Aug 12, 2026 · 1h 29m · news
Canva Slashes Growth | Talent Exodus at Google | Revolut's $50B CEO Package | Musk's $55B Terrafab
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this episode of The 20VC Podcast, host Harry Stebbings leads an in-depth roundtable discussion with venture capitalists Rory O'Driscoll and Jason Lemkin covering major shifts in tech and venture capital. Together, they analyze AI infrastructure costs, changing SaaS unit economics, high-profile talent movements at Google, massive founder compensation packages, and private market valuation dynamics.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Harry holds 9.8% of the talking time here. How this is scored →
Harry as informed peer 5.0 Guest teaching 4.7 Guest disagreement 3.9 Harry pushing back 3.9
speaking balance: gold is Harry, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 1:08:30 Jason and Rory clash over $50B CEO compensation and non-founder leadership
Jason forcefully argues that any venture investment not led by a founder is destined to be a zero, while Rory sharply objects to handing over a 16% cut of $300B in value creation without operational performance hurdles.
Hardest push from Harry ▶ 1:28:33 Harry rejects Rory's private equity acqui-hire strategyHarry aggressively shuts down Rory's idea that PE firms should acquire Y Combinator startups for AI talent, bluntly arguing that young founders have zero loyalty and that PE-owned companies are unattractive destinations.
Biggest teaching moment ▶ 36:36 Rory explains internal hyperscaler compute allocation to HarryWhen Harry assumes Jeff Dean had carte blanche compute access at Google, Rory breaks down the harsh corporate realities of public hyperscalers prioritizing high-margin cloud compute and core products over speculative science moonshots.
Harry holds his own ▶ 13:49 Harry counters Rory's ease-of-use thesis with UI disintermediationHarry leverages direct industry insights from interviewing the president of Uber to refute Rory's argument that product ease protects incumbents, demonstrating that algorithmic UI aggregation bypasses standalone apps entirely.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Harry as informed peer | Guest teaching | Guest disagreement | Harry pushing back | Why |
|---|---|---|---|---|---|---|
| Canva's AI Cost Crunch & Growth Deceleration | 5 | 4 | 3 | 2 | Harry introduces Canva's growth slowdown and introduces the concept of the 'fortnightification' of tools where consumer subscriptions bundle single-use functions. The guests expand into a nuanced comparison of prosumer vulnerability versus enterprise workflow defensibility. | |
| The Public vs. Private Market Dilemma for Tech Unicorns | 6 | 4 | 4 | 6 | Harry challenges Rory's premise that retaining users is merely an ease-of-use problem, citing his interview with the president of Uber regarding the existential threat of UI disintermediation. Rory acknowledges the point while exploring whether aggregation layers can survive on top of frontier models. | |
| Incumbent Adaptation & the Palantir AI Playbook | 4 | 5 | 2 | 3 | Harry presses the guests to name pre-AI incumbents that successfully pivoted into the AI wave. Rory and Jason explain why infrastructure vendors naturally benefited while highlighting Palantir's forward-deployed engineers and outcome-based pricing as a rare app-layer exception. | |
| Valuing Late-Stage SaaS & the Power Law of Holding Winners | 5 | 5 | 4 | 3 | Harry brings up LP portfolio dilemmas and cites secondary-sale strategies from Freestyle and Emergence. Jason and Rory push back on online conventional wisdom, explaining that venture returns are dictated by power laws where the rare compounders drive almost all fund returns. | |
| Google's AI Talent Exodus & The Rise of "God Tier" Compensation | 6 | 6 | 5 | 6 | Harry pushes back on the narrative around Jeff Dean leaving Google, arguing Dean had unlimited resources and access to founders. Rory dismantles this view by outlining corporate compute allocation trade-offs between cloud revenue and moonshot research, while Jason introduces the emerging 'god tier' compensation dynamic. | |
| AI Infrastructure Bottlenecks, Data Center Politics & Musk's TerraFab | 5 | 4 | 3 | 4 | Harry criticizes Ro Khanna's proposed data center bill of rights and questions whether local opposition will stall AI rollouts. Jason and Rory offer counterpoints regarding state-level regulatory competition, local electricity subsidies, and the economic scale of Musk's TerraFab project. | |
| Revolut's $50B CEO Package & Founder Governance | 5 | 5 | 7 | 4 | Harry details Revolut's leaked ratcheting CEO compensation structure, prompting a fierce debate between Rory and Jason. Rory argues that a potential $50B equity payout lacks operational governance protections, while Jason insists that non-founder-led bets in venture inevitably go to zero. | |
| Whatnot's $20B Valuation and Investing in Non-AI Disrupted Sectors | 4 | 4 | 2 | 2 | The group explores Whatnot's $20B valuation and the value of investing in categories insulated from direct AI disruption. Rory likens Whatnot to modernizing QVC and eBay live commerce mechanics while Jason notes the durability of e-commerce platforms like Shopify. | |
| Atlassian Earnings Beat and Warning Signs in SaaS Monetization | 4 | 5 | 3 | 2 | Harry prompts a review of recent SaaS earnings beats including Atlassian and Datadog. Jason highlights hidden stress indicators, pointing out that Atlassian's decision to restrict free Loom seats represents short-term base monetization rather than organic product expansion. | |
| HubSpot's Future and the Threat of AI-Native SMB Competitors | 6 | 5 | 6 | 7 | Harry provocatively asks when HubSpot will be acquired by Bending Spoons, leading into a discussion on AI-native SMB competitors. When Rory suggests private equity firms should acqui-hire Y Combinator startups to gain AI DNA, Harry aggressively pushes back, calling PE companies 'shit heaps' and questioning startup loyalty. |
Statements from this episode (28)
Canva lowered its 2026 revenue growth forecast from 30% to 20%
“The facts are that Canva You know, large privately held company in the kind of creative suite space discloses its revenue, even though they're private, and, you know, they were at three billion in gap revenue last year. Going into this year, they're growing at…”
Lemkin's team churned off Canva and Notion due to AI agents
“My, me having churned, and Amelia having churned, we churned just not because, this is a tough one, Canva and Notion, both we churned, not because they're not great apps, we just no longer had any need for them in the agentic area, just no need. Canva and Noti…”
The era of no-code software is slowly ending due to AI
“The era of no code, of things that we can, humans can do without engineering resources, it's slowly winding down. And if it's in chat, GBT, man, I'm just worried.”
Uber president's biggest fear is ChatGPT routing rides purely by price
“I just interviewed the president of Uber. What is his single biggest fear? It's actually the disaggregation of UI or the removal of UI where you say, I want a car, and ChatGPT automatically routes you to Lyft, Uber or another provider based on price.”
Figma missed market share later captured by Replit and Lovable
“I firmly believe a big chunk of Replit and Lovable could have been Figma's if they'd done it.”
Datadog benefits from AI scale without changing its core business model
“Because if you think about Datadog, they sell observability. They sell it to infrastructure vendors. Nothing in their model has changed except there is now an infrastructure vendor who needs to buy a hundred times more Datadog than anyone else has ever bought,…”
Palantir accelerated revenue growth from 18% to 98% following AI pivot
“Palantir went from 18% growth to 98% growth, right? Unprecedented in our lifetimes, right? It may be the one of n or the n equals one.”
Enterprise software companies only pretend to offer outcome-based pricing
“No one does that outcome. They talk about it, but they just tip, they just pretend that they do it. No one puts a two billion dollar deal on the line for an outcome based resolution.”
Canva is currently worth only $12 billion in public market value
“I'd say it's probably worth twelve billion right now. 20% growth at four billion ARR in the current public markets and not decelerating. There's some sort of rule of 40 number that's better, but I'd say it's worth about twelve billion.”
Vibe coding a custom CRM makes no sense for most companies
“This makes no sense for 99.9% of the world, ok? You can't maintain it, you can't build the integrations, it's more complicated. Most of the folks that say that have never used a CRM”
Lemkin's lifetime exits split evenly on holding versus taking early secondaries
“And I literally just did this analysis across my whole lifetime of all the things I've been involved with, personal, angel, venture, who should have sold, and who should have sold early or taken than secondary. And for me, it broke roughly fifty-fifty.”
Google Cloud generates 30% operating margins selling compute to Anthropic
“Every dollar of, every piece of compute that you give to Google Cloud turns into 30% operating margins in a day because they can sell it to Entropic.”
Anthropic will IPO soon after pulling ahead of OpenAI's ChatGPT
“They will go public because this is peak brass ring moment, and you know, you could argue the trends in 27 are tougher. You've pulled ahead of ChatGPT, so just put a nail in the entropic pit, right? You've pulled ahead of ChatGPT comfortably, OpenAI, to a poin…”
Tech compensation features three tiers including seven-figure superstar AI packages
“Now we're seeing three bands of compensation. We're seeing the regular human beings, then we're seeing the AI guys, and then we're seeing the one to five superstars that we're talking about, right? That I have to that I have to find a seven figure package for …”
A $1 million Anthropic stock grant from 2023 is worth $51 million
“Someone who got a million in stock in Anthropic in 23, it's worth fifty one million now, right?”
Representative Ro Khanna is proposing an AI data center bill of rights
“Rep Ro Khanna said he will introduce a data center bill of rights that will give local communities the right to say no to AI data centers.”
Intel completes first equity raise since 1971 IPO for TerraFab consortium
“Intel is part of the TerraFab consortium in such a way, and I just saw it today. I did not know this. Intel completed an equity round, which I read someone, I haven't verified it. It was the first time they raised equity since, like, they went public in seven …”
Any startup investment not run by a founder will be a zero
“Any investment I've made that is not run by a founder is a zero. It's going to be a zero in this age. It's going to be a zero. I look across, now you, we have different portfolios, but the ones I have that are not run by founders, whether they're at twenty mil…”
Seed models must assume 75% dilution and 4x effective entry prices
“When I started, my model was I'm actually paying twice my entry price. That's how I model. Now it's four X. Now I'm playing, now I'm paying, I'm going to suffer 75% dilution, and it's not, and that really means my entry price is four X what it looks like on, b…”
Founders will continue getting Musk-style compensation packages from willing investors
“The baby Elons are gonna get these packages, and it don't really matter what I think, or any of us think, because Enough investors are going to go along with it that they're going to get these packages.”
Late-stage investors design mega CEO incentive packages around 10x return hurdles
“In my limited experience with my portfolio, these mini Elon packages, they're basically all focused on 10 X what the last guy paid. That's what all the late stage investors do. Whatever I paid, 20 dollars, 20 dollars, and I just want 200 dollars for you to get…”
Whatnot recently raised $500 million at a $20 billion valuation
“Background for people, Whatnot raised about half a billion at twenty billion in valuation, right?”
Whatnot's GMV is doubling to $16 billion with a 12% take rate
“You measure GMV, which is about eight billion last year, going to sixteen billion this year, and then they get a 12% take.”
Investors ignore dubious revenue metrics like GMV if growth exists
“I think whether it's literally GMV or not, I think there are plenty of folks getting benefits of revenue that make no sense. It's not even just lying or cynical. I think there's a lot of, I think investors to some extent don't care as long as the growth's ther…”
Datadog's largest customer, OpenAI, slashed its $150 million software spend
“Datadog's story was just, everything's amazing, but our biggest customer, and everybody knows it's OpenAI and know what it says, right, is suddenly realized they maybe don't need to spend a hundred and fifty million dollars and they're spending less.”
Developer-focused SaaS seat pricing faces permanent assault from AI agents
“A lot of their revenue is still developer focused, which, where I think the seat is under assault, a permanent, permanent assault, the seat is.”
Emerging low-end CRM startups are seeing unprecedented revenue growth rates
“The low-end CRM competitors are exploding. The revenue growth from Monaco, Lightfield, Auracel, and others, they're like nothing we've seen before.”
Scaleups acqui-hiring failing YC startups for AI talent is no longer viable
“When rippling is hoovered up 30, and owners hoovered up 20, and Revolut's hoovered up 10, you just, that strategy worked three years ago. Like, it's too late. Like, everyone is just sitting there hitting refresh, hoping these smart YC companies fail so they ca…”