May 8, 2024 · 1h 17m · news
Tom Hulme: Lessons from a 24x Angel Track Record, 275x on Robinhood & Making Billions on Uber |E1150 · 20VC with Harry Stebbings
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
This video features an in-depth interview with Tom Hulme, Managing Partner at Google Ventures, as he shares critical insights on angel investing strategies, venture capital market dynamics, and founder psychology. Through personal anecdotes and frameworks, Tom explores the realities of portfolio management, the risks of overfunding, and the future of the artificial intelligence landscape.
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 25% of the talking time here. How this is scored →
Harry as informed peer 5.4 Guest teaching 3.9 Guest disagreement 2.3 Harry pushing back 3.4
speaking balance: gold is Harry, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 23:22 Tom calls Harry's $20M learning framing 'bananas'
Tom directly and forcefully rejects Harry's assertion that it takes $20–30 million to learn venture investing, calling the dollar-based metric bananas.
Hardest push from Harry ▶ 58:14 Harry challenges the underwriting logic of late-stage StripeHarry aggressively pushes back on Tom's $100M bet on Stripe at a $32B valuation, challenging the financial math and opportunity cost of capital.
Biggest teaching moment ▶ 48:53 Tom reframes the Microsoft/Inflection deal as a hardware grabTom educates Harry on the reality of the Inflection deal, pointing out that Microsoft was simply acquiring GPU compute rather than buying an obsolete model.
Harry holds his own ▶ 54:42 Harry demonstrates incumbent dominance with hard financial dataHarry counters general AI startup optimism by citing Microsoft's $350 million daily cash generation, proving why incumbents hold insurmountable distribution advantages.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Harry as informed peer | Guest teaching | Guest disagreement | Harry pushing back | Why |
|---|---|---|---|---|---|---|
| A Long-Awaited Interview | 3 | 5 | 4 | 3 | Harry shares personal reflections on childhood bullying and conflict avoidance, citing Jason Lemkin to argue that founders do not want feedback. Tom playfully counters that top founders are attuned to feedback and teasingly suggests the issue might just be Harry's specific feedback. Harry accepts the ribbing lightheartedly. | |
| The Angel Experiment and Three Types of Investors | 6 | 4 | 4 | 6 | Harry uses his sample size of 2,700 VC interviews to aggressively lower Tom's optimistic 25% estimate of 'smart' VCs down to 3%. He also pushes back on the idea that bad VCs actively damage companies, prompting Tom to counter with deal structure traps. | |
| The Danger of Deal Structure and TVPI Inflation | 6 | 3 | 2 | 4 | Harry quotes Charlie Munger and Warren Buffett on incentive alignment regarding TVPI inflation and LP reporting. He challenges Tom on practical founder realities when cash is needed, leading to an agreeable dialogue on deal ratchets and liquidation preferences. | |
| The Pitfalls of Overfunding and 'Foie Gras-ing' | 6 | 3 | 1 | 3 | Harry introduces the vivid concept of 'foie gras-ing' startups with excessive capital. When Tom admits his inability to stack-rank his own top angel investments in hindsight, Harry immediately connects this to his thesis on why reserve models fail. | |
| Lessons from Winners, Losers, and 'Hot' Rounds | 5 | 3 | 1 | 2 | Tom breaks down historical exits like GoCardless versus hyper-hyped write-offs like Fab and Jawbone. Harry synthesizes the pattern into a clear thesis on the inverse correlation between seed round heat and ultimate company success. | |
| The Importance of Liquidity and Taking Money off the Table | 6 | 3 | 1 | 2 | Harry demonstrates high domain familiarity by detailing the 'social validity' trap where multi-billion dollar funds write tiny checks that trick naive angels into co-investing. Tom strongly agrees with Harry's explanation. | |
| Caliber Questions: Unfair Advantage and Healthy Paranoia | 5 | 3 | 3 | 4 | Harry asks whether founders really need a unique unfair advantage or if simply spotting an unfulfilled market need is enough. Tom firmly rejects the idea that ideas alone matter, emphasizing that execution is everything. | |
| The True Cost of Learning Venture and Check Size Consistency | 5 | 4 | 4 | 4 | When Harry asks if it takes $20–30 million of burned capital to learn venture investing, Tom explicitly rejects the dollar figure framing as 'bananas'. Harry adjusts his estimate and shares his own strict rule of fixed check sizes. | |
| The Dilemma of Founders Raising Side Funds | 5 | 4 | 1 | 2 | Harry introduces the distinction between active founders doing personal angel checks versus raising external side funds. Tom expands on the 4 S's of venture and compares VC to being a founder on antidepressants. | |
| The Rise of 'Tourist VCs' and Lack of Operating Empathy | 6 | 2 | 2 | 3 | Harry delivers a strong critique of 30-year-old VCs from Bain writing $10M checks without operating empathy. The two share playful banter regarding Tom's Wired articles and Harry's TED talk. | |
| Optimism, Pre-Mortems, and the Multi-S-Curve Fallacy | 5 | 4 | 3 | 4 | Tom references Daniel Kahneman's pre-mortems and questions how anyone in VC can claim complete conviction. Harry pushes back, arguing that projecting conviction is necessary salesmanship to win over founders. | |
| Conviction, Honesty, and the $10 Billion Founder Formula | 6 | 5 | 4 | 5 | Harry argues that stellar founders can easily turn a $2B business into a $10B outcome through second-act product expansions. Tom counters with statistical reality, citing market persistence and competition constraints. Harry pushes back using the Hopin case study. | |
| Sifting Through the AI Frenzy: Why Foundation Models are Bad Investments | 6 | 6 | 3 | 4 | Tom details foundation model economics, comparing them to rapid-depreciating power stations and framing the Inflection deal as a simple GPU cluster acquisition. Harry presses Tom on underwriting OpenAI at $90 billion. | |
| Capitalizing on the AI Application Layer and Incumbent Distribution | 7 | 4 | 1 | 3 | Harry cites specific cash generation numbers for Microsoft ($350M/day) and Amazon to highlight why incumbents are unbeatable in AI distribution. Tom reinforces this with estimates on capital loss across AI tiers. | |
| Psychological Biases in VC: Navigating FOMO, FOOLS, and the Stripe Late-Stage Bet | 6 | 4 | 3 | 6 | Tom describes GV's $100M Stripe investment at a $32B valuation. Harry pushes back hard on the underwriting math, arguing that the opportunity cost of capital makes a 3x return over 7–10 years unappealing compared to index funds. | |
| Founder Archetypes, Startup Clock Speed, and Generating Real Market Signals | 5 | 4 | 1 | 2 | Tom discusses founder archetypes (Lemonade vs CurrencyCloud), startup clock speed, and why free pricing destroys signal. Harry enriches the segment by citing Ryan at Letterdrop on velocity versus speed. | |
| Managing Cultural Debt and Navigating the Remote vs. In-Person Dilemma | 6 | 4 | 1 | 2 | Harry articulates the informal mentorship lost in remote work, describing 'the value in the cracks' during post-meeting taxi rides with senior mentors. Tom agrees, criticizing hybrid work mandates. | |
| Quick-Fire Round: AGI Hype, Defense Tech, Robotics, and the Nature of Founders | 4 | 5 | 2 | 2 | In the quickfire round, Tom reveals key insights, including a strong correlation between founders hyping imminent AGI and their immediate need to raise massive capital. Harry facilitates a smooth conclusion. |
Statements from this episode (61)
Hulme: Venture capital is like being a founder on antidepressants
“Venture capital. It's like being a founder on antidepressants. You basically have all of the highs, just not as high. All of the lows, just not as low.”
Hulme: Founders must avoid passive investors who think they are smart
“There are basically three types of investors. You've got smart investors that know they're smart and they're going to add value. Then you've got passive investors that are going to stay passive and they're not going to get in the way. Both of those are absolut…”
Stebbings: The majority of startup founders do not want feedback
“The majority of founders actually just don't want the feedback.”
Hulme: Second-time founders strongly prefer passive investors over value-add ones
“First time founders are looking for investors that add a lot of value. Those that are second time around understand it, bias, and skew wildly towards the passive.”
Hulme: Heavy deal structure in venture investments damages businesses
“I think a lot of structure in a deal can actually damage a business.”
Hulme: Priced rounds are declining as founders avoid valuation write-downs
“Oh, I think we're seeing that firstly price rounds are less common because people aren't willing to accept the new and fair valuation. So you get convertible notes.”
Hulme: Founders should accept realistic valuations instead of complex deal structures
“I think actually, if you want to maximize value for the business, you're much better off accepting the valuation That's closer to reality in the moment and building from there.”
Hulme: The tech IPO window will not open anytime soon
“Particularly since kind of the IPO window isn't open anytime soon.”
Hulme: Premature scaling inflates costs and destroys startup adaptability
“And I think what you're talking about is when the money's used for premature scaling, it's really damaging because costs go up. When your costs go up, ironically, you actually become less adaptable. So your clock speed goes down. So if your primary job in a bu…”
Hulme: Second-time founders handle overfunding better than first-time founders
“Again, one of the reasons second time founders do so well is they would be less affected. No one is unaffected, but they would be less affected.”
Hulme: Pre-2015 angel investments achieved 4.5x DPI and 24x TVPI
“You could take a batch of my angel investments Pre- twenty-fifteen. And I tracked this roughly 27 portfolio companies, about 4.5 X DPI, about 25 X or 24 X TVPI.”
Hulme: Angel investors should refrain from making follow-on investments
“It led me to draw the conclusion as an angel, I shouldn't follow on. So I had examples of companies that would go up sort of 50 X, my pro rata allocation in the next round would be a million dollars plus, and they went to zero.”
Hulme: GoCardless has reached nine-figure annual recurring revenue
“Look, GoCardless would be an example where the business kind of nine figure ARR, they have just continued growing.”
Hulme: Angel investments in Fab.com acquired firm and Massive Health went to zero
“So I invested in a company that was acquired by fab.com. So I thought I had a lot of money in equity in fab at one point. I invested in really brilliant team in California called Massive Health that were acquired by Jawbone. I thought that was going to be very…”
Hulme: Founders and VCs should take secondary liquidity to minimize regret
“And my recommendation often to founders, often to VCs is to take some money off the table. It's a classic place. I like to apply a kind of regret minimization framework. Will you regret taking 10% or 20% off the table in this round? Probably not. Will you regr…”
Hulme: Angel investors frequently rely on others to do due diligence
“I think too many people in our industry, particularly angels who have other jobs, full-time jobs, We'll take the view that probably someone else has done the work. I actually think often it's surprising how people haven't done the work.”
Hulme: Asking founders about product is a lens to evaluate thinking
“I spend a lot of time with founders, and I'll ask them a lot about the product they're going to build. I know that that's probably not what they'll end up being successful with. In fact, empirically, the product often changes a lot. But I'm asking about that p…”
Stebbings: Great entrepreneurs do not start in corporate Bain or Oxford jobs
“I don't think great, great entrepreneurs first made money from getting a job at Bain after three years at Oxford. Like they did something before.”
Stebbings: The best founders openly acknowledge what terrifies them
“The real pattern is the best founders say, are you kidding me? Here's 10 things that are challenging me, and I'm terrified about them, and I'm like, okay, let's go through them one by one.”
Hulme: Being too early in a market equals being wrong
“Being too early is tantamount to being wrong.”
Hulme: VC learning depends on deal cadence over time, not cash losses
“The learning cycle is number of deals and time To see how they do. And so for me, it looks more like five deals over five years to figure out, actually, if you're building the muscle and you're learning, then it looks like an absolute quantum of cash.”
Hulme: Aspiring angels should write $5k checks rather than waiting for $200k
“If your job is to learn and to demonstrate you add value, then just write small five K checks. I see so many people that never start angel investing because they think they have to invest 200 K at a time. It's crazy. If you believe you've got to learn, you wan…”
Stebbings: I write $25,000 angel checks consistently across all investments
“And I just do 25 every time. Every time I'm 25.”
Hulme: Active founders angel investing personal capital provides high-value empathy
“Yeah, so the former, I think is actually great for the portfolio companies they invest in, because I think they're in our category of smart, smart. Like, they will be specialists at something, and the founder will be able to ask them for advice, and there's no…”
Hulme: Active startup founders raising side funds is concerning
“Now, when you start talking about raising funds, that worries me a lot because building a company is ridiculously hard. Like there's only a handful of people that have ever managed to kind of build multiple companies concurrently. And this idea of having a sid…”
Hulme: Salesmanship is the essential fourth pillar of venture capital
“The fourth S that I increasingly think is important is actually selling salesmanship. Like, great VCs are selling to LPs, they're selling to founders to take their money, and then increasingly, they're also selling to, you know, exec hires to go into portfolio…”
Hulme: Great board members act as shock absorbers for founders
“Great board members are like shock absorbers. They kind of reduce the highs and they will cushion the blows and let founders know the negatives aren't as bad.”
Stebbings: Young VCs from Bain lack real operating and fundraising empathy
“I just see so many thirty-year-olds come out of Bain who are writing ten million dollar checks at large funds, and I'm like, you have no fucking idea how hard it is to raise money to run a business.”
Hulme: Venture capital is undoubtedly a game of access
“Look, without doubt it is. I think it would be really unfair to say that actually everyone has even access.”
Hulme: Investors underestimate winners by assuming single S-curve growth trajectories
“We have a sort of bias that whenever we're looking at businesses, we see them on an S curve. It's very difficult to imagine new S curves, but the best businesses will continuously put themselves on new S curves and create new opportunities.”
Hulme: Great founders collect data early to preserve option value
“Most of the best founders I've ever worked with have collected data without knowing really what it might be used for, but they've instinctively known there's option value in it.”
Hulme: Absolute conviction in VC investments is irrational and unrealistic
“I don't understand in our industry how anyone can have complete conviction on anything. That makes no sense to me. Like I studied physics at university. We would go through a proof and I still didn't have complete conviction that I got it right. And then now w…”
Stebbings: Great founders and teams separate $2B from $10B companies
“I think the difference between a two billion and a ten billion dollar business is a fucking great exec team and founder.”
Hulme: Ex-rocket ship employees make the best startup founders
“I think one of the best groups of founders are those that have come out of the rocket ship companies because actually they kind of instinctively know they need to aim at big opportunities and they know how to run a Fast growth business.”
Hulme: Requiring every VC investment to be a fund-returner is flawed
“I think, ah, a lot of VC strategy is a lagging indicator of what did work in the past, and the test, Or the experiment that worked very well in the past is funds with 25 portfolio companies, power law of returns, and one or two return the whole fund, and then …”
Hulme: GV manages $10B with 10 to 20-year holding periods
“So our approach at GV is primarily be founder first, and we can take a very long time. You know, we have ten billion dollars under management. Alphabet does not put us under time pressure. We're less worried about IRR than we are absolute returns. And so we fo…”
Stebbings: Hopin founder Johnny Boufarhat cashed out $200M in secondary sales
“And so Johnny took two hundred million off.”
Hulme: Virtual event platform Hopin had product-COVID fit, not product-market fit
“Is Hopin, does Hopin have product market fit or product COVID fit? It clearly had product COVID fit. It was, you know, no one was excited to do events online. They were forced to do events online.”
Hulme: Mark Zuckerberg will own 14% of world's H100s by year-end
“He'll have 350,000 H-one hundreds by the end of this year. That is 14% of the world's H-one hundreds, and he's going to open source the result.”
Hulme: GV has avoided investing directly in foundation model startups
“Now we have made investments in Gen AI, but more in infrastructure, more in the application layer, more in the sort of picks and shovels to support, but we've not thus far invested in foundation models.”
Hulme: Generative AI is a sustaining innovation, not internet-scale creative destruction
“I think one of the frustrations with Gen AI, as the technology is commoditizing so quickly, is it's a sustaining innovation. It's actually going to get sprinkled across all businesses to lower costs in call centers or to improve the product in personalization.…”
Hulme: Microsoft acquired Inflection primarily for its 12,000 H100 GPU cluster
“I do not believe that Microsoft were buying the Pi model. I think Microsoft are GPU constrained, and we're buying a cluster of 12,000 H-one hundreds.”
Hulme: GV would not invest in OpenAI at a $90B valuation
“I would struggle To make that investment today. No.”
Hulme: Expecting AI researchers to build profitable businesses is ridiculous
“The idea that they can all go out and create highly profitable, fast moving businesses seems ridiculous.”
Hulme: 90% of AI foundation model capital will go to zero
“Foundation models, 90%. Application layer, 70%. And in the incumbents, the value going into the incumbents, only 20%.”
Hulme: 60% to 70% of Fortune 500 companies use Microsoft Copilot
“If you look at the proportion of the, like, Fortune 500 that are using Copilot, I think it's 60 or 70% I heard last week.”
Stebbings: Microsoft generates $350 million in cash daily
“Microsoft is throwing off three hundred and fifty million in cash a day.”
Hulme: GV invested $100M into Stripe's Series G in 2020
“We invested a hundred million bucks into Stripe in 2020. We extended the series G round”
Hulme: GV invested in Lemonade Series A despite zero insurance experience
“We invested in Lemonade, Series A, wonderful founders. One, Shai, had done Fiverr, product and design guy. Daniel, lawyer, had done hardware, and they were going into insurance. Combined insurance experience between those two? Zero.”
Hulme: Currencycloud sold to Visa for $1 billion
“We invested in CurrencyCloud. Mike Lavin, the founder, I think had 30 years experience in fintech. He was so well placed to understand actually what those buyers wanted and ultimately sold the business to Visa for a billion dollars.”
Hulme: Execution clock speed is the primary driver from zero to one
“I think every startup is a series of unanswered questions and the best founders choose the order in which they answer the questions, and they answer them extremely efficiently, and that is basically speed of execution.”
Hulme: Free products generate bad feedback because users don't value them
“I see a lot of businesses, and it's slightly different in consumer, but they will not charge their early adopters, they won't charge their design partners, they won't charge their ICPs, and the problem with that is you don't know if it's valued. And generally …”
Hulme: Generative AI minimizes the long-term impact of technical debt
“I actually think technical debt is almost less of an issue over time, particularly with generative AI, which is actually incredibly forgiving of how you integrate it. You don't have to have like perfect systems like you perhaps used to, or certainly not as mon…”
Hulme: Best founders execute aggressive cuts rather than repeated shallow cuts
“I think it's one of the reasons you've seen the best founders are quite aggressive in cuts. They never do shallow cuts too often. They see that burning platform as a way to actually shift the culture and maybe move this cultural debt to the side.”
Hulme: Staggering hybrid attendance in small offices is a false economy
“That's a classic false economy, where people say, we can have a small office, and then we can have a third of the team in at any one time. Much better to have the full office and have all of the team in a third of the time.”
Hulme: Early-career professionals must work in-office for critical learning
“I think people early in their career, being in the office is a ridiculously important learning opportunity. You learn the interpersonal stuff. You can shadow people in a way that's pretty hard on a Zoom call.”
Hulme: Nothing will reach $600M revenue in 2024
“I think they'll do six hundred million dollars revenue this year. They've sold three, three million devices so far”
Hulme: AGI hype correlates directly with founders' capital fundraising needs
“I think there is a Correlation between how aggressively people predict AGI is coming and how much capital they need to raise. So you've got Emodi from Anthropic, you've got Altman, OpenAI, Musk, XAI, all of these guys are saying that it's just around the corne…”
Shorts cut from this episode
▶ 3 roles of a VC 💪 · 20VC with Harry Stebbings (@27:04)
▶ Why Microsoft is winning AI 🤖🚀 · 20VC with Harry Stebbings (@55:14)
▶ How to become an Angel Investor 💰🪽 · 20VC with Harry Stebb (@24:12)
▶ Why working from home is BS 🏡❌ · 20VC with Harry Stebbings (@1:08:40)
▶ The 3 types of investors 💰 · 20VC with Harry Stebbings (@0:14)
▶ Biggest lessons to win in Venture 🏅 · 20VC with Harry Stebb (@0:01)