For whom the Gong tolls
Welcome to Cautious Optimism, a newsletter on tech, business, and power. Our regular editor is off today, so expect a typo or two. — Alex
📈 Trending Up: Media bargains? … the stablecoin wars … VA employment … taking on Brex … vertical SaaS … iPhone prices … AI competition (HuggingFace) … coercive marriages? … teamwork …
📉 Trending Down: The Russian economy … the Fed’s balance sheet … MongoDB, after earnings … job tenure … domestic jobless claims … the stock market (again) … interest rates in Europe …
Interest rate cuts in Europe come as the German state works to unshackle itself from its rigid ‘debt brake.’
📚 Reading List: Superintelligence Strategy: Expert Version, by Scale AI’s Alexandr Wang, former Google CEO Eric Scmidt, and the Center for AI Safety’s Dan Hendrycks. Global Sector Trends on Generative AI from SimilarWeb.
The Big Customer Problem plaguing today’s IPOs
At first blush the CoreWeave IPO has lots to like. Rarely has the ever been a company that has grown as quickly as the GPU-based neocloud, and its underlying economics are improving over time (visible in gross margin expansion).
Sure, the company’s debt load is high, its debt servicing costs steep, and its capital expenditures towering. But so long as its core customers keep buying from it, the model kinda works. That’s the pitch, as I understand it.
However, as CO noted during our first look at the company’s S-1, CoreWeave is effectively a Microsoft vassal from the perspective of customer spend concentration. Given that Microsoft’s CEO has stated publicly that his company’s deal with CoreWeave is not something he intends to invest in forever, and new reporting from the FT says that Redmond has “withdrawn from some of its agreements over delivery issues and missed deadlines,” that main account appears unsteady.
CoreWeave is a super interesting company. But I have no idea how to value it, thanks to its customer concentration issues. And I doubt that I’m alone. The same problems present with Cerebras Systems — our full notes here — namely that it depends on a single account (G42 in the case of the chip maker) for a huge chunk of its top line. Cerebras derived nearly shocking 87% of its revenues from its top customer in the first half of 2024, it reported.
The apparent reason why Cerebras has yet to price and list is that after it posted its IPO filing, investors balked. Here’s a headline from last October: “Cerebras IPO has ‘too much hair’ as AI chipmaker tries to sell Wall Street on Nvidia alternative.”
Can we please get a simple, strong tech listing?
American foreign policy is now pro-Russian, anti-Ukrainian:
There’s too much going on to cover everything here, but I think it’s worth stating plainly that:
The United States is undercutting Ukrainian defense by cutting off military intelligence, by sabotaging equipment in the field, and ending military support.
The United States is reducing pressure on Russia by ending cyber attacks, working to limit sanctions, and destabilizing European deterrence at a pace that seems designed to weaken the bloc.
We’re now helping Russia and harming Ukraine. It’s enough to make you weep.
For whom the Gong tolls
Here’s some pretty darn good SaaS news from TechCrunch’s Mary Ann Azevedo dealing with Gong, a startup that sells AI-empowered sales tools:
Gong, a startup that helps companies predict their revenue from potential sales, has surpassed $300 million in annualized recurring revenue, the company announced on Wednesday. [The company’s CEO] didn’t share Gong’s revenue growth, he said it’s in the range of “top-quartile public SaaS companies.”
Excellent! Gong is clearly an IPO-in-waiting. With emphasis on the waiting part of the phrase. The CEO went on to say that an IPO would be “interesting” but not “the most important thing,” with TechCrunch adding that “Gong is nearly profitable and still has plenty of cash from its 2021 round.”
Damn.
We tend to whine about investors wanting to put more capital into private companies, reducing the need for the same concerns to list to raise capital. We also point out that secondary transactions for founders, employees, and investors reduce pressure to list.
But one thing we should probably keep in mind is that a lot of startups that raised historically large rounds took the advice to get close to breakeven when the bottom fell out of the venture market back in 2022. Those companies could look a lot like Gong: Not burning much from atop their nine-figure mountain of ARR.
Gong will eventually list or sell, but I cannot keep the following tweet from a venture investor out of my head when I consider the startups out there like Gong, but with a bit less revenue and perhaps a slightly slower growth rate:
