Rick Heitzmann

The through-line

Rick Heitzmann's consistent frame is that venture capital is an asset management business, not a taste-making one. "As much as VCs like to be, you know, uh pickers or they like to be style makers, they like to be podcasters, you know, we view our job as being asset managers who actually have to think about not only buying but also selling securities" [1]. Because the power law companies "usually generate those [returns] in the public markets," an early-stage investor who is not fluent in capital markets does not actually understand where the money comes from [1]. That belief runs through everything else he says: the reading of the IPO calendar, the use of secondary trading as a pricing signal, the judgement about which software businesses survive AI. It is also the base from which he built a network-driven seed and Series A firm in New York, backing consumer and enterprise technology from a city that was not the obvious place to do it [1][3].

On what actually opens and closes the IPO window

His model of the IPO market is a model of uncertainty. "The market hates uncertainty," and the companies about to go public are already the most uncertain assets in it; layer on "uncertain regulatory environments, things like tariffs and things like an uncertain presidency," plus "wars, you know, unstable macroeconomic times, very volatile currencies," and the window narrows [1]. He describes 2022 through 2024 as "as bad of a time as you've seen in years," followed by a genuine reopening across crypto, fintech and traditional SaaS, with Figma as the marker, and a backlog "as good as a backlog as you've seen in years" [1]. Then the macro turned again and hit "a pause button" [1].

The mechanics matter as much as the mood. Companies get ready in Q2 because that is when annual audits clear, with SaaS businesses on a January fiscal year-end taking longer [1]. And the smaller deals have a second problem beyond macro: the gravity of the megacaps. "Even if you're doing a billion dollars in revenue, you're profitable, you're growing, well, no one cares cuz everyone's just staring at SpaceX, OpenAI, Anthropic" [1]. Banker attention follows the fee pool. A banker thinking "if I could raise 50 billion from Anthropic, this is going to be a huge year for me, for my firm" will not be "bothered with a which was historically a great IPO, a $500 million IPO" [1]. He is sympathetic to the founders caught in that squeeze, people who "have worked very hard for a decade or more building a couple hundred million dollar company, even a billion dollar company" and might still slide through in Q2 [1].

On why the giants have to go public

Heitzmann's central claim about the current cycle is that the largest private companies have finally hit the ceiling of private capital. "SpaceX, Open AI have accessed every bucket, every pocket, every sofa cushion of private capital," and the only remaining pool, sovereign wealth, is constrained by the Middle East being "tied up" and the world being more uncertain [1]. He calls this a first: for as long as he has followed the markets, back to the 1990s, companies have not run out of private money [1]. His ranking is SpaceX first, driven mainly by Elon and by the capital needs of providing internet from space, then OpenAI as "the company that needs to go public just being capital consumptive," then Anthropic, which he thinks goes despite Dario's line that "maybe next year I'm out of money or maybe next year I'm a trillion-dollar company" [1]. Stripe he puts off to next year, and probably never by choice: "Stripe never wants to go public," because it has been "really good at" raising capital and giving early employees and investors liquidity privately [1].

That last point generalises into his test for whether any company needs the public markets at all. If a business already has liquidity and capital depth privately, there is no other reason to list [1]. He notes the argument that a private investor class has its own incentive to keep companies private and price them itself, and does not dispute that the depth is real: sovereigns and large institutional LPs co-investing in big rounds simply did not exist twenty years ago [1]. He also expects the eventual S-1s to be the most interesting reading in the market, given Nvidia's deals, the Amazon investment in OpenAI, and off balance sheet private credit structures such as the reported x.ai deal to buy Nvidia chips in which Nvidia also took equity: "the disclosure around them will keep you busy for weeks" [1].

On secondaries as the best available price signal

FirstMark actively watches the secondary market for its top positions, roughly the first twenty, of which fifteen or so beyond the household names have markets liquid enough that "the trading price matters" [1]. The pattern he reports is blunt: "the buzziest ones traded a premium to last round. The ones who were not as buzzy traded at a discount," and he treats that spread as "a better predictor of public markets than anything else" [1]. Part of why it works is who is setting the private price. When a Baillie Gifford, a Fidelity or a BlackRock leads a large private round, "they're kind of pricing it like an IPO" [1].

On venture quietly absorbing small-cap tech

The disappearance of the small-cap public market is one of his more structural observations. Going back to the 1990s there was a whole ecosystem of analysts and investors following $300 million market cap companies from 300 to 3 billion, "an somewhat inefficient part of the market, really good part of the market that billions of dollars were invested in," and it is gone [1]. Two forces replaced it: companies staying private longer, which pushes bigger outcomes to the eventual listing and rewards the venture investors who hold on, and public market institutions moving back into private rounds to get the exposure they can no longer find on the exchanges [1]. His conclusion is that "the venture capital industry has totally subsumed that whole sector," done "somewhat slowly and quietly" and, at least in FirstMark's case with its growth fund, "unintentionally" [1]. The message LPs have drawn: "if I want access to the best companies, I have to get it through the best venture capitalists" [1].

On what survives the SaaS apocalypse

He rejects the blanket verdict. "People have blanketly said SaaS is dead. So, uh some SaaS companies are dead. I think Figma is a very good company, which is not dead" [1]. His underwriting test for software now has two parts. First, pure workflow is not a business: "you can't just be workflow, cuz that's being commoditized incredibly quickly," so a product needs network effects and collaboration [1]. Second, and more important, proprietary data. "Data is the the oxygen for AI. It's also the moat in AI," and whether the software is horizontal or vertical, access to that data is what compounds competitive advantage over time [1].

The companies he expects to be taken out are the ones that stopped evolving, particularly those bought by private equity firms that "cut R&D, tried to milk that that user base" [1]. Those customers are "probably tired of getting milked," an AI-forward alternative will offer more value at similar cost, and a balance sheet burdened with debt cannot fund the response [1]. Figma is his counterexample: down roughly 70% from the IPO, still growing about 40% with 136% net dollar retention and back to being a rule of 40 company, run by an excellent CEO, competing at the edges with the replits and lovables and with Adobe on collaboration [1]. His rule of thumb is that "if you have a big market, you have an excellent CEO, um, hopefully, when somebody else throws the baby out with the bathwater, that's a good opportunity" [1].

On incumbents, distribution and the courage to cannibalise

He is unwilling to assume incumbents get decimated by a new generation. The Oracle versus Salesforce story is instructive because Oracle "has also done tremendously well since the rise of Salesforce" [1]. Salesforce today, he argues, is not passive: it is loud on agents, aggressive, and buying its way to product, including acquiring FirstMark portfolio company Momentum, which he describes as AI-forward. Plugging great products into amazing distribution is "kind of how Salesforce has grown over the last 25 years," and the open question is whether it is being thoughtful about what to build, buy and partner on [1]. "I know they're not crying in their beer" [1].

The harder constraint is investor patience. The problem with public markets, in his framing, is whether a company "can slow down to speed up," and whether shareholders grant the grace to cannibalise your own pricing or product for a better long-term position [1]. His view is that the best CEOs earn that grace, naming Benioff and Elon as examples [1]. He sees the same tension inside product roadmaps generally: "my new product that I'm going to take years to develop, I might be able to sell it to an existing customer for less. That's a bitter pill to swallow" [1].

On concentrated market power in AI

Beyond capital consumption, what strikes him about the leading AI labs is the sheer asymmetry of their influence. The consumer adoption of OpenAI is "amazing," Anthropic's revenue scale is amazing, and its market power "really amazing": "I've never seen a a private company just kind of announce a product and crash a whole public sector," pointing to Anthropic entering security and the sector falling more than 10% [1]. He also flags the political volatility around all of this. The current administration "has had back and forth with every single company where they've been on the naughty list and on the nice list," and a demand for a government stake can reorder a sector overnight, whether prediction markets or chip companies [1]. Anthropic's friction with the Defense Department looked damaging, yet its consumer pull and download numbers cut the other way, and he is candid that the net effect is unreadable [1].

Takeaways

• Treat venture as asset management, not picking: power law returns are realised in the public markets, so the job includes selling securities, not only buying them [1].
• The 2026 IPO calendar hinges on uncertainty from war, tariffs, regulation and an unpredictable presidency, plus the gravity of megacap listings that pulls banker attention away from perfectly good $500 million deals [1].
• SpaceX and OpenAI have "accessed every bucket, every pocket, every sofa cushion of private capital"; Heitzmann ranks SpaceX most likely to list, then OpenAI, then Anthropic, with Stripe waiting and never really wanting to go [1].
• Liquid secondary prices are his best public market predictor: buzzy names trade above last round, unbuzzy names below, and crossover investors like Fidelity or BlackRock effectively price private rounds like IPOs [1].
• The old small-cap public market from $300 million to $3 billion has vanished, and venture has "totally subsumed that whole sector," which is why LPs now believe access to the best companies runs through the best VCs [1].
• Software needs more than workflow, which is being commoditised fast: network effects, collaboration and above all proprietary data, because "data is the the oxygen for AI. It's also the moat in AI" [1].
• SaaS is not uniformly dead; the vulnerable ones are debt-laden, PE-owned businesses that cut R&D and milked their users, while a big market plus an excellent CEO makes an indiscriminate selloff a buying opportunity [1].
• Incumbents with real distribution can buy their way forward, as Salesforce did in acquiring Momentum, but public companies need shareholder grace to slow down, self-disrupt and cannibalise pricing [1].