#170 Miguel Pinho, Seedcamp - Part II

0:07Hi, and welcome back to the European VC, the go-to podcast for everything European VC. If you love the show, share with your friends and join our newsletter at eu.vc. This is the second part of Miguel's episode. Today, we will explore the common misconceptions about the VC tech stack and check his top tips for emerging VCs building their tech stack. Miguel is the head of technology at Seedcamp, the European seed fund launched in 2007 that helps European entrepreneurs to compete on a global scale. He works on internal and public phasing projects, ensuring they leverage the right tool suite to continue to identify, invest and support the most exciting companies out there.

0:55If you enjoy our content, do support us by hitting the follow button, giving us a review and following the European VC on LinkedIn.

1:11in. Let's transition into our segment about common misconceptions. And I would ask you to dive into the first one, which is FOMO and hurt mentality in VC. You have a perspective there that I think is super relevant to bring to the audience. One of the stories I hear a lot is we should use a specific tool because someone else is using it right and i think that that yes it is important and okay this is not oh we're so independent it's not a topic this is not we're so independent we think about ourselves no it is important to understand what the industry uses and we take that as an information point right it's an information point it's a data point on what the industry is using and we should look at those and take them seriously so that's valid 100 % The part where it becomes invalid, to my view, is that is the reason why we should use them, because the industry uses them.

2:06We live in an industry where FOMO is very useful. It's very useful for investors. It's very useful for founders. And leveraging that to your advantage is very useful and good for the industry. I would dare to put the word good behind it. So I think it is good. I just don't think that that should play the key role when selecting your tech stack. So those are two very different things. When you're selecting the tools you're meant to use, you need to be pragmatic and objective and engineering minded. And FOMO should play a very small role to help you identify the tools that the industry use, but not justify you using such tools.

2:44There are so many articles being written on tech stack. There's even full sub stacks dedicated to the VC tech stack and so on. And I'm curious to hear your take on, so it's not the media's coverage, right? but the VC industry's obsession with the tech stack and also the way that you sometimes then end up having specific solutions being glorified. I'd actually say that maybe there's one that you just described as being pivotal to you, which is Airtable, which for sure, I guess two years ago or so, really was all the rage in VC, LinkedIn, Twitter stuff you're in Europe, right? Everyone talking about, you won't need to use Airtable to manage blah, blah, blah.

3:26about right obviously you'd be one of the the fan guys of that proposition are there places where you're seeing that we're being overly hurt mentality driven on the tech stack in many ways i'm quite excited about this entire newcom buzz around the tech stack for vc if you look at it like even when i did the first report of the vc tech stack one of the reasons why i did it was because there was fuck all online to help people you know i could say it in a smarter way but it's Like there was very few articles. There was a few articles in the past from really well-known investors that had been open about it.

4:02And that was it. But some of them were like three years, five years, six years old. And tech does evolve. And I thought that it deserved proper attention and it deserved someone taking the insane amount of time required to just map out what's in there, building the questions, grouping tools into the ones that are required. So that was a big exercise that benefited me personally in my day-to-day job. but also the end report readers right you know so i'm very happy essentially the way i'm trying to say is i'm very happy to see all this buzz around the vc tech stack i think that that shows that you know like to your earlier question about like should funds have people you know looking into their vc tech i think this hints that people are taking it more seriously and spending more time and i'm happy about that world where vcs actually use tech more and they take it more seriously and they spend more time looking into the tools that they use and how they operate and optimize.

4:55So I think that's very good. I think that is actually the perfect transition into the next common misconception, which is you've said it before as well, but I think that there's something there that needs diving deeper on, which is you're saying tech stack is not a magical bullet that people tend to think that it is. And I would then add the part, which is what you're really saying here is there's too little deep thinking around the tech stack and there's too little of thinking of your tech stack from a product angle. And I'd love you to dive a bit deeper into that and share with our audience how you think about the tech stack from a product angle.

5:32I think that the tech stack can be your silver bullet, but you have to build it up. No one external can come in for like half a day and tell you these are the three tools you should use and that person leaves and that was it you now have a silver bullet that will magically solve most of your problems right and kind of going to the whole crm thing which is something that everybody speaks about all the time what crm should you use i think part of the problem is if you ask in a team what a crm does and who it serves you'll get different answers from every single person that's it you'll get the people that manage the founders you'll get the people that manage the lps you get the people that want and they will want different things they want different features.

6:12Then you get the marketing people and then you get the person that does social media and they will all want different things from a CRM, right? So what does a CRM do? Which CRM should I use? Whichever suits you best, but then you need to make a choice. And maybe it's not having one, maybe you need more than one. So you have this one tool that serves this customer and this part of the journey and fixes part of the problem and this other one that serves this other part and that's it and that's fine you just have to think through it i think the delusion and the part that that sometimes drives me crazy and i'm trying to stay well poised here i think it's the expectation that just choosing one tool will fix it all and that you know someone saying a combination of tools will fix it all because then that reverts back to the topic of the combination of tools that will be suggested by someone that doesn't know your business well is the latest trendy bunch of tools.

7:05Oh, these are the latest cool new shiny toys I've heard about. You should use them. You don't even know how that firm operates. Do they engage with their founders? How do they engage? Like literally, what do they do for them? Ideally, everyone just helps their founders a lot, but that's very utopic and magical, right? I'm saying like, how literally do they engage? Do they run events? Do they not run events? Do they do get togethers? Do they have an online community? Do they not have an online community? Do they do virtual events? Do they do hybrid events do they do events across one geography or multiple geographies all those are variables that you need to take into consideration into anything in your tech stack and and trying to dump that down drives me crazy right the part what i'm quite passionate about and i am passionate about is the dumpification of something that is so impactful that is your tech stack you can't you can't dump it down to be like just use the coolest new toy in the block what i mean no that makes no sense.

7:58It's like, I'm a cars guy, right? I love cars. I will not use a Lambo to go in the street, in the mountains, full with potholes. A Lambo is a great car. It doesn't mean it's not a great car, but you know, like just be context aware. One great car for a specific context is not necessarily a great car for another context. You need to know what it is you're going to do with it. And that's it. And I think the same applies to the tech stack. Just understand yourself and only you can do it. and it will require you it will require the people at the company to think through it no external person will come and answer those for you you need to put in the effort you need to put in the work i don't care what it is you know block a day a week block a week and quarter don't give a shit what it is but like you need to do the deep thinking you need to ask those questions there's no work around no no external entity will come and answer that for you so let me let me ask you question, Miguel, because on the one hand, it is a real pain point of our own company, UEC, right?

8:56But I also think some emerging GPs and earlier stage in terms of time they've been in the industry as a firm, teams might also have, which is I know very well internally where our tech stack lacks, right? And I know very well, what are the problems that I would, in an ideal world, like to see solved. But I don't have the skillset, right? Writing code or scripts or just understanding enough of, you know, the coding side of things or whatever you want to call it, to deploy it myself. But I am very able to give a brief. What I'm struggling with is, how should I go about this? Is this something that I will solve by trying to get someone in-house who's just going to add to my cost structure that I know in the beginning will add very little value?

9:36And then if we're super lucky and if things go really well, maybe in two, three years, if the company scales, this person will be a superstar, right? Huge risk here. Or should I try and invest hours and hours and hours and hours in finding a fucking service provider of a guy who does this as a service who knows nothing about venture by the way so i'll have to educate him on that right so it feels like it's it's really a tough decision to make in turn i'd love to hear your reflections on how you know these less and smaller teams could go about kind of dividing work and understanding you know the people running the investment shop running the operations on a daily basis need to do the deep thinking and come up with what problems need to be solved and ideally to some extent how but then deploying that is a very different thing here is the framework i would probably suggest the group of people the leadership of the fund of the company should get together and put together the deep work of identifying those problems clarifying them you know literally you know like there is a thing about like after five whys you find a real why or something like this Like find those real whys and what are the real problems you're trying to fix.

10:42And that needs to be the deep work that you do together as a group. Then if you have one person that is more technical in that group, then just make that person the leader of fixing those problems or trying to solve those problems. And that one person will have to do the deep work initially. I think getting external help can help you shortcut part of it. But I think there's like simple things that you can learn. If you're a technologist already, So if let's say, and this is based under the assumption that you have a leadership group that is say two, three, four people. And within that group, you have one person that is like highly technical or more technical prone or more engineering minded than the rest of the other.

11:22The others are like marketing and magical shit, which is very important. But like, but one is like, one is like very structured and this is me, right? Like my brain is, is extremely structured and organized. Like not only I'm an engineer, but I grew up in Switzerland. So you get both. It's like my brain when it comes to work, man, it's very organized. So if you have this one person that is more organized, then that person should become in a way like the internal champion of the tech stack. And it needs to be given those power. The rest of the leadership needs to understand and trust them over time.

11:55It's like, look, clearly this is not my vibe. You understand this better than me because we see that as part of your core skill set. and clearly you're going to do a better job than if we would try so you get allocated that time and you deep dive into that and i would say like in terms of things that that person would have to learn would be look at database design learn how to design databases so that they're useful to you look at automation tools look at no code and low code tools so that they can help you sort that out so that's in terms of what that person could learn the skill sets that would help them become very fast, much more useful for the team.

12:35If you have someone with a team that can design databases in a very long lasting way, that will reduce the little problems that you will have two years, three years down the line. That person should be the one designing them. That's it, the end, right? The others provide their requirements, but that person should be the one designing them. That's it, because they will just save you so much headaches and make that data set so much more useful. So I think the framework would be then to summarize, get their leadership together, find those whys and those problems as a group, then identify who's the highest tech skilled person within the organization that can do that deep thinking on the tech angle.

13:11That person, depending on their skillset, could benefit from learning database design, low code and automation tools and the tech stack. There's lots of resources online that person should be allocated time to do so because this will be very fucking time consuming. There is no shortcut around that. And I think that would be the steps I'd suggest. And once that person starts to solve some of the problems, if there is value in bringing external help, that person should be the point of contact with whoever comes in, right? Whoever comes in, you know, like as a junior, more junior person that will become amazing in three years time, that person that became the deep vertical expert in that, in the tech within the fund, because that will take a lot of time to get, right?

13:56Because you need to understand the values, the culture, where things come from. And that's why I'm saying it needs to be a leadership person in that role. There's no work around that because you'll have the holistic view about where the company is going. That someone that just comes in has no idea about the historic part, has no idea where you are, about some of the problems you're struggling, about why you can't do this, why you can't do that, why you can't do that. You'll just spend like the first few months will be like, no, you can't do this. No, you can't do that. No, you can't do this because of X.

14:24And the person will be like, okay, no, I get it. Like they're smart. They'll learn. But the first few months will be like you educating that new person on why they can't do X, half of the shit they want to do. I would suggest that framework around work and then bring in that person if you can afford to. Make sure you keep them as well. Don't let them go early on. Otherwise, you're essentially training them to go to the other company, do it better. So like make sure that whatever you bring will stay in for years, ideally. Like you said, there's very few VCs that have a tech person. I've met a few people in my role at the similar stage.

14:55Few, there's not that many. And usually like there is me's, but it's at PE later stage. I have a team of 60. I'm like, that's cute. We're not, we, the entire team is not 60. Like, so, you know, that's kind of a different level of, it's literally impossible, right? So I think, you know, if you go through that journey of identifying the key problems internally and creating that internal champion and bringing someone in, Make sure whoever you bring in has the incentives, has the buy-in, has the support for being there for a few years so that you can get that ROI on that person. Even objectively speaking, it's like literally you're going to turn and teach them so much.

15:34Like the point at which they pour knowledge onto your organization is what? After a year, a year and a half? So if they leave after two years, you got six months of them delivering at maximum speed and you want to keep them for longer. So if anyone's listening in with a 50 million euro fund or 75 million euro fund, look for someone with a tech person employed in a slightly smaller fund and then poach them.

16:01That's the hack if you're in that size. Okay, but let's get into the last. I will get in trouble, man. I will get in trouble. That's what I'm trying to do. So let's get into the last common misconception here, which is I want to ask you about the rise of the data-driven VC that we're seeing. Just as we spoke about all throughout this interview, really, a lot of people are paying more attention to this. But, you know, we're maybe seeing a few too many saying that they are super data-driven. Definitely, that's something that David and I see in our LP due diligence slash meeting new managers world.

16:40and I'm sure that you're also seeing it across the internet where you're kind of like, I wonder if that's just like you're seeing AI in every single startup that you're meeting. So Miguel, please take us into that whole space and tell us what you think about where we're at right now. Top level, I think it's very good that this industry starts to make more data-driven decisions and uses data to either just support and guide or to at some point even make the decisions themselves. And there's interesting about how Google even built that entire system that they ended up shutting down after a few years.

17:13But I think that the fact that the industry is using data more is very good. I'm very curious about how data-driven and many of those that claim that are, are, and where that data comes from, for example, right? I think there's lots of external data sources that you can pull information from, and there's a lot of value in there that tends to work better for later stage investors versus compared to SeedCamp at the stage we invest in, for example, from deal identification purposes at the stage we invest in. If it's like two people and they didn't give any public signals that they're building something, only so much you can try to guess.

17:52That's why deal sourcing and deal identification in the earliest stages of a venture is super difficult. It's difficult because there are no public signals or very few public signals. And I think if you go slightly later stage, then the public data set becomes more interesting. And there are quite a few tools that you can use to pull in that data. I am also very curious about how those companies build their own data set and for how long they've been building it. And do they have it? As in like, for real, like, do you have a data set? How big is it? For how long have you been building it? How did it come to be?

18:25And how big is your data science team? You have a data scientist, you have more. Like, do you have people assisting that person? Do you have an engineering team around them to help them build the stuff that they need to do magical stuff? Like, I'm very curious about that. I think it's very easy to say that you're a data-driven VC. Just the same way, it's very easy for a founder to say that they have AI, right? It's easy to say, but then you need to deep dive into a bit and to what they mean by that. Because, like, I think the answer is very different from investor to investor. And I'm not saying that there's just shades of gray.

18:57Some of them might be bluntly lying and they're like, no data-driven VC at all. That I'm sure is the case in some cases, but I think just it's about it being in the shade of grain. How much effort did they put into it? Which sources are they using? Internal data set, how they build it for how long, how did they essentially design all that? But I think overall, I think the trend is very good. I think that's going to become non-optional almost in a few years. If you fast forward a few years, I think having some level of data-driven decision or guidance will become non-optional. and you will need to have a data set that helps you guide you on your decisions.

19:33And that has to do with many things. I think the speed of innovation is also increasing so much and the amount of noise will become so big. I think if you look at the current macro, I think it's very exciting now. I think we're going to see so many companies pop up for the next few years. It's going to be amazing because you have all these really, really intelligent people that were really, really unluckily, you know, kind of kicked out from their companies that were doing magical shit at and they were doing a great job. But sadly, because of, you know, whatever macro climate shareholder pressure, they had to be fired.

20:04And those projects they had on the side paused. Now they're going to take this time to make them happen. And that's going to like the next few years in terms of innovation are going to be glorious. Like it's not even a question. Like so lots of great, great things are going. But what that translates into as well, this means more noise in the market. Right. You will need to use as an investor data and information to help you go through that noise, identify those good potential investments and invest on those, double down on the winners and all that. I just want to add a small comment, which is thank you for contributing to our R-rated status on Apple Podcasts by making everyone think about the shades of gray.

20:46but I love what you just said actually on a more serious note that innovation in the next couple of years if we're going to see your words being glorious I think that is a really, really nice way to end that kind of reflection there I actually just wanted to ask you one final thing before I pass it to David for the quickfire round and that is you wrote the go-to piece 2021 VC tech stack on your medium and I'm wondering are we going to see one for 2023? We are. I've decided to make the 2023. It's my own pain. I'm very, very excited about it. I think it's 100 % worth it. That's why I'll do it. And the reason why I said my own pain is because the amount of work behind it, the amount of hours required to make that go live, you never see it in the end result of any product, right?

21:35Because you see the final version, but this, any image, any report, any list is version 375, right? You see the final one. You assume that was, but I can tell you there's many versions halfway through. So it is going to happen if anyone wants to help, whether that's, you know, just participating in the forum. So this goes out as a survey to investors, essentially, so to VCs. And if you want to participate in the forum, ping me. My email is crazy2guestmiguelatseedcamp.com. And feel free to ping me if you want to participate in that way. If you have ideas, shoot. Happy to take feedback and thoughts.

22:13You have our unwavering commitment to helping you distribute that survey. So for sure, we'll do everything we can on that front. I appreciate that very much. Thank you. Yeah, David, take it away. Go to the quickfire, my friend. Awesome. So Miguel, we always end our episodes with a quickfire round. And today we're going to shift it up a bit. So we're not going to do a quickfire round. We're going to do a top tips round in a quickfire manner, right? Because we're running out of time. And what I will do, Miguel, is I'll echo a tip that has been initially and originally shared by you. And I'll ask you to comment on it in 30 to 60 seconds if possible.

22:48So let's see if we can do it. Let's give it a try. So top tips for emerging VCs building their tech steps. First one, learn to design databases. Definitely probably the most important thing you can do. If you can design a database, well, it will serve you for years. If you design a database poorly, your life will suck because you'll have to migrate data, redesign. You'll have data parts that you wish you had that data, but you didn't do it. Probably one of the most important things you can learn if you're building a data set, essentially, which we all are trying to. Second top tip for emerging VCs building the tech stack, which is think about scale from the onset.

23:30Don't think about where you are now. Think about where you're heading. I mean, this sounds very Yoda-ish, but think about where you're going in terms of portfolio size. Think about where you're going in terms of not only the current fund, but the next fund. Think about where you're heading and plan for that. And that applies to everything. So the databases, which we just spoke about, but also for the tools you're using. Make sure they serve you for at least, at least, at least the deployment of this fund. Ideally, if they can last for longer, even better. Which takes us to the third top tip for emerging VCs building a tech stack, which is reassess tech at least at every fund rate.

24:10I think that's a very important habit. I think the reason why I suggest that cycle is because it will feel natural. The reassessment of the tech stack never ends. And, you know, I could have picked arbitrarily, you know, one year, two years, whatever. It doesn't matter. But that one will feel very, very natural. If you think through things and you take a few months, you know, as you're gearing up towards raising the new fund and starting to deploy it, if you take the six months before to reassess the tech stack you're going to use for that fund, it will feel natural. It is the right time to have that clean sheet.

24:43At least by fund, you'll be able to look at that data set and it will be spot on for that fund for the entire duration of the fund, right? So I think that cycle will feel very, very natural versus something that is random and, and, you know, picked out of the blue. I just want to do a special thank you, Miguel, because this was the episode and we're going to put it out as a two-part episode. So we need to bring them together for this. This was the episode where we had more references to like pop knowledge and my brain was everywhere. So I thought about Katy Perry. I thought about Lord of the Rings.

25:16I thought about Chains of Grey. I thought about a Macklemore music called Glorious. I thought about Star Wars. It was really fun. Thank you for giving my brain a good time today. My pleasure. And more than anything, David thought about Fifty Shades of Grey. Let's leave it at that and say thank you, Miguel, for joining us. Thank you very much, guys. It was a pleasure. Thank you for listening to this episode of The European VC, the go-to podcast for everything European VC. If you love the show, share with your friends and join our newsletter at eu.vc.