All Colors vraagt door - Stijn Christiaens - Co-founder & Chief Data Citizen bij Collibra

Thanks, Christof.

I think they'd probably say something in software, and more specifically around data. But if you asked them to really explain it, I don't know if they'd get very far.

I think it's also part of our problem that it's very hard to explain simply. And of course, because it keeps changing. But we're in a fairly new domain in a way. We had a business strategy of category creation, which means you're constantly trying to explain your position in the market. We've been doing it for a long time, and for example, because of AI, new needs are emerging, so you have to adapt a bit.

But if I try to explain it, I have two ways. One is our official way, and two is an unofficial way I learned that actually works better for people to understand and remember. So the official way is about data as an asset. Every asset in a large organization—our market is enterprise—has a system of record. For example, the money asset has the ERP, the pipeline asset has the CRM, and then the data asset is actually a fairly new asset for most organizations. They've been doing data for ages, but they've never really managed it as an asset. And that's now a rising trend in the market. So it also needs a system of record. That means it's a place in the organization where that asset is registered, but where responsibilities are also defined. For example, who is responsible for that asset to organize the process of access, but also many other processes and responsibilities. And those often vary from organization to organization. And we are that player in the market. So the system of record for your data and now also AI assets. We call that unified governance. So that's a bit on a conceptual level what we do.

And when you explain that, people get it, but five minutes later they've probably forgotten. So that's why, when I'm in front of an unfamiliar audience and I don't know how well they know our market, I take a much more generic explanation. And it goes like this. It's a bit of a game. So you have to play along, Christof, if that's okay. And I always say, let's make it a bet. So in 30 seconds, I'm going to explain it to you. And the clock has already started.

Okay. So I bet you €1,000 that it's easier for you to buy a book online than to find data in your own company.

Yes, exactly. Because then you see people start thinking: 'Ah yes, finding it, what is that actually? And who should I go to to ask about that?' And for customer data, for example. And then you see them thinking, which company should I go to, and then you see them going to Slack in their head, or Teams, or bothering someone at the coffee machine. And then you actually see people realizing there's very little organized for that.

And what we do then, in that metaphor at Collibra, we are that sort of Amazon experience or bol.com experience for data in a large organization. So you have a place where people can start that search. But then also the whole baggage that comes with it. Because that data needs to have responsibilities with an owner. There needs to be a process organized for how people can request access or how they can register new data products. So we do all that baggage. But it's very easy for people to remember that we are a sort of marketplace for data assets in a large organization. That's a concept people can grasp. But to translate that practically into the large organization, that's of course still a big conceptual step. But everyone understands a sort of Amazon, bol.com marketplace for data. And that's the role we play, and much more, of course.

Correct. Yes, a deliberate word or an important word in that entire title. We had a deliberate word for our entire culture and our proposition in the market itself, I would say. I'll get to my role and that title, because it's my third role in the company. But within Collibra, we believe that the boss of data, the Chief Data Officer or the organization as a whole, they don't succeed necessarily by having the best data or the most data, but they succeed by winning the hearts and minds of what we call the data citizens in the organization.

And a data citizen is a person. We define that as someone who needs data to do their job. And especially in this AI reality that is unfolding over the world, it's been three years but it will continue for a number of years, data is becoming more important for almost everyone in their job. So in that sense, we see almost the entire employee population in an organization as data citizens. Whether they know it or not is another story. And so we believe that an organization is successful with data if it wins the hearts and minds of all data citizens. So that's our promise to the market, to our customers. We're going to help your company find, create, and grow those data citizens, etc. And so when my role shifted a number of years ago, I was thinking, okay, how can I continue to add value in the changing Collibra? The organization is getting bigger. There are new needs in the market. What can I do as an individual in the company to add strategic value? Then I say: 'Okay, our customers have data offices. Our customers have a Chief Data Officer. So let's put our money where our mouth is. Let's eat our own dog food.' That's Monday morning. 'Let's drink our own champagne,' that's Friday evening. 'Let's practice what we preach.' So we set up our own data office. And then I was discussing with Felix, the co-founder and CEO, is that a good proposal? And then Felix said: 'Yes, okay, but then let's call it Chief Data Citizen.' To also make a statement instead of Chief Data Officer, we signal to the market, we believe the Chief Data Officer is successful if they win the hearts and minds of the data citizens. And that's why the Chief Data Citizen title.

That's going to be quite messy, I think. For example, this morning I arrived here from Paris on the first train because there was an event in Paris last night about European Sovereignty. That's a very popular topic for the last 12 or 18 months, let's say. So it's very messy. But in the role itself, there are a number of things that recur. We do all sorts of data things. We have data engineering, which is about pipelines, data warehouses, and source systems. It's about data products like reporting, for example, but also now AI. So there's a technical aspect to it. Another part of the role also has a human side. Because even within Collibra, we have to bring that data value to our colleagues. We have colleagues who joined yesterday, they don't necessarily know what Collibra does. They come from the broader market perhaps, but they also need to realize what this software actually does and how it will evolve. So there's a lot of internal communication involved, but also external. We go to the market and say: 'This is how we approach it.' And apart from that, in an organization of about 1,000 people, in my role, there's also a managerial responsibility. So we also have our management meetings or middle management meetings, workshops, etc. So the agenda is very varied and changes from week to week sometimes.

Well, we've been lucky that we've been working on it for a while. We're almost 18 years old now. We started in 2008. So we've been lucky that we've gone through each of those phases of the organization. You can't really see Collibra as one company. There are different generations of that company. The generation of the four founders in 2008. The Collibra of 20 people, the Collibra of 50, 200, 400, etc. And through all those phases, we've always known that we have to set out a vision, communicate it, align, iterate, etc. Yes, that requires work. With four people, it also requires work, but then it's around the table, so to speak. That's still work, because sometimes people are away for three weeks on sales trips, then come back. So you have to align your perspectives, of course. But that's still fairly easy. But then you have to do it with 20 and 50. So through all those generations, let's say, we've always had to practice getting that vision sharp, communicating it, repetitively aligning, and repeating it a lot, and iterating on it, of course. So we still do that today. For example, today there's clearly a lot of change in the market. For example, on sovereignty, but also on AI. The gorilla in the room, so to speak. Then we see that we also have to repeatedly sharpen our focus there. What do we want to be as Collibra in that changing market? What position do we want to take? What strategy do we want to set out? What direction do we want to go? So that requires thought work, alignment, communication. And that requires getting it sharp in a story, in a number of slides. And then that requires repetition internally, but also to the market, to customers, etc., but also internally at all-hands meetings where the whole company can be present. Which is already an issue over time zones, because there's actually no time zone where the whole company can really be present. So some people then have to watch the recording or the follow-up email or the intranet pages, whatever. But then you also have that repetition, the repetition. And then you also have the further iteration on it, the elaboration of a higher-level vision into, yes, certain parts of the company. For example, last week there was a product workshop, product and engineering leadership in Prague, and then you have to try to refine that higher-level vision over two or three days into a more practical level. So that requires a lot of effort.

Yes, absolutely. Yes. It's enormously different. Many things are the same. You still have a lot of problems, challenges, opportunities to tackle. But ultimately, many things are significantly different. So my argument is always, if you do a startup, but also in a larger organization, if you don't grow, then your business can't grow with you. Whether it's your own business or the department you work for, you have to grow with it. And that starts with yourself. It starts with the Growth Mindset, for example, which we also gave our organization as an exercise a number of years ago. Where we were sitting with 50 or 100 people in the room about the growth mindset. And then you see that many of your own experiences, you might have looked at them as, 'Yes, that was someone else's fault.' But actually, you can also look at it as, 'What can I do differently next time so it goes better?' But it actually started very early on, in 2008. We're a spin-off from the Vrije Universiteit Brussel, and yes, we were academics who ended up in a market that is much less forgiving, let's say, than the academic cocoon of the university. And we had to start changing immediately. We could no longer be the preaching professor, so to speak. We had to effectively learn more with open ears and eyes. What is the real need in the market here, and how can we respond to that? So for 18 years, we've had evolution and growth as individuals, I would say. Now, I'm going to be realistic about it. There are things you can grow into as a person, there are strengths you can develop, there are weaknesses you can mitigate. But ultimately, you're never going to completely solve your weaknesses. So you can grow in them and have an awareness of what are my strengths, what are my weaknesses, where do I run into myself? And ultimately, you'll still ensure through good teamwork choices that your strengths are amplified and your weaknesses are actually covered for a part in the team. Because some of your own weaknesses, no matter how much you want to grow, will always continue to exist in yourself.

No, exactly. So when I speak about preaching, I don't want to make it about religion or anything. It's really about evangelizing your proposition, let's say. So in the very beginning, it was very academic. We did research on semantics, which is now gaining importance again because of AI. And then, yes, we were actually not so much preaching about data to organizations, but about the niche of semantics in data, because we came from that angle, and we were really the semantic guys. So because of our passion, people sometimes listened to us. But the gap between what we were preaching and practice was also enormous. So it took us many, many years to get that technical proposition right. But it's true what you said, today we still have to preach a bit. To such an extent that at a certain point, I noticed in myself that I was getting tired of repeating myself a bit, if I may say so. And I still repeat myself a lot in the market. There's still a lot of the same message needed, let's say. So that awareness still needs to grow, and I'll come back to that. But you see there, yes, we were so tired of repeating the same thing that we literally said, let's just record these things and let's share those recordings with the market. Because it's better for us and for our business. The rising tide raises all ships. So then we just set up university.collibra.com. Where we effectively share our knowledge, our learnings on our data governance topic with the market, and now also AI governance. Because we said, yes, we see that there's still very low maturity. And we still do that today. There's still a lot of awareness needed. For example, we have a collaboration with another Belgian startup, DataCamp, where an AI governance course was made in collaboration with Collibra. So it's also freely available because we realize, yes, those learnings are needed.

I'll try to sketch it for you, Christof. So if you look at other assets in the organization, the money asset, materials or sub-products as an asset, then you look at mature disciplines. You have natural systems of record like CRM and ERP systems that have been around for multiple decades. And the basis of those assets has also been around for multiple decades. So when you talk about a Chief Revenue Officer or a Chief Financial Officer, yes, those are roles with decades of experience. So you know very well what's expected of that role or what such a process should look like around those assets. For example, around money, it's the process of budgeting, accounting, or for us, expensing. So those processes are mature, they've developed over two, four decades in the market. If you look at the data asset, you're actually at a good 10, 15 years maybe that it's been running as an asset in the market. So in terms of maturity, it's actually still, let's say, kid leaving home for college level. So it's not a baby anymore, it's not a child anymore, but it's also not yet the mature professional with the mature discipline. I can sketch that with a number of examples. For example, the Chief Data Officer role. Gartner has done research on that. So they trace the evolution of that role. And then you see it's existed for a good 40 years. And when it started, around 2010, there were about 10 or 15 Chief Data Officers worldwide. And none of those people had done that role before, because that role didn't exist. So now there are over 150,000, 100,000. But many of those people who are starting today are still doing it for the first time. Some for the second or third time, but those are exceptions. So you're actually dealing with a fairly young discipline in the market. And that's also confirmed by many data maturity studies. So if you ask organizations how their data maturity is, you'll see they always score themselves quite low. There are pockets of maturity. I was, for example, on a podcast with Ruben from Colruyt, and he described it beautifully, because he plays that role within Colruyt. And he said: 'Yes, look, when I have to work with my internal clients, the business stakeholders, I actually have to meet them in their maturity where they are.' I'll give an example. For some, it might be, we are very AI advanced. We know SQL queries inside and out. And a report in Tableau or Power BI or Looker, no problem. Right? So they have a high data maturity. But that also means that that business stakeholder, the decision-maker, yes, will also process the insights in those reports and, for example, do analysis based on bias. Is there bias in your insight? Yes or no? So you can ask a bit more to think critically and say: 'Okay, now I'm going to translate that insight into this decision, into this action.' That requires maturity at that level too. It's not just the SQL query. But some people, then you can meet them with high maturity, but some people you then encounter in another part of the organization, they might say: 'Yes, what is that cell? For me, that's my Excel file.' And then you have to meet those people where they are. And when you then speak for our market to very large organizations, yes, that maturity curve is all over the map. And if you work with one group of data citizens and make them champions, then you have a group of people who have now increased their maturity. But they are also champions to take the next set of people who need to increase their maturity along.

I've been shocked by it. Yes. And I still get shocked by it, but I've also gotten a bit accustomed to it. So I actually expect that there will be pockets of maturity and also pockets of, yes, sometimes embarrassingly low maturity. That's a reality. And that's because it's not an accusation towards people or organizations. It's just a consequence of the young discipline, the field of sport. Treating data really as an asset is a fairly new phenomenon in the market.

If I were to say we only hit one wall, I would be doing injustice to the 999 other walls, of course. No, so we always focused on the importance of data and the importance of doing data the right way. That was also because we did research on it at the VUB. We were in a semantics technology and applications research laboratory, a computer science laboratory where we effectively did research on the technical aspects of data, but also on the social aspects of data, making meaning between people and between people and machines. So we always knew that the way data was handled in organizations, there were better possibilities. What we didn't know at that time, due to our youthful naivety, let's say, was how big the problem actually was and how we should position ourselves in the market to sustainably tackle the problem. Because in the very beginning in 2008, we had started with a technical proposition around semantics to do data better, which worked technically but not commercially. So we had to go into the market and search. Okay, but what is the proposition that has a commercial chance of success? Because we want to solve this problem in organizations, even if they don't know how big the problem actually is. So yes, let's say our first wall, one of our first walls we hit, was the pivot. Our proposition works technically, but not commercially. How do we solve that now? And then we had to make a bit of a pivot to find a proposition that worked commercially. We also kept in mind at that moment, I'm going to... we're going to just question people and see what works, what doesn't work, or is that literally diving into the void and starting to think about how can we fix this ourselves here? No, no. So I think we've done enough diving into the void and did enough of that at the university there too. But we still do that now. That means thinking strategically about your market, but that's only possible if you open your ears and eyes sufficiently. And that was also one of the lessons in the very beginning. So in the beginning, our mouth was too big and the preaching about semantics was too high, and our ears and eyes were too closed. But we learned those lessons in the very beginning and we still have to keep teaching them to ourselves and to the rest of the organization. Because ultimately, the customer's problem is still where it starts. Okay, but where is the problem? And in our proposition, it could be a technical problem. For example, an integration, a proxy, whatever. So that can be very basic, but it can also be extensive. Yes, we're starting a new data warehouse or this or that. How are we going to work with that? AI comes with a whole range of new technical challenges, of course. But listening also means that you can discover: 'Yes, the problem might be cultural at the customer's organization, organizational, even political sometimes.' I want to give an example of organizational. So when an organization starts its data program, it could be that three years later, the program can run perfectly, but the larger business does an M&A or is part of an M&A or has some business transformation. And what often happens there, in the technical world we all know that, of course, is that pendulum swing of centralization and decentralization. So that means there are changes there that are not technical but have technical consequences. But you also see other pendulum swings, of course, from cloud to cloud with AI, without AI, etc. So all those dynamics in the organization are things you have to develop sensors for, as an individual but also as an organization, to pinpoint them. Because our value proposition to the customer is, we want to understand your data problem and solve it together.

Daily. Yes, daily. So I'll give an example, just from last week, also in follow-up to the previous questions. That awareness, one of the things we recommend to our customers and the market also does itself is organizing data and AI days, for example. So that data office, the champions of data in the organization, which are typically a very scarce resource, by the way. So a small group of people in a very large organization, which means they can only be successful as a scarce resource by being focused and going in the right direction. But what they have to do is develop a sort of data and AI days where they then, with that team, a broader population of the company, a few hundred people, maybe even a few thousand, for one day, maybe even sometimes multiple days. Some people then do a data and AI week. And that's about creating awareness in the organization. What is data actually? What does AI actually do? What can we do with it? What are examples of what we do with it? What are vendors we partner with? And for example, last week, if you mean with your feet on the ground, I was at ASML in the Netherlands. They had such a data day. And the same in London at Schroders Investment Management. That's a FTSE 100 company. And they each had such a data day. Being present at that data day means you have your own presentation, but there are also booths, there are also various conversations, there are also other presentations from the customer themselves or other partners, like Snowflakes or Googles or whatever, Databricks and Microsofts, all sorts of Amazons. So that's feet on the ground. But there are also many other ways of feet on the ground. For example, later at the office there are all sorts of meetings about how we do our data program or which new AI features that just emerged yesterday we might need to evaluate if we can do something with them.

It's an organization of 1,000 people. So if it only came from me, that would be a problem, of course. So you try to make each of those people a sort of sensor on that changing playing field in a way. And so you try to open broad channels internally and externally with both your colleagues internally but also customers and partners. So you try to keep your ears and eyes open. Look at it yourself in other ways, but also through the eyes and ears of other people. And we have internal channels for sharing that information via a Slack message, for example, but also via monthly meetups, all sorts of things. So we do a lot of those knowledge-sharing sessions. And apart from that, I also try to keep my finger on the pulse. We see that the technology around AI, for example, is changing very fast. It's very visible and very loud. It's almost deafeningly loud. And it's also clear that compared to previous technology waves, and let's take the cloud for a moment, yes, there were many changes, but they weren't architecturally significant. Also not in a way. Data on-premise or data in the cloud is still architecturally quite similar. So that was also easier to understand. And that change was less fast. With AI, that change is noticeably faster and louder. So we're also definitely in a sort of hype cycle. There's the reality of the technology, but also a lot of noise. So being on board also means how can we get through that noise a bit to understand what applies to us and to our market and to our customers. And that's not everything. And there are two ways I handle that. So one way is, yes, sometimes you just have to get started with it yourself as an individual or as an organization. Okay, MCP is a new protocol. Yes, it still has all sorts of shortcomings. That's clear, because it's a new protocol. Maybe 18 months old or so. But we also have to do something with it. The demand in the market is such that we have to do something with it. So let's do it. And we know it's a protocol that's not ready, so there will be all sorts of problems with it, but let's bet on it. And then we'll see, one, two, three, ten quarters later, we have to bet further on it or not. Right? So some things you have to do with the chance that your experiment also turns out negative. Around MCP, I expect it will turn out positive, but the proof is in the pudding, of course, is eating it. And then on an individual level, it's also a matter of trying to filter that noise you receive with other sources, and then also thinking about, yes, okay, it may be that this new AI widget is sexy and automating like never before. But you also have to think, okay, how do I take that in as an individual, and how is the organization going to be able to take it in, and on a broader level, how is society going to be able to take it in? Because it's not because there's a sexy new technology widget somewhere that does magical things—and there's no magic, of course—that you can suddenly adopt it quickly. Even much slower your organization, and even much slower society. For example, let's take the LLM as a widget. Yes, actually it's like saying, here's Stijn or here's Collibra, the organization, or here's Belgium, society. Here's an engine. Do something with it. Yes, okay. I can't ride on an engine. I have to build a motorcycle around it, or build a car around it, or build a tractor around it, or build an airplane around it. I don't know. But with the engine alone, I can't do it. So I have to get started with that engine. And that means I have to set up the machine, the agent around it in this case, yes, we still have to set those up. We still have to teach them to our people. And when that agent, that engine, that car or whatever starts to sputter, how can we decide if it's too slow...

Eh ja, dat helpt, want voor ons is het zo: we hebben het voordeel, Felix en ik, dus twee co-founders die nog altijd aanwezig zijn in het bedrijf, wat toch redelijk uniek is in deze VC-backed snelle wereld, laten we zeggen. We hebben het voordeel dat we zowel het perspectief als de ervaring hebben van al een paar keer door zo'n platform shift gegaan te zijn.

Platform shifts zijn van die, ja, dat is een fenomeen dat is beschreven door een analist, Ben Evans, goede analist. En die ziet dat eigenlijk als vijf of tienjarige golven die door de technologiesector gaan. Dus bijvoorbeeld de mainframe als een platform en die is dan geshift naar de desktop computer en die is dan geshift naar web en open source. En dan hebben we de shift gehad van mobile en cloud. En nu hebben we de shift van AI. Dus dat zijn telkens vijf of tienjarige technologiegolven.

Die oude dingen gaan niet weg, maar het individu, de organisatie en de maatschappij gaat wel inzetten op die nieuwe golf. Dus je kunt als business wel plannen op hoeveel moeten we inzetten op verandering versus hoeveel van die oude dingen gaan er nog blijven. Want die situatie is onze realiteit. Wij kunnen onze software niet alleen maar verkopen met het nieuwe AI stuff, want onze klanten zitten nog altijd met big data platforms van 10 jaar geleden of data warehouses van 20 jaar geleden of mainframes van 30 of 40 jaar geleden. En die hebben ook data nodig. Dat wil niet zeggen dat wij die als bedrijf allemaal evenveel moeten prioritiseren, maar dat is wel een realiteit waar we rekening mee moeten houden.

Dus wij, Felix en ik, hebben het geluk dat we die golven al twee of drie keer hebben moeten meemaken in onze organisatie. En dus dat we nu kunnen zien: oké, wat betekent dat dan vandaag, de afgelopen drie jaar, maar ook de komende drie jaar voor Collibra, wat moeten wij dan precies doen qua aanpak, qua strategie, qua execution om op die golf mee te surfen?

Ja, klopt. Ja. Dus uw vraag is hoe overtuig je die mensen, hoe krijg je die mee in dat verhaal? Want dat gaat ook niet altijd rozengeur en manenschijn zijn. Nee, nee, maar die zouden dat moeten weten, hè. Dus VC is eigenlijk een categorie van risicokapitaal. Dus die zouden moeten weten: er zit een risico. En de goede VC, want ge moet die een beetje opdelen, de goede VC die heeft ervaring en die heeft typisch een thesis rond de markt.

Dus nu rond die AI, dat AI-fenomeen, ga je zien dat de venture capitalist eigenlijk gaat proberen ontwikkelen: wat gaat er gebeuren in die markt? En wil ik dan meer inzetten bijvoorbeeld op chips of optimalisatiesoftware op chips of vind ik dan meer in zitten op de applicatielaag en hoe gaat die veranderen? Dus een goede VC, die is daar al in mee. Maar je moet ook uw VC's openen in domeinen, geografie en stage.

Dus bijvoorbeeld, we hebben VC Patrick van Nuen in Nederland. Ja, die gaat al mee met Collibra sinds ik denk 2012 ofzo. Dus dat is ondertussen al bijna een persoonlijke relatie die je dan opbouwt en die heeft ook zo lang dat perspectief op die markt en die kan dan ook zien hoe kijken jullie naar die veranderende markt.

Terwijl als wij nu zouden zeggen: wij moeten een nieuwe visie aan boord nemen. Wat niet het geval is voor ons, want wij hebben een vrij gezonde business met nog veel uitdagingen voor de boeg natuurlijk en veel opportuniteiten. We hebben geen liquiditeitsnoden zo in vergelijking met in het begin natuurlijk toen we net startten en onze business nog aan het zoeken waren.

Maar stel nu dat een startup of een scale-up nu een VC moet meekrijgen. Ja, dan is het zeer eenvoudig. Je kunt niet tegen de VC nu zeggen: 'Ah, we gaan een B2B SaaS startup ontwikkelen', want dan ga je eigenlijk op de vorige platform shift een propositie aan het ontwikkelen zijn. En dat is niet noodzakelijk verkeerd. Je kunt investeerders vinden die daarvoor een thesis hebben en een investering willen wagen omdat ze verwachten dat daar nog altijd returns in te vinden zijn. Dus die bestaan.

Maar de meeste, als je naar de markt kijkt, de meeste VC's, jammer genoeg is er ook in VC-land een lemming-effect soms rond AI. De meeste kijken naar: oké, we moeten op AI investeren. Dus als gij vandaag een startup of een scale-up zet, dan is het vrij eenvoudig. Dan moet gij een visie ontwikkelen rond hoe gaat de wereld de komende drie jaar ontwikkelen. Wat gaan mijn positie daarin zijn? En hoe, waar sta ik vandaag? En hoe kan ik dan die afstand overbruggen en welke resources, aka in dit geval, heb ik daarvoor nodig?

Wel, je zijt typisch wel iets langer bezig met uw investeringsplanning, hè. Dus ik raad altijd aan dat ge toch, stel nu dat je zegt: ja, ik heb een investering nodig om gelijk welke reden dan ook. Dan raad ik toch altijd aan: ja, plant drie tot zes maanden, een true end-to-end, hè, van eerste gesprek tot cash op uw rekening. Want het is dat moment dat telt natuurlijk. Dus dan moet je toch drie à zes maanden rekenen. Als je het sneller probeert te doen, oké, maybe it works, but more likely it won't.

En idealiter geef je zelf toch liever 12 maanden. En in het geval van Patrick was het ook effectief 12 maanden. Want Patrick heeft bij ons geïnvesteerd op een bepaald moment en dat was niet de eerste keer dat hij ons zag pitchen. Dat was de tweede keer. En dan natuurlijk is er nog een heel due diligence proces, want er is een investment committee erachter en zij moeten van alles nakijken over markt en klanten en contracten en producten en technologie et cetera.

Maar wat er gebeurd was, is dat Patrick en team hadden ons 12 maanden daarvoor op een of ander VC pitch event in België, denk ik, had hij ons al zien pitchen. Dus toen hadden wij al met zijn firma gesproken en 12 maanden later opnieuw. En toen in het begin had hij niet in ons geïnvesteerd, maar 12 maanden later wel. En dat was puur omdat hij ook twee datapunten had. Dus hij kon zien: zij staan daar en 12 maanden later staan zij daar. Die evolutie, dat is een trend waarop ik wil wedden.

Nu, dat is een early stage VC ook, hè. Dus als je kijkt naar later stage VC's, we hebben bijvoorbeeld Capital G, dat is een dochter van Alphabet. We hebben Sofina, een soort van Belgische verankering ook op een manier. Dat zijn eigenlijk late stage investors en die dynamiek is helemaal anders. Dus die opereren al bijna eigenlijk zoals je een investering zou overwegen in een publiek bedrijf, omdat er ook veel meer data en business beschikbaar is om door te pluizen. Dus die VC's moet je toch opdelen in stage, geografie en een domein waarin ze investeren.

Ja. Oh ja, voor 10 jaar geleden. Dat is ja, dat is meer dan 10 jaar geleden. Goh. Eh ja, ik weet niet meer wat ik toen dacht. Waarschijnlijk niet veel. Maar ik heb wel al het gevoel gehad dat ik iets wilde creëren, iets wilde maken.

En dat maken ging dan altijd over iets in de computerwereld. Dus we waren daarnet bezig over de lokale Vlaamse geografie, hè. Dus onze gemeenschappelijke punt die we delen in het Vlaamse landschap is Zele blijkbaar. Dus ik ben daarin opgegroeid in Zele en mijn vader, wij hadden thuis een computerwinkel, hè. Dus wij, dat platform van de personal desktop, die heb ik meegemaakt in de zin dat wij in Zele dan de eerste computers bij gezinnen thuis gingen brengen.

En dus in die zin was het voor mij altijd fascinerend om met computers bezig te zijn en iets te maken rond die computers. Gelukkig voor mij ook, want mijn handen staan op praktisch vlak nogal zoals een mol, hè. Dus voor de luisteraars, ik ben een mol aan het uitbeelden. En dus ik ben in de computers terechtgekomen via die winkel. En dan later opleiding in de elektronica, waar terug mijn handen zo en mijn kleurenblindheid en die resistors, de praktische kant van de elektronica was voor mij een excuus om in de software terecht te komen.

Ja, inderdaad. Ja, de computers waren thuis vrij vroeg aanwezig. Ja.

Ik weet niet wat ze thuis zouden zeggen. Ik kan alleen maar gokken, maar ik heb geen idee. Toast, wat zouden ze thuis zeggen? Humor, fingers crossed ook, ene die je professioneel blijft gebruiken.

Ja, uiteraard. Ja, dan op een andere manier. Bijvoorbeeld humor van thuis. Als ik dan toch een beetje schaamrood op mijn wangen moet zetten, is dan bijvoorbeeld over ja, wat heeft het jongetje Stijn? Die heeft geen honger meer en dan komt er taart op tafel, maar ik heb wel nog taartenhonger. Dus een beetje onnozelheid uit het verleden.

Nee, nee, nee, nee. Maar die taartenhonger breng ik natuurlijk niet mee naar de bedrijfswereld.