the time on doing like really making sure that we focus on what is helping our investment decision become better and better. And in that we then built more and more process around it, where we, as you say, involve the human in the loop a lot more. So we understood that if we need to have data points of what we are interested in, because we started off to be like rules. So companies, we should look for statistical rules, but we then realised that we need more than we need more data points to make it better. And then we figured out that if we actually use it as a complete CRM, so if we use not the brain for everything we do, we gather data indirectly from all the work we do. And then we can use that data to train algorithms to help us to see the patterns and what we like and then suggest new companies to it. Much like you know, Spotify recommendation engines work. And that was a key thing to do exist that took it from like being a research tool, which you use on the side, but actually one platform to do everything you did. Because that also kept you in that platform all the time. So we can move more and more of your actually, you know, investment decision work into the platform. So all the graphs, you need to make your decision in there, all the notes, everything is there. Because all of that also in turn generated more data. So that's how it evolved. And then throughout those years, we went from doing rule based to to like statistical models to machine learning driven models. And all the time measuring the accuracy or how well we could predict all the different rounds that we wanted to see out there ahead of the time. And we use that as the golden goal of what we wanted to improve all the time. And the more data we got there. And the better more work we put on into the algorithms, the better the those accuracy metrics became. So can
Calin Fabri 29:44
I ask like, what were the accuracy metrics? And also, as a follow on like, did you start using any subjective opinion, for example, someone talking with the intrapreneur and let's say high energy, low energy, something that maybe sometimes is not very quantitative. If but as you use as the CRM, yeah,
Henrik Landgren 30:03
exactly. So there, there's two stages of the investment decision is first to find the company. And after you have engaged with them, you get access to first party data for like data from the company, but also data from us as people assessing the deal. And we know that the investment decision has to be a combination of factual data driven analysis, plus also judgments on like, what we believe about the team, and also the sector and everything like that. So it's has to be a combination of that. Mother brand was built primarily for finding the companies first, but more and more also helped us to go into and help eating. After we got like in contact with them, we got access to the first party data, too. So we did build like Team assessments features, too. So like, as soon as you had left meeting with a founder, it came a push notification to your phone and said, which, how was that meeting? What do you think about this founder? What do you think about the end, and we had like a structure framework for how to rate the teams using certain dimensions that we had created ourselves, things like that, to help to aid that process, too, even though we didn't use like third party that but still tried to make us as data driven as possible. Even in that evaluation, too. It's really the end to end platform, there is so much in that investment decision, that if you use data, you will make better decisions in the end, but also respect and that it has to be combination of judgments and data.
Calin Fabri 31:29
Got it. Got it. Love it. And As Ted mentioned, you've made your analytics, your trademark to remember today, when you decided to leave equity and joined the team at ARC, what was the debate in your head?
Henrik Landgren 31:42
Yeah, that's, in one way, a very tricky decision. And another way not because as I learned, I like building things I want to build new things from the beginning, I want to build the the analytics team on the build Spotify, I want to build the new ventures fund, I want to build my brain like I love that early phase where you are, anything is possible and you like it's up to us to create something from the beginning. And then I'm less and less interested and less and less energised after some time has gone on, you see it's working, you know, now it's more about, you know, using it and like leveraging the platform that we built, then I'm less engaged, I've realised I can track that back to the childhood when playing with Lego, I thought it was super fun to build. But I was never the guy who played but it's which is, I've seen that people are different. They're like the building parts. After I think it was the same like, you know, being a Spotify for five and a half years, I also kind of felt the same, I had a team of 75 people in the end, not as fun as the in the beginning, building the whole thing. So as you could see, when I had been there for five and a half years, I had that demon, like same feeling that mothering thought from his 25 people it works, we've got to like it's now a central piece of equity. We know exactly, you know, this is going to be huge. The next level is going to be to scale this up internally. And within Unity. That's amazing. But the teams are working, we have all the processes, like everything is rolling, it's just to continue in this phase now. Whereas I felt like want to build new things. And also, after seeing all the founders as a part of the investment process, I've felt tempted that I also think I want to try something myself, I think I do have what it takes at least on paper, I thought I should, with my background, if I would evaluate in theory myself. That sounds interesting. I would back myself,
Calin Fabri 33:35
tested the algorithm on yourself, like, oh, very high scoring.
Henrik Landgren 33:39
Exactly, I scored myself. So then I thought okay, I might actually I you know, I've got confidence that could actually be worth a try, maybe I can make it. And then I have all the time a lot of ideas of stuff I wanted to do. And then one of these ideas was about creating something and new platform for building really good analytics model in the way that it should do in the way that I've always been coaching people to do it. Build that like one time as a platform and give that to all founders and from that position give really good financing options to founders that I think they should like deserve. If you do the data rights. I felt that lots of founders that I met with, couldn't get the didn't match, like the VC mandate. So we couldn't invest as VCs, because the VC mandate is like, you have to invest in a power law fashion. So you can, you can only invest if you believe that upside is, you know, 100x or something like that. That's not all the time. And there's a lot of companies and founders that build great companies that we can't invest in as a VCs. So I wanted to create a new company that actually could fix that to finance many, many, many more founders and give them the opportunity to grow faster in the way that VC backed companies can do, and also give them the tools and for the companies to have VC backing. There is part of a big part of that that I think shouldn't have to take back the High dilution of VCs to give away their companies when their companies are actually growing predictably, if you do the analysis, right, all those cases, I felt like they should just get access to loans and then fund their predictable parts of their growth with with loans instead, and then, you know, focus their high risk money on really high risk. So I felt this is something big. And this is something that is lacking. And this is something very scalable, if you will, this, right, we can scale this to all of the world because a b2c startup in Sweden looks the same as one in France and the US and Asia and Africa. It's, it's extremely scalable if we make this right, and founders are everywhere, and like I was blown away by myself about that mission, that while we can make a very, very big impact here, this is probably the biggest impact I can have in my life. And much bigger than having people listen to music. I felt, yeah, this is something too big to not do. And I think also on paper, I have, like this summer, what I've done in my life is doing that. So I have to pursue this. But then I had to find someone who know the financial parts of that mission, because I didn't know much about creating a financial institution and the regulatory requirements and everything around that. So then, when I met with my two co founders, Oliver and Axel, they had exactly what I didn't have. So we complemented each other really well. And they had a similar idea, not as technically deep, but financially. So it was the perfect combination of combined joining our visions together. Yeah, after a lot of red wine during last year, I took the decision to jump ship from you can see even though it was super fun, and you know, we were very high performing. And as you can see, and, yeah, it's a very tough decision to leave. But an easy decision to start building the new company.
Calin Fabri 36:43
I mean, like, if you look at your CV or HR background, it almost like it tells you that you have to be here and to build Ark at almost all the learnings that you have, you know, until now it like all of those are for Ark. And also if you connect it with the tooling that we discussed a little bit of that actually it improved so much during the last years now every startup has so many tooling and all the finances are so well connected with all the metrics, you can have a high degree of predictability, actually on the churn and how well they're going to do in the future. It makes no sense that they can't raise capital. It's just like silly.
Henrik Landgren 37:17
Exactly. You really get it, you've come and join the team. No, but I think it's really true, right? I mean, the we know also that the best, most fast growing tech companies out there they are have developed really good ways of working and modelling to get there. But that's given. So if we just take are inspired by the leaders in every different segments, and we just bring that in house, we can give that away to founders immediately. So they get that from day one. And why shouldn't they actually, it's just hindering people. Why should they have to wait years and years and years until they get to a point where they get attractive enough so that they can hire the right people? Who knows what tools to use and how to use them? And that's, you know, I just want everyone to be able to have those kind of tools, as well. So yeah, I think it's very interesting mission here. And yeah, I think the first six months has been proving that focusing on building a prototype going out fundraising for it, all the interest we got from there, and now also haven't been able to attract really, really good people into the team and start to build the real platform, and show that the customers have been really, really high satisfaction to see the how this really now seems to work. And when founders see it, I think it's has proven the thesis over and over again, that they're like, Wow, I never learned so much about my company in this short amount of time. And things like that is what they say. So very pleased and happy and, and humble that it seems to work so far. But I also know from my business life that this is the honeymoon period, right? Just 30 days, and we're going to hit a lot of crisis moments. And but I'm a problem solvers actually enjoy that part. So I'm looking forward to that, too. So yeah.
Calin Fabri 38:56
So like talking about fundraising, you just made an announcement on LinkedIn. So where are you guys in the journey? And where are you heading next?
Henrik Landgren 39:03
Yeah, super proud about that. We raised 165 million euros seed rounds, which is a combination of equity and that which was super happy, because our vision to get to what we just described, you have to be deep and tech that build the best possible analytics engine, you also need to have financial innovation to make sure that you can convert insights from your analytics engine into financial products as well. Those kinds of financial products are not they don't exist today. So we need to create new ones, we have to have that parks. And the third step is we also need to have the right backend of funding ourselves so that we can actually offer these financial products at good attractive rates. So that's important. That's why we're so happy to have that large round now. So we have the combination of equity and debt together. Because this means that we can really fund all these different products that we now are going to are offering to our customers. This is the first step and we've as I mentioned, I joined this Six months ago, and so it's very early, right, we're just having the first customers up to the door and the first product seeing that they have light right now. And this is really just the beginning. So we are very happy now to be able to, to roll this out and really start to iterate on this product and towards our vision. And then hopefully, we'll later down the line, things will work out well. So we can, I don't know if we're going to raise another round later on down the line. But if we're going to get to our vision that we're talking about, it's very grand, and it's probably going to go even faster if we can raise more capital later down the line to
Calin Fabri 40:33
exciting, exciting times ahead. Just such a pleasure to talk to you, you are just so passionate about what you're doing. And also, you know what you did in the past with Spotify, and also it EGT. Also, Ted just gave a stellar review about yourself. He said that he misses you a lot at Ekiti. And something that he said he's action oriented a thinker for sure. But he gets frustrated with two more abstraction, or if things take too long time. He's one of those guys who opened the tool, and 1015 minutes into a meeting to start building. Were you always like this? Like, were you always very eager to build things. You also you come across as a thinker. So you would think a lot? How do you combine the thinking and the doing?
Henrik Landgren 41:19
Wow, good question. I think probably impatient is a good word. And I think fast and not but not all the time the right way. I'm probably creative in my way of thinking as I get a lot of ideas all the time. And the best way to to know if you're thinking in the right way is to try I think it's not worth it to just overspend time on thinking, because I usually think that we don't know the answer to it. So I think it's mostly a waste of time to overthink a lot of things. And I think that's what happens in many of the meetings that I've been in the past with, sometimes that you'd like have workshop hours and hours and hours, but you feel like we don't know. So we're not going to get further we can get we're getting further only in the conceptual layer. But why don't we just try things since that would take us faster to understanding which direction is the right one. And of course, that's both good and bad, I think a bit more chaotic. And you might miss something sometimes, but on the other hand, you will also get faster to new ways. That's experimental and iterative thinker. That's what an impatient that's what makes me me, I think. And it's also fun to see, I want to hear that comment from Ted. Also thinking about reflecting on how I work now, with in many meetings now, I've been, we've been talking about the nine minute man here that I stopped meeting after that is being scheduled for one hour after nine minutes. And like came straight to the conclusion, like great done next meeting. And that comes from what I just mentioned, but also added now with the founders life where every minute almost counts, it's like stressful, right? We have so much opportunities of decisions that has to be taken strategies on like, where we should go in terms of like building tech platform recruiting, right people doing sales, you know, all sorts of decisions, so many of them, they just have to be made fast. Because if they take too long, it's going to take us much longer. And we know that the clock is ticking with, you know, when we raise money. So it's just so important for myself to all the time evaluates, should I spend more time now on this decision or should just make the decision and then move on to another? Because that will give me more value as a company or not. And I think that is what has pushed me now in this direction to be even faster in the decision making. Yeah, I think that makes me really high energy. And I think it's just I really enjoy it. And but I'm not sure if everyone around me. Probably a bit stressed
Calin Fabri 43:48
now you're five minutes, man.
Henrik Langren 43:50
Calin Fabri 43:53
What was the most challenging period in your career? And how did you overcome it?
Henrik Landgren 43:58
One challenge that I had that I resolved quickly was back in the days when I before I came to McKinsey, it was straight after school. I go back that early. That's the first thing that came to mind. Then I had my background from I've been coding since I was a kid. Then I wanted to move away from Tech because I wanted to learn more about the business side how the hell does that work. And then after four and a half years of studies and Masters of Science, I wanted to really get like into business. So I just picked Accenture and one of those quick decisions like yeah, just pick one of the management consultancy firms and then go for it typically made, didn't do much research on that. And then I came to a project at Accenture, which was really a large scale project for two years. My role was given was very technical. So I was like a project leader of a technical team, which was very, very far from what I wanted to learn. So it was completely not at all what I wanted. And then I was like shit that I'm stuck. I was feeling really bad about that, like, I got now going to do this for years now being this person that I don't want to be I want to learn more about the business side and decision making strategy and fast moving things. I actually the second day on my, my new job, I applied to McKinsey instead, after five interviews in one day, I got that offer. And then I moved on. That was a shocking part of my career for myself, because I remember like physically feeling like really anxious, like shit, what did I do here, really to make the wrong decision. But looking back on that, I think, you know, essentially, is a great company too. And I did not express what I wanted myself, cuz I just, you know, picked the brand and went for it, because Accenture has like a really good strategy department suit. And they just took me from my CV, oh, he can code, let's put them into a development project, which, you know, I could do that very well. And I did too, during those months there was there and before I switched, but it was a Yeah, interesting part of my life. And I think the sense morale for myself has been from that moment, I will always, always listen to you know what, if you're happy with what you do every day, you should continue to do it. If you feel like you're not getting 100% of pleasure out of your day to day job, you should, you know, do take a step back, think about what you should, if you should switch jobs. I mean, life is too short to have a job which you don't really like, you should always act on that feeling,
Calin Fabri 46:23
act and feeling. So what is the best and worst advice that you have received?
Henrik Landgren 46:27
Someone told me in recruiting that you can't always recruit the people that are like you, which is very true, right? I want it's so much better when you have a diverse and rich team from different backgrounds, which is amazing. I've heard that being coached to me sometimes in interviews, processes during my career where my favourite feedback has been, you know, as you can hear I'm I'm very like energetic person. And I love working with people that are also high on energy that you feel I get energy back from. And some people that are really good, highly skilled, but they don't have this energy. I've been saying like, Yeah, this is a great great candidates, but I'm not sure if this person doesn't have my the energy level that I have felt like we should probably find another one. And then the coaching has been no, you should not find the person that are just like yourself. And I've been listening to that. And but now when we recruit our team, now, we actually hire people that have all of them high energy, and are all of them very positive. And now I feel like what this is like, and I could just feel like that coaching to my past were has been really bad. I should always have listened to myself and hired people that do have high energy that actually match what I like to work with. That's an advice that I have got that I have not realised it has not been the right one for me. Because that would be one example of that. Well, good advice, there is a lot of good advice all the time, I think the people that were with me, supported me to say that it's worth it, you should follow your passion here to move on. Even though the moving from a big corporation, like you could see that where you're very successful, and have a great trajectory, drop all of that, and then move on to this high risk world of startups and the people that pushed me to do this and that were supportive. I think that was a very, very nice and good coaching from them. Because the I think the most standardised way of advising a person like me, that position would be of course you should stay on you should find ways of you know, achieving what you want. Within that you have a great track record a great career, great everything. But I think people that really understood what I really wanted, they were supportive of this move. And I'm thankful for that, of course,
Calin Fabri 48:35
funny to see like how the best advice actually ends up being people telling you to believe in yourself. So then in the end, the advice is actually believe in whatever you think is the best thing to do instead of, you know, coaching you to do something that it comes natural to you like hiring someone that you just don't feel that it's the right energy that you could work together with.
Henrik Landgren 48:58
Yeah, that's a good reflection on that. Yeah.
Calin Fabri 49:02
So most thoughtful investor and or intrapreneur that you have met and admire. Well, then
Henrik Landgren 49:07
I have to lift up I think both Ted that you have spoken to before this and also Jamar bindaas, who was the founder, one of the founders of Aikido ventures, like The Godfather. He's also an investor in Arc now has been extremely inspirational person and really good investor, specifically early stage taught me a lot lots of learnings from him.
Calin Fabri 49:31
Got it? And the last question your advice for founders listening to this episode that they want to implement a more data driven culture.
Henrik Landgren 49:41
There are two dimensions we spoke about that before of being data driven. One is to have to invest to make sure that you have the tools, but the other is to work in a data driven way, but you're humble. Define what you really want and then evaluate your actions using that framework in a truthful way. to overcome your own biases towards yourself. So I think those two dimensions are really good to think about. It's not that you only need the tools, you can actually be very data driven without any tools. People are like, truthfully, always trying to figure out, did it work or not? What is the impact on whatever metric I have, even if it's simple metrics, that's one dimension. The other is the tooling. So invest in the right tools that helps you to be driven, but probably even more importantly, think about yourself, how you make decisions, how we evaluate and how we learn from it. That's equally important. I
Calin Fabri 50:30
love it. Thanks a lot for taking the time. Henrik. It was such a pleasure. Thank you very much for sharing all the experience with us and look forward, you know, to see what you guys are at our car building. It sounds that it's a very exciting future ahead.
Henrik Landgren 50:44
Thank you so much for having me. Great questions. Thank you. Thanks.