Nvidia Without Huang, INVEST Without Leverage: Capital Follows People, Not Programs
Source: Startup Insider – Original Episode (episode in German)
Two topics, one thesis: Capital follows people, not programs. After 33 years, Jensen Huang has stepped down as CEO of Nvidia; CFO Colette Kress is taking over on an interim basis—ironically, the very executive who has been internally critical of risky financing deals and the circular AI economy. Björn Rieckhoff explains why the sense of security that Nvidia instilled in startups’ cap tables was never an Nvidia asset, but rather a Huang asset, and why corporate venture capital lives and dies by its sponsor. Next, we head to Eschborn to discuss INVEST, the venture capital subsidy program, whose budget—30, 44, or 74 million euros—is currently the subject of debate. Björn believes the debate is too narrow in scope: According to the 2016 ZEW evaluation, the program is most effective among virgin angels, while experienced investors have received more in grants than in additional capital—and according to BAND, 80 percent of angels in Germany invest in a single startup over the course of three and a half years —his alternative model is a matching scheme similar to that of the European Angels Fund. To start things off: PitchBook is opening its platform to founders for free, and Jan Thomas talks about his own INVEST application.
Key themes in this episode
- PitchBook goes free: PitchBook is opening its platform to startups before their first institutional round. Björn’s take: when you get something for free, you are often the product yourself – founders hand over their metrics, and the monetization happens with investors. Nephilia deliberately takes the opposite approach: founders pay a flat fee, because that is where the value is created.
- A Huang asset, not an Nvidia asset: After 33 years at the helm, the CEO practically is the company, with around 60 direct reports. The big startup deals went across his desk. According to Crunchbase, Nvidia led only about one in five rounds – the logo on the cap table mainly conveyed security and perspective, and that hinged on one person.
- The brake takes the wheel: Colette Kress has been CFO since 2013 and was seen internally as a counterweight to Huang’s aggressive financing deals. As an interim solution she calms things down – whether she can steer Nvidia into the future remains open, but that may not even be her job.
- The player and the coach: Startups chose Nvidia because of Huang – like players who join a club because of its coach. Now the coach is gone, and the new coach is the CFO, who always thought the player was too expensive. Corporate venture capital lives and dies with its sponsor; in the worst case, past commitments get revised.
- Small potatoes in Eschborn: Whether INVEST gets 30, 44 or 74 million euros is the wrong question for Björn – the sums are too small, and the administrative effort may outweigh the effect. Jan Thomas knows the program from the startup side: short expert reports on innovativeness, phone queues, annual renewals.
- It works for the wrong people: According to the 2016 ZEW evaluation (funding years 2013 to 2015), 2.7 million euros in grants to virgin angels triggered 14.25 million euros in additional capital. But according to BAND, 80 percent of angels in Germany invest in just one startup within three and a half years. Among experienced investors, 16.5 million euros in grants were matched by only 14.5 million euros in additional capital – a net effect of minus 12 cents per euro of funding.
- Euro follows euro: For Björn, the better model is the European Investment Fund’s matching, as with the European Angels Fund: for every euro an experienced angel invests, one public euro is added, and the decision stays with the angel. Instead of piecemeal funding: framework conditions that make startup investments easier and cut red tape.
Transcript
This transcript has been edited for readability. The content and statements have not been changed. The original conversation was in German; this is an English translation.
Jan Thomas: Björn, hi, nice to meet you.
Björn Rieckhoff: Hi Jan, nice to meet you too.
Jan Thomas: I hope you’re doing well. I’ve been thinking about you, Björn—as I mentioned during our preliminary conversation—because PitchBook is probably no stranger to you. They’ve launched a free service, which surprised me a bit, since PitchBook is actually relatively expensive. And at the same time, I thought about your new baby, if I may say so.
Björn Rieckhoff: Right, right. PitchBook is, of course, Nephilia’s biggest competitor in my view. Or rather, the other way around.
Jan Thomas: The other way around, right.
Björn Rieckhoff: Just to set the stage here. No, you’re right. I started Nephilia to help founders find investors, specifically for their funding round. The whole thing is integrated into the founders’ assistants via an MCP—whether it’s ChatGPT, Claude, or something else, it doesn’t matter. It runs through a connector that’s easy to use—and I get the impression that founders are increasingly warming up to this. And yes, you’re right: PitchBook has now started offering its service to founders essentially for free. That’s coming from a different angle, though. PitchBook is a household name to many and has an enormous amount of data. When you get things for free, you’re often the product yourself—that’s the social media analogy from back in the day. And that’s actually the case with PitchBook, too. I don’t want to sound too negative—it’s an absolutely legitimate model. But what you’re doing there as a founder is this: you’re providing your own financial metrics and the like to give PitchBook the opportunity, behind the scenes, to monetize that with its actual target audience—namely, investors.
Jan Thomas: What you actually meant to say was: It’s good that your service is a paid service.
Björn Rieckhoff: I think that’s the fairest way to monetize the whole thing. Everyone knows where they stand, and in the end, you can count on getting what you’re promised. PitchBook positions itself by offering particularly early access to potential deals and then monetizes on the investor side. You can structure the service either way. Sure, it’s harder to monetize founders than VCs, for whom that extra euro might be easier to part with, thanks to planning certainty and the like. But …
Jan Thomas: I don’t know if that’s still the case. VC isn’t exactly the easiest business right now, is it?
Björn Rieckhoff: Fair enough. But once you’ve closed your fund, you at least have budget certainty for the time being, and that makes longer-term contracts possible. Still, I always think it’s fair to ask: Who am I actually providing value for? In my case, that’s clearly the startups, first and foremost. Accordingly, I charge them a flat fee, which makes sense. I found that to be an interesting development in recent weeks. For Nephilia itself, however, this announcement doesn’t really change anything. Crunchbase, PitchBook, and the like were already there before, will continue to be there, and likely have their own MCP connectors on the market as well. For me, it’s about carving out a kind of boutique business that stands out particularly for its data quality and relevance to founders.
Jan Thomas: Do you know how many employees PitchBook has?
Björn Rieckhoff: To be honest, I really can't say.
Jan Thomas: I looked it up because I’d heard about this a year or so ago and couldn’t believe it: over 3,000 employees. I think that’s totally crazy. That’s on such a massive scale. You just see a news portal or data portal, but I wouldn’t have thought that … I guess a lot of them are analysts or some kind of entry-level analyst—data entry clerks? I don’t even know what the right word for that is.
Björn Rieckhoff: I know what you mean. I actually feel that same sense of separation—I mentioned that in our preliminary discussion. On the one hand, you want to make the best possible use of the connector, including in terms of how it’s used. I’m increasingly involved in supporting founders during their funding rounds—including providing feedback to Nephilia: Who’s giving what kind of feedback? That way, you can adjust the prioritization of individual funds and continuously expand the list of contacts you should speak with. The other level is the data level. There, you have to implement a whole series of routines to ensure the quality of this data. Whether it’s an announcement like “Sequoia is pulling out of Europe” or “new fund from XYZ closed”—these are all things that need to be updated daily and with a certain level of quality, ideally using registry data and the like. You can spend a lot of time on that. Even if you can use AI to support the research or tap into the registries via APIs, it still requires a certain amount of oversight. That’s a time commitment that’s always there in the background, and I’m really noticing it now.
Jan Thomas: I was looking it up at the same time: data collector or data entry clerk or something like that. I don’t like it when I have to search for a word like that—there’s no fancy term for it, but you know what I mean. I think they have this massive army of 1,000 or 2,000 employees based in India or Pakistan who do exactly that. And—this might be motivating for you—there’s no exact revenue figure published, but estimates range from 618 million to 1.8 billion. So you’ve still got a ways to go, but I think that’s a great goal.
Björn Rieckhoff: Hey, Jan, at this point I’d like to remind you again that last time you said what Nephilia is building—even down to the smallest details—isn’t exactly rocket science—we said that, just for fun. That’s right. Then I’ll check in with you in a few years when I hit 680 million. [Editor’s note: Jan Thomas had cited 618 million as his lower estimate.]
Jan Thomas: Cool. I’ll listen to it on the go. I’ll check in with you every now and then.
Björn Rieckhoff: Right, right. “Build in Public”—I’m a big fan of that. We’ll see if I ever publish revenue figures. If I do, it’ll be on your show, Jan.
Jan Thomas: Cool. We were already halfway through the last episode, so let’s pick up where we left off: Last time we talked about Nvidia—it was still a high-profile event, with the red carpet rolled out. I have no idea what they’d roll out this time—or for whom, to be honest. In any case, let’s talk briefly about Nvidia in the—how do you say it now—post-Huang era?
Björn Rieckhoff: Yeah, it’s hard to imagine. Someone who’s been CEO for 33 years is, in a sense, the company itself—you could almost say that. That’s a pretty drastic change. I don’t want to get into all the personal stories in this episode, so I’ll just say this objectively: After 33 years, it’s difficult to separate a person from an organizational structure—with all the dependencies and the centralization around that person that simply existed.
Jan Thomas: Totally. I was just amazed. There are so many rumors right now and so many facts that are gradually coming to light. I think at some point, the Wall Street Journal or the New York Times will have to do a full investigation into this. Right now, you don’t really know what to believe. There’s a rumor going around that Elon Musk has set up a sort of “war room” to get him reinstated—just like with Sam Altman back then, who was only gone for a weekend. And someone has already joked that it would be totally funny if he came back—wearing a white leather jacket. That would be an extremely funny image. If that really happened, it would be the perfect picture.
Björn Rieckhoff: That’s really funny, yeah. I’d almost leave it at that.
Jan Thomas: It’s really hard to imagine Nvidia without him. I read recently that he has around 60 direct reports. He’s like some kind of sun god over there. And I think all those big deals with startups we’ve been talking about somehow went through him, because he also understood how to ramp up the game a bit overall.
Björn Rieckhoff: Exactly. And that brings us to the right starting point. My main goal in this episode was to look at this centralization from this perspective: Just how much was this actually top-down at Nvidia? This was triggered by a few press reports in which people said: Of course, a great many people are involved in Nvidia’s startup investments, but that sense of security wasn’t really an Nvidia asset—it was a Huang asset. He actually made the key decisions, especially on the more important deals, and pushed them through internally with considerable determination. What I find fascinating now is the tension—between the board on one side and Huang, but also between Jensen Huang and Colette Kress, who is now taking over as interim CEO and served as CFO for many years. If you take a closer look, you suddenly realize there’s almost a kind of counterpart to Jensen Huang who was critical of certain investments. Sure, she was conservative as CFO—but she had definitely criticized the riskier financing deals in particular, especially the circular AI economy, which we’ve discussed frequently in recent episodes. She wasn’t exactly its biggest advocate. In that sense, I believe that putting her at the helm—at least on an interim basis—could also send a signal: one of a less aggressive approach, including less aggressive customer financing. I find that an interesting nuance in this whole issue.
Jan Thomas: It would be pretty wild if there really had been such a rival within the company. That’s all speculation—I don’t know anything for sure. But by the very definition of her role, she has to make sure everything is above board. And she’s probably also keeping an eye on the impact Nvidia has on the global economy. She knows: Anything can backfire, anything can lead to awkward questions, to a congressional hearing, and eventually her head could end up on the chopping block. So I can certainly understand if someone goes around saying, “Computing time—anyone who wants it can get computing time from me in exchange for a stake in a startup.” That at some point a CFO might raise their hand and say, “Wow, that sounds a bit half-baked”—I can understand that.
Björn Rieckhoff: Yeah, and she hasn’t exactly just started at Nvidia yesterday. I think it would be a bit harsh to say now that …
Jan Thomas: It’s already here in 2013, I see, right?
Björn Rieckhoff: Exactly. What’s happening here isn’t black and white either. The two have worked very well together in the past. You simply need these different perspectives to end up on a very solid path forward. I’m just saying: their approaches and ways of thinking are definitely a bit different. And of course, that’s also a board decision: Who do you put in that role? Especially when someone has been your CEO for 33 years, it’s not such a bad idea—at least for an interim period—to put someone in place who has distinguished themselves in the past primarily through substance and a certain conservatism. That helps calm things down for the time being. Whether that person is the right one to steer the company into the future remains to be seen. But that may not even be their job in the first place.
Jan Thomas: Donald Trump called Tim Apple right away.
Björn Rieckhoff: Right, right. And it’s kind of understandable—in hindsight, it’s all just a big commotion. But I’m curious to see how this plays out; it’s sure to be an interesting development in the coming months.
Jan Thomas: You’re absolutely right. And yet I still don’t quite understand this whole 60 direct reports thing. That’s what I find so fascinating: We’re talking about what is—at least at times—the most valuable company in the world—with an organizational structure that any normal manager would have deemed completely impossible. How many direct reports can you have? Seven? In a really extreme, founder-centric company, maybe 15 or 20—but 60?
Björn Rieckhoff: If you rely on standard management practices, you can’t handle that. But if you think about what a typical workday for Jensen Huang looked like until recently: those aren’t 60 employees who need supervision, but rather 60 people who spend 99 percent of their time working completely on autopilot and might just need to check in every now and then. Formally, you’ve put him in charge as CEO, but I can’t imagine that this would be comparable in any way to standard management practices. There simply isn’t enough time for that. So, we’ll have to wait and see. The CFO certainly hasn’t taken on those 60 reports.
Jan Thomas: Yeah, no way.
Björn Rieckhoff: That’s going to be restructured in some way.
Jan Thomas: Of course, people have been paying a bit more attention to him lately, and I’ve seen interviews time and again where he complains about how little time he actually has. There he is, sitting on a stage somewhere in the afternoon, saying, “I worked until three in the morning and only had time for half a sandwich today.” The guy looked perfectly healthy, but he’s not exactly young anymore either. I find all of this fascinating. But if I had to bet on it, I’d say: He’ll be back. I’m almost certain of it. He’s so well-connected.
Björn Rieckhoff: You need a figurehead like that, a poster child, exactly. Let’s see: Today, November 17, at the time of recording …
Jan Thomas: Exactly, we’re recording on the 17th, yes.
Björn Rieckhoff: … Nvidia is releasing its quarterly results. I think they’re expecting 91 percent growth again. That’s basically a great first call you can make as CEO when you announce numbers like that. Let’s see what happens to the stock price then. Of course, a certain level of expectation was always factored into the price. But over the past few days, we’ve also seen just how much of the investor narrative hinges on him personally. There’s no other way to explain the short-term dip. It wasn’t really driven by fundamentals, because as far as the financial results go, there were no developments other than the CEO role itself.
Jan Thomas: The situation right now is really exciting. It actually started with the midterms, then there’s Anthropic, then OpenAI—which isn’t quite filling its funding round—and now Nvidia with its financial results and Jensen Huang’s departure. It’s definitely an exciting phase for the overall market.
Björn Rieckhoff: For the overall market, and I think especially for Nvidia’s portfolio companies. In the last episode, we talked about Nvidia’s presence here in Germany and the investments it has made in Europe, and we also asked: How does Nvidia actually get involved with its portfolio companies? These were often companies that would potentially become very large customers of Nvidia in the future, and that’s where they got involved. With that in mind, I took a look at how Nvidia actually operates—Crunchbase had looked into this. Only about 20 percent of the investments that Jensen Huang was essentially responsible for were as a lead investor. Most were, in later funding rounds, more of a stake—an absolute minority stake—though still of a certain magnitude and likely often linked not only to equity but also to deals involving computing power. This didn’t necessarily mean there was a central dependence on Nvidia as a shareholder in the cap table, but it did provide a certain degree of security because a key player in the AI ecosystem was involved—and that, of course, also gave the individual investments a sense of direction.
Now imagine you’ve picked out Nvidia as an investor for your startup and, just to be sure, even spoken with them in person—the organization was very centralized, after all. I picture it like this with a soccer player: He joins a club because a certain coach is on the sidelines there, and he’s exactly the type of player that coach needs for his lineup. Then the coach gets fired. The player continues to wear the club’s jersey—but the new coach is, of all people, the CFO, who has always viewed his transfer with skepticism.
Jan Thomas: Because it’s way too expensive.
Björn Rieckhoff: Exactly. And every startup in Nvidia’s portfolio is in that situation right now. As far as Nvidia’s future involvement is concerned, the founders are probably putting a little question mark next to the name on the cap table for now. Not necessarily in terms of future investment activity, but: How does a person react now? You’ve built the entire business relationship on Jensen Huang, so to speak, and now you have to fundamentally question that. It will be interesting to see whether certain investments are scaled back—including those related to computing capacity—if he doesn’t return. But I also believe that, in principle, we’ll see a certain shift in policy. A future CEO will certainly start sending signals.
Jan Thomas: Do you know anyone who’s received an investment from Nvidia? I’m asking because we all still remember the media-for-equity deals we’ve always compared this to. Back then, ProSieben—and later, I think, RTL as well—did tend to undervalue those deals a bit, but it was still a currency they set themselves. There’s a list price—let’s say an advertising budget of one million euros—that ProSieben, because they might not have been able to sell all that ad space anyway, could have written off, or on which there’s still a 90 percent margin. And that’s how it is with Nvidia afterward, too. They’re a monopoly and feel relatively little cost pressure. Take Avelios Medical, one of the four startups in Germany set to receive an investment: They definitely need the capacity; otherwise, they wouldn’t be bringing Nvidia on board—just like Black Forest Labs. And if Nvidia now says: “That capacity actually costs—you name it—10 million; we’ll invest 7 million, but you’ll get the full 10 million”—then that’s actually a great deal for both sides. And a CFO will understand that, too—that’s my point. I’m not even sure if these deals—depending on how they’re financially structured—are really that bad for Nvidia in the end.
Björn Rieckhoff: Certainly not. After all, it’s almost always exactly the case you’re describing—potentially very large future customers of yours. In that sense, it makes perfect sense. But what I also see—you mentioned ProSieben—is this: When it comes to corporate venture capital or strategic investments, the project often stands or falls with its sponsor. If you have that one top manager driving the whole thing, and that person leaves, then that’s at least a disruption. Whether you can pick up the ball and keep the momentum going remains to be seen. But first and foremost, it’s a situation that calls for a thorough assessment: What do I do with this now? There are various examples from the past where top managers stay with a corporation for three and a half years, build something up, and then the CVC fizzles out again—to put it bluntly. Because there’s no internal sponsor, and because the results aren’t there yet after three and a half years. At Nvidia, as you say, the situation will be a bit different given the allocations and commitments to potential customers. But it will still be interesting to see how this plays out. The mildest scenario would be: If the budget is cut, deals like this simply won’t be made with the same level of aggressiveness in the future. The most drastic scenario would probably be that certain past commitments are revised because someone says, “Nah, that was actually way too aggressive.” Who knows—we’ll see how it all plays out.
Jan Thomas: Maybe even up to the Hugging Face acquisition. That might be a topic of discussion at some point—I think that was around 13 billion.
Björn Rieckhoff: Yeah, it’s not finalized yet, exactly. It was just announced for now.
Jan Thomas: I’d also read about Kress: The Wall Street Journal reported that she had already confirmed in an interview that virtually all of these deals went through Huang, and that he often closed them with a handshake. At OpenAI, I believe she cut a data center commitment in half. So she’s already gone a bit overboard. I can imagine that for a CFO, it’s a bit of a nightmare afterward when someone is running around all the time handing out checks to whoever they feel like. We’ll see.
Björn Rieckhoff: Sure. That’s going to be a really exciting story, and, as you say, it will have a certain influence on market trends in the coming years. Everyone is aware of that—otherwise, someone like Musk, Trump, or whoever wouldn’t step in to set the tone for who should take that position.
Jan Thomas: And that’s a wonderful transition, Björn, to our second topic, because that’s what will shape the markets in the coming years like nothing else. I’d say let’s head over to Eschborn. I believe that’s where the RKW and the BAFA are located.
Björn Rieckhoff: Exactly, that’s right. And that’s where we’re starting to aim very, very, very modestly. We’re talking about INVEST, the venture capital grant.
Jan Thomas: Which others also refer to as a rounding error in Jensen Huang’s balance sheet.
Björn Rieckhoff: Right. And that actually brings us to the topic at hand: There’s a very, very high-profile debate going on about whether we have 30, 44, or 74 million euros in the program. And I think to myself: If you’re talking about these kinds of sums, you might as well just shut it down and put the money toward something else. Because the administrative burden this whole thing causes is probably almost greater than the positive effect. I’m now deliberately making myself unpopular with a great many business angels and, above all, probably with a great many startup supporters and associations in this country. But I just find it absurd and ridiculous how fragmented these figures have become over the last few years. I think the idea behind the program itself—to promote business angel financing in Germany—is absolutely excellent. It’s just that, as is so often the case: you don’t need to come up with 44 million, 30 million, or 74 million euros, as the Greens are demanding. That’s a bit too small for my taste.
Jan Thomas: Although the question of 30 or 74 is certainly a relevant one. That’s two and a half times as much. Maybe we should take a moment to step back and ask: What’s the point of this program anyway? What’s the idea behind it? I don’t know if you’ve ever applied for it—I’ve applied for it, though. As a startup, you have to fill out a lot of applications—it’s a real administrative hassle. When you see the paperwork and how often I had to call them and so on—wow, I don’t know if there isn’t a way to streamline that process. Couldn’t we say: Every business angel who invests gets a 20 percent reimbursement or so through their tax return? I think it’s 20 or 25 percent—I’m not exactly sure what the current rate is. The goal would be to motivate them to invest more.
Björn Rieckhoff: Right. And as is so often the case, this has also been evaluated multiple times by independent third parties: What are the effects of this program? I took a look at it, and it’s quite interesting to see how it has developed. It’s now an explicitly stated goal of the program to trigger—I don’t want to say “generate”—new business angels, that is, to stimulate investment in startups. And if you look at it in terms of just that one variable, it works quite well. But I’ll get right to where the whole thing falls short on a deeper level. In 2016—ten years ago—the ZEW examined this and found that 30 percent of the supported investors were investing in a young company for the first time. So far, so good.
Jan Thomas: That’s exactly what it’s all about.
Björn Rieckhoff: That certainly fulfills the objective. In 2022, as many as 41 percent of the supported investors were active as angels for the first time. So far, so good. The study also found that among “Virgin Angels”—a term I came across for those investing in a startup for the first time—a 2.7 million euro grant triggered an additional 14.25 million euros in capital. Which is great. If you look at the net effect—that is, 14.25 million minus 2.7—for every euro from the federal government, you get a net effect of 4.26 euros on top of that. That works; that’s actually quite good. But it must be said: A great many angels investing for the first time—with all due respect to the individuals—aren’t necessarily the ones with the most experience as investors or the most relevant network of clients, follow-on investors, or whatever else characterizes a very good business angel. And a 2023 study also found that 80 percent of these angels invest in a single startup over the course of three and a half years. Just one. [Editor’s note: This figure comes from BAND’s 2023 Business Angel Report and refers to all active business angels in Germany, not just those supported by INVEST. According to the 2026 INVEST Evaluation, 47.8 percent of funded first-time angels did not make any further investments after their first one.] So you think: Okay, the goal of the funding program is to increase the number of business angels in Germany. But if 80 percent of those who invest for the first time through the program have still made only a single startup investment after three and a half years, then I can’t really speak of a sustainable expansion of the business angel base.
Jan Thomas: That’s crazy, Björn. I’d love to know the reasons behind that. I would have thought that someone making their first investment would know right away: As a business angel, you always need some kind of portfolio approach. You can’t stick with just one—not for three years. If you have 50,000 euros, you’re better off writing three checks for 16,000 each rather than one for 50,000, I’d say.
Björn Rieckhoff: I don’t know either. I spent some time in venture capital and spoke with countless LPs who had extreme concentration risk—extreme risk tied to individual vintages. It’s exactly the same thing, just one level higher or lower, depending on how you look at it. It’s the same with business angels. Rationally, everyone knows this, but that doesn’t necessarily mean people behave rationally—or even can. Perhaps most people come to realize over time that it’s much more work than they expected, even with an INVEST grant: a ton of administrative red tape, notary appointments, shareholder resolutions, and the like. And they realize: If I really want to scale this up now, I’ll have to quit my full-time job. Based on these dynamics, I’ve repeatedly come to the conclusion that anyone considering investing in startups is, in my opinion, always better off finding an early-stage fund and investing there as a small individual investor in an entrepreneurs’ feeder fund or a pool, rather than doing it on their own. Because—sorry—you suddenly want to play in the same league as the pros, and for that, you need a certain foundation. I’m absolutely convinced that anyone who accumulates private capital should invest in public equities and real estate, and only after they’ve exhausted all those options should they start with alternative investments—and please, only with capital they truly no longer need to live on.
Jan Thomas: Sorry to butt in, Björn: I’m getting the sense that this is a bit like “friends and family”—people who transfer 50,000 euros to their son or grandson once and are then just grateful that the INVEST program exists, because it’s a perk afterward. That’s how I tend to see it when you say they won’t invest any further for three and a half years. I think that’s extreme.
Björn Rieckhoff: The same studies reveal a second effect: What actually happens with experienced investors? Thanks to INVEST, they actually invested more—specifically, an additional 14.5 million euros. So without the program, less money would have flowed into startups. But the net effect—additional capital minus the grant that was spent on it—is that this 14.5 million euros was triggered by a 16.5 million euro grant.
Jan Thomas: So that means the effect …
Björn Rieckhoff: … the net amount left over is actually negative. In other words: For every euro of funding invested, you have a negative effect of 12 cents with experienced investors. What’s interesting is that you paid out more in grants than the amount of new capital that was triggered. The fundamental question remains: Did you invest additional money based on this INVEST program, or would you have invested it anyway? I would certainly question whether that actually fulfills the program’s objectives. [Editor’s note: The figures are from the 2016 ZEW evaluation, covering the funding years 2013 through 2015, when the subsidy rate was 20 percent. The 2026 evaluation by BF/M Bayreuth, based on a survey of 1,761 investors, shows a net effect of approximately 94 cents per euro of funding for the program as a whole. It does not include a net breakdown by experienced investors, but confirms that the mobilization effect is higher among first-time angels. For 34.5 percent of the recipients, the grant had no influence on the investment amount.]
Jan Thomas: But we also don’t know exactly how much money business angels invest in Germany overall. I tried to find that out, but there aren’t any figures available. Just to put it in perspective: EY says that last year, I believe, there was 8.4 billion euros in venture capital. When you hear those numbers, that amount is negligible. As you know, we have a large Cap Table and a broad group of shareholders. First of all, I find it difficult when someone bases their decision on whether we’re “INVEST-worthy.” That actually shows me what level someone is operating at. I don’t mean that disrespectfully at all—it’s obviously cool when someone knows how to handle money and where the funding sources are. But on the other hand, I know this: The really big-time investors on our cap table would never ask me that. Some of them don’t even know this program exists. Well—they know about it, but they don’t use it. That’s how it is, I think, and that’s the right way.
Björn Rieckhoff: Exactly, that’s right. As I said, I’m a fan of startup funding. I’m absolutely in favor of having a broader base of business angels and raising awareness in society that people can—and should—invest in startups to drive innovation. And of course, this INVEST program also has secondary—and probably tertiary—effects on the economy and society—such as job creation and the value generated as a result—which I’ve completely left out of this discussion. But my main point would be: When we talk about these figures—whether it’s 74 million euros or 44—it really doesn’t make a huge difference. Perhaps we should instead pool the 44 or 74 million euros and channel them into something else where we feel there’s greater leverage. And perhaps channel that into creating the right framework conditions—or use this money to establish conditions that simplify investment in startups, reduce bureaucracy, and make it easier for founders to get started in Germany, and so on. Perhaps the impact of that would actually be greater than if we focus on the minor details of this program. After all, the sums involved are simply too small to justify discussing tax savings, the number of business angels attracted, and the like.
Jan Thomas: We should take a look abroad to see if there are countries that are a step ahead and where people have already gained experience. To be honest, I don’t know what the right number is here—I don’t really have an opinion on that at all. Maybe it would make a difference later on if the number were higher—say, 200 million or so. I have no idea if things would be different then. But maybe it’s just as you’re saying. For example, I recently had Robin Haak from Robin Capital here—you know, the solo GPs; there are quite a few of them. Maybe that’s more along the lines of what you said earlier: that angels first get to know this whole landscape and don’t make it dependent on a grant afterward if they want to play the game.
Björn Rieckhoff: There was a business angel program in Europe run by the European Investment Fund called the EAF—the European Angels Fund. Essentially, it was a 1:1 co-investment by the European Investment Fund—or, I believe, even the Investment Bank—designed to support successful and experienced business angels. They didn’t look at individual deals to determine what would be financed or whether it was eligible for funding; instead, the business angels were given—in quotes—“a blank check”: For every euro you invest, we’ll invest one euro as well. In principle, this is the well-known matching of funds by the European Investment Fund, which they have also been practicing very, very successfully at the fund level for years. There, it’s not a 1:1 match, but is subject to other ratios that, I believe, depend on the specific case. But it’s a well-known system that works very well: The euro from government funding follows the euro from private-sector funding. Because I’m firmly convinced that the government isn’t always the best entrepreneur—and probably isn’t the best investor either. So you leave it to those who know what they’re doing to allocate the capital, and simply ensure greater leverage for that capital. That’s a very, very sensible dynamic. However, this must also be carried out with minimal bureaucracy, and with the EAF, a program was launched back then that operated exactly that way. To my knowledge, however, it is no longer active in Germany. I believe it’s still active in other European countries, but not currently in Germany. [Editor’s note: According to the German federal government’s ERP report (Parliamentary Document 20/12935), the European Angels Fund Germany has not entered into any new co-investment agreements since 2021. Existing partnerships with business angels will continue until the respective end of their contracts.]
Jan Thomas: And this lack of bureaucracy—I’d like to briefly come back to what I was saying earlier about the application process. It has to be verified… I think this is another example of that hidden sponsorship by certain coaches and consultants, because there’s always a brief assessment. You have to have the innovative business purpose and the innovative business model verified. We briefly talked about Black Forest Labs earlier: Do you really think a founder of Black Forest Labs—of such an innovative company—has the time to sit on hold with the BAFA? And then every time you’re told: “Nope, Ms. Meier isn’t here right now; she’s on vacation today or working from home and can’t be reached—please send an email.” Does any of these top founders have the patience for this kind of hassle? That’s why I really just want to say: That process needs to be significantly streamlined, too. Either you’re innovative or you’re not, and that can be determined by other criteria as well. You could also say there’s a “lighthouse” angel investor who’s accredited and can briefly confirm it—that could be a two-liner. But there’s always this whole string of red tape attached to it. And then it’s only valid for a year and has to be renewed every year. We’ve renewed it twice now—it’s a nightmare. That’s the part that really gets me—from an entrepreneur’s perspective, not an angel investor’s.
Björn Rieckhoff: Hey, but in principle, you’d prevent exactly those kinds of dynamics by applying the familiar technique of mirroring.
Jan Thomas: The Super Angels do that, too—it’s actually a similar model.
Björn Rieckhoff: Exactly, exactly. Just give the capital to the people who know what they’re doing. You don’t have to become an expert yourself, because you might not be able to build up that expertise over time—or at least not in the time it would take for a specific deal. Otherwise, you’ll end up with adverse selection again. How to design and structure funding programs is, of course, a topic in its own right. But for me, when it comes to this INVEST grant: I see the program’s advantages, I know how to appreciate them, but I’ve come to find it simply too fragmented. In that case, it would be better to discontinue it and allocate the funds elsewhere—or to fundamentally ask how to create a system that can actually promote business angel activity in Germany.
Jan Thomas: You’re available if the BAFA has questions about what something like this might look like.
Björn Rieckhoff: Exactly, so I’ll start by talking about association work. No, but of course it’s true that the federal government, the states, and the ministries are really making an effort here. I’ve actually been to the ministry a few times in the past and have weighed in on the matter. All of that is happening, and it makes sense and is a good thing. As you say, the key is to establish unbureaucratic structures. I’m not saying: Make it as easy as possible to spend money, with as little regulation as possible. That’s not it. Rather, with moderation and a certain level of expertise—but please, let it be done by the people who do this every day anyway.
Jan Thomas: Let’s take that as our closing remark, Björn. Last time we ended up on such a rant, so we have to be careful that doesn’t happen again.
Björn Rieckhoff: Exactly, every episode all over again. Constant dripping wears away the stone.
Jan Thomas: But let’s see. As I said, I have absolutely no opinion on whether this should be cut or increased to 72 million. [Editor’s note: The Greens are calling for 74 million euros.] It probably won’t make much of a difference, at the end of the day.
Björn Rieckhoff: When you’re talking about the amounts that are typically being invested in the AI boom right now, that’s really negligible. I’m curious. Maybe there will be a few controversial comments about this episode on social media. I’d love to get a sense of the general sentiment there as well.
Jan Thomas: I’ll send it to Ms. Meier at the BAFA. And that was just a made-up name, by the way. But I’ll send it to the BAFA anyway—you guys can have some fun with it.
Björn Rieckhoff: Right.
Jan Thomas: Cool, Björn, thank you. That was fun.
Björn Rieckhoff: Me too. See you next time.
Jan Thomas: See you next time, ciao.
About Björn Rieckhoff
Björn Rieckhoff is the founder of Nephilia and a business angel with ten years of experience in early-stage venture capital. He helped build Cavalry Ventures as its first employee and later became a partner of the fund. Nephilia grew out of around 100 transactions and board seats from seed to Series B: an information service that delivers investor selection for a founder's round directly inside their own AI assistant. Alongside it, he remains a sparring partner for entrepreneurs.
About Startup Insider
Startup Insider is the industry portal for the startup scene in the DACH region. It covers news from all regions and industries, along with an overview of key players and events in the German-speaking startup world.
Raising a round? Reasoned investor prioritization for European early-stage rounds: $ open nephilia.ai
Links
- TechCrunch: Jensen Huang’s departure leaves a huge hole in AI startup funding (November 13, 2026)
- Al Jazeera: Why did Nvidia CEO Jensen Huang resign, and who is Colette Kress? (November 12, 2026)
- SiliconANGLE: PitchBook opens its platform to startups for free and launches Origin (November 10, 2026)
- Börsen-Zeitung: Greens Call Plans to Cut Startup Funding “Completely Incomprehensible” (November 15, 2026)
- ZEW: Evaluation of the INVEST Funding Program (2016)
- ZEW/Technopolis: Ex-ante Analysis of INVEST (2022, Executive Summary)
- BF/M Bayreuth on behalf of the BMWE: Evaluation of the INVEST Funding Program (2026)
- EPISODE: Business Angel Report 2023
- Bundestag, Printed Paper 20/12935: ERP Report 2023, Section on the European Angels Fund Germany
- Nephilia – Information Service for Founders
- The Nvidia Pact and EU Inc: Is Europe Buying Sovereignty Instead of Building It? (previous episode)
- Launch Nephilia: Select investors via MCP directly within your own AI assistant