Summary

Mitchell Green sees the SaaS selloff as an estimates reset, not an extinction event, and is buying Procore, Workday, Appian and Toast. He also likes Clearwater Analytics, although it is being taken private. The host counters that Workday is growing only 6.8% while AI attacks seat-based pricing; Green answers that its AI business is growing quickly and that the company still produces roughly $10 billion of revenue and $3 billion of free cash flow, with distribution, data and a balance sheet challengers lack. Expect “dead money for a little while” as analysts cut forecasts, then potential upside once companies beat reset numbers.
AI will create enormous businesses, but Green thinks many of today’s highly valued first-generation companies will fail before the real winners emerge over the next two to five years. In 1999, nobody would have framed social media as the internet’s eventual multi-trillion-dollar outcome; similarly, AI’s defining company may not resemble another call-center vendor or Workday. His updated view is that AI will be “even bigger than we thought,” particularly through productivity gains across support, distribution, manufacturing and drug development.
Green calls ByteDance “the most advanced AI company in the world” and bets China may win the AI race through power, technical resources, scientific talent and execution speed. ByteDance reportedly grows 25–30% annually with substantial profits; Green thinks it could generate $70 billion, $80 billion or even $100 billion of earnings within five years. China can build power plants quickly and repeatedly engineer products more cheaply, although Green stresses that ByteDance winning would not mean Google, Meta or other Western incumbents lose.
The investable dividing line is cash generation, retention and capital structure—not whether a company carries the AI label. “If you don’t have earnings or EBITDA, there is no floor”; for software, Green treats gross dollar retention around 90% as good, 95% as great and 98% as exceptional. He is equally concerned by excessive stock-based compensation because shareholder value is price multiplied by share count, making dilution a hidden but very real cost.
Green’s return discipline is built around being “in the money” within 18 months at a reasonable multiple, then continuously re-underwriting the position. Lead Edge targets two to five times invested capital over three to seven years, roughly a 25% IRR, rather than underwriting every deal as a power-law moonshot. “Buying is glamorous, selling is the job”: at a sufficiently rich ByteDance valuation—he offers $1.3 trillion as an example—he would sell a meaningful portion despite believing the company could ultimately be worth $2 trillion.
Venture’s liquidity drought is partly self-inflicted: funds held marks without returning enough cash. Green advises young managers to sell 5%, 20% or 30% when windows open, because LPs may withhold commitments from funds that cannot return capital. Secondary sales have represented roughly one-third of Lead Edge’s deals, and Green accepts being called a trader because “the investor is my client.”
Green believes at least 50% of venture investors should not be in the business, with too much capital chasing too few assets at undisciplined prices. He calls billion-dollar valuations for people spinning out of OpenAI or Anthropic with little beyond “an idea and a napkin” complete lunacy, while questioning the math of $10 billion-$15 billion funds that require multiple companies to reach extraordinary earnings scale. His preferred setup is to preserve capital for a major downturn within the next decade, avoid many “Gen 1 AI companies,” and invest aggressively in the stronger businesses created afterward.

Deep dive

1. The SaaS selloff is resetting estimates, not erasing incumbents

Green is actively buying Procore, Workday and Appian, while rebuilding Lead Edge’s former Toast position. He also likes Clearwater Analytics, although it is being taken private. His premise is straightforward: “These companies aren’t going anywhere.” Incumbents possess distribution, data and balance sheets, although disruption will still produce new winners, adaptable incumbents and “some incumbents that blow up.”
The host’s Workday challenge is the sharpest test: growth is only 6.8%, seat-based pricing faces cannibalization and incumbent agent products look unimpressive. Green responds that Workday’s AI business is growing quickly and the company produces roughly $10 billion of revenue and $3 billion of free cash flow; comparing its growth with loss-making frontier labs ignores both scale and profitability. He characterizes the company as a high-single-digit to low-teens grower at enormous scale.
Green’s diagnosis of the broader drawdown is more mundane than an AI apocalypse: Wall Street assumed companies that grew 20% last year might still grow 19.5% this year, insufficient deceleration for the law of large numbers. Analysts will cut estimates, investors will avoid names while numbers fall, then companies may begin beating lowered forecasts within a quarter or two. Until then, “it’s probably dead money.”
The host invokes Howard Marks and Duolingo to argue that falling knives may have “no floor.” Green partly agrees: “If you don’t have earnings or EBITDA, there is no floor.” For desired positions, he recommends buying incrementally over a month or on hard down days—accepting that the position may never fill completely because consistently timing the bottom is nearly impossible.

2. Growth-minded leadership and unlevered balance sheets determine who adapts

Green only partially accepts the host’s claim that non-founder-led companies are inherently disadvantaged during AI transformation. The essential trait is a growth mindset: businesses run primarily for margins often lack the willingness or financial capacity to fund reinvention, while entrepreneurs more commonly—but not exclusively—keep investing for growth.
The 1999 retail analogy carries his argument. Six or seven of today’s ten largest US e-commerce businesses, he says, are established retailers such as Walmart, Target, Home Depot and Lowe’s; Sears, Kmart, Montgomery Ward and Bed Bath & Beyond failed. The critical distinction was often leverage: Walmart could “bet the company” on e-commerce, while debt-heavy competitors could not innovate.
When the host argues that 2008 lacked today’s potentially incumbent-destroying technology shift, Green points to the roughly $5.5 billion mainframe market: technology introduced in 1950 still runs many banks. Oracle, Microsoft and SAP are all legacy vendors that remain giants. His wager is categorical: most incumbents will not disappear, particularly those with 90%, 95% or 98% gross-dollar retention.
The Big Tech capex dispute remains unresolved. Apple is spending comparatively little while Google, Microsoft and Meta spend enormous sums; Green’s “honest non-answer” is that the right level is probably somewhere between them. Zuckerberg nevertheless “deserves to go for it”—he built Meta, owns the strategic choice and is compelled to invest while competitors do the same.

3. AI’s largest payoff may arrive after today’s first wave

Green compares the present moment with 1999, when “social media” would not have appeared in an internet investment discussion even though Facebook later created trillions in value. AI’s defining businesses may emerge over the next two to five years, not from today’s obvious call-center or enterprise-software wrappers. “I don’t even know what it is.”
Strategy should match the return model. An early-stage fund seeking 100x outcomes or zeros should keep investing; Lead Edge instead targets two to five times capital in three to seven years, roughly a 25% IRR, with very few zeros and few 10x outcomes. At extreme entry multiples, even 18 months of extraordinary growth may leave an investor underwater.
Green rejects imminent mass unemployment as historically and operationally implausible. A senior banker told him hundreds of thousands of trained back- and middle-office employees would be retrained, not simply discarded. If unemployment somehow reached 10%, 20% or 30%, government would intervene; meanwhile, many regulated companies still cannot let employees use Claude or ChatGPT at work.

4. Retention and operating leverage matter more than an AI narrative

Grafana Labs illustrates Lead Edge’s sourcing model rather than a fashionable consensus deal. It was a bootstrapped, roughly $12 million software company when Lead Edge cold-called the CEO and invested alongside Lightspeed; today it is a much larger, fast-growing beneficiary of AI infrastructure spending. Green’s team includes 18 people aged roughly 22–24 “pounding the phones” for such companies.
PaceMate, a cardiac-monitoring software company, was near $20 million of revenue when Lead Edge invested after raising only $8 million and burning roughly $3 million. It grew about 50% annually and probably did roughly $45 million of revenue the following year. AI can let its service team analyze more device data without proportional hiring—productivity growth without requiring immediate layoffs.
Software economics are not semiconductor economics. Green estimates that cumulative research and development accounts for only about 30% of spending, measured across the life of the average software company reaching public markets; much more goes toward sales, marketing, distribution and customer support. AI can therefore strengthen incumbents by making those functions materially more productive.
Green calls gross dollar retention “the most important number in tech companies” and dismisses the distraction of net retention. Of one year’s customer revenue, 90% remaining without upsells is good, 95% is great and 98% is exceptional. Businesses retaining only 60–80% become especially fragile near $150 million of revenue because sales spending must continually refill a leaking bucket.

5. Casino-like markets reward earnings discipline—and expose dilution

The host cites a Citron Research report that erased billions in market value; Green is astonished that “some random person” can attract 25 million views while investors ignore figures such as Stan Druckenmiller, Howard Marks, Ken Griffin, Steve Cohen, Marc Benioff, Mark Zuckerberg and Jensen Huang. He agrees this is “casinoization,” but sees emotional selling as an opportunity for long-horizon buyers.
Green’s 2008–09 comparison adds scale: indices then moved roughly 8% intraday, with swings approaching 15%, making today “amateur hour.” The host’s alternative—that rational investors should avoid markets detached from value—meets Green’s repeated answer: buy fundamentally strong, cash-generating businesses on earnings multiples when fear makes them available.
Stock-based compensation is his under-discussed reason many internet stocks are still not cheap. Silicon Valley companies can issue enough equity to create substantial dilution, and “market cap equals number of shares times price of shares.” Reported free cash flow is less compelling when employees continually receive a growing share of the enterprise.
Capital allocation provides a signal. Green admires Larry Ellison’s debt-funded Oracle repurchases because Ellison retained his own shares while shrinking the share count; Salesforce’s buyback is less certain in the transcript—Green recalls that it had said it would buy back roughly $50 billion, “or some crazy number.” He expects more companies to establish large buyback programs. The host contrasts that with ServiceNow’s CEO buying only $3 million personally, while Green asks how much stock the CEO already owns and whether the company itself is repurchasing shares.

6. ByteDance’s discount is closing as China’s AI advantages compound

Green values ByteDance against Alibaba and Tencent, which he said had been trading around mid-teens earnings multiples despite limited growth—not revenue or EBITDA multiples. ByteDance grows roughly 25–30% or more and generates substantial earnings; Meta, by comparison, was trading around 25–30 times earnings. Green sees a plausible path to $70 billion-$100 billion of earnings within five years.
Lead Edge bought ByteDance at prices below $200 when China sentiment was much worse, judging the earnings-backed risk/reward exceptional. Alibaba’s stock had roughly doubled off its lows over the prior year. On an IPO, Green first says there is “zero percent chance” of a US listing, then immediately restores uncertainty—“there’s a chance; I have no clue”—before identifying Hong Kong as the likely venue.
“Don’t count China out. I bet they win the AI war,” Green says, while clarifying that “win” is not winner-take-all. China can build nuclear and other power plants quickly, has abundant resources and PhDs, prizes science and technology, and repeatedly reverse-engineers and engineers products more cheaply. The release he calls DeepSeek was therefore unsurprising to him.
Power may become America’s bottleneck. Green imagines a local town hosting a giant data center that employs many construction workers but only 50 permanent local staff while electricity prices triple; resentment follows as coastal investors capture the gains. He expects local regulation and pushback in both the US and Europe, while acknowledging that many climate businesses have been difficult to underwrite because they are capital-inefficient.

7. Selling and secondary liquidity are core investment work

“Buying is glamorous, selling is the job.” Lead Edge continually re-underwrites whether an asset can still double and whether its expected return fits the fund. At ByteDance’s reported $550 billion secondary valuation, Green remains confident and says Lead Edge has been offered higher prices; at a hypothetical $1.3 trillion valuation, he would sell a meaningful amount despite a possible $2 trillion outcome based on $100 billion of earnings at 20 times.
The operating test is whether an investment becomes “in the money 18 months out” at a reasonable multiple. A company growing rapidly is not enough if the original price already capitalized several years of progress. Green distinguishes a good company from a good investment: many A+ companies were purchased at “D- prices” in 2020–21 and still lost investors money.
Lead Edge seeks the intersection of asset quality and price: an A+ company at a B-/C+ or even D- price can deliver excellent risk-adjusted returns, while a D- company at an A+ price is unattractive. Structured secondaries may secure an A-quality company at a B- or C+ price. Its special-situations business recently invested $200 million in one company that began raising again at roughly twice that price only a month later.
Liquidity windows “open and close,” so young funds should sell 5%, 20% or 30% rather than wait indefinitely. Roughly one-third of Lead Edge’s deals have involved secondary sales. Green’s operating rule is blunt: “Marks are opinions, DPI is math,” and managers who consistently return capital are the ones likely to stay in business and keep raising funds for 10–20 years.

8. Too much venture capital has weakened both pricing and governance

Green thinks 50% of venture investors should not be in the business—possibly 70%—because “there’s too much money and there are too many tourists.” The starkest specimen is a person spinning out of OpenAI or Anthropic and raising at $1 billion-$2 billion with “nothing more than an idea and a napkin.” He genuinely asks whether such starting valuations have ever worked.
Fund size creates its own impossible hurdle. Discussing $10 billion-$15 billion platforms and Thrive’s roughly $9 billion growth fund, Green notes that a $100 billion investment may need to reach $250 billion after dilution merely to double. At a mature 10-times-earnings multiple, that implies $25 billion of profit—an achievement available to very few companies.
He nevertheless rejects the claim that only “big or boutique” survives. Benchmark and Index can remain smaller and earlier, accept dilution and still discover the next OpenAI or Anthropic; Index’s attraction is not branding but decades of fund returns and DPI. Menlo’s Matt Murphy also earns credit because its Anthropic investment looked “are you nuts?” when made.
Negative-value investors urge indiscriminate burning, pretend they know how to operate companies and force growth because of the prices paid in prior rounds. Green’s alternative is humility: help a $20 million company recruit leaders who have already scaled from $20 million to $200 million, connect founders with experienced operators, then “get out of the way.” Under-promise, over-deliver and keep helping even after selling.

9. Public markets restore strategic flexibility—and the next crash restores returns

Private valuations can become detached from public comparables: the host contrasts a roughly $150 billion Stripe with $45 billion Adyen, and $9 billion Ramp with $4.5 billion Wix. Green calls the discrepancy “kind of silly” and is surprised more growth funds cannot buy publics such as Workday, Atlassian, Salesforce or Toast when former portfolio companies fall 60%.
Public status also supplies acquisition currency and enterprise credibility. A $150 billion public Stripe could combine with a roughly $50 billion PayPal using liquid shares; private Stripe stock would not satisfy PayPal holders without an enormous cash backer. Likewise, customers assessing vendors such as Veeva, Datadog or CrowdStrike value evidence that the supplier is large and survivable.
Green says one major historical mistake was not matching software-market prices of roughly five to seven times revenue in 2016–18 while Iconiq paid about ten times and won better deals. Lead Edge passed on leading Procore’s next round, then sold its small holding; a $2 million Shopify IPO allocation could have returned its second fund roughly twice over had it simply held. He identifies Procore as the biggest miss.
Green thinks companies below roughly $1 billion-$2 billion can be a nuisance for public-market investors, while $5 billion-$10 billion companies are legitimate public businesses even in a market dominated by passive ETFs. AppFolio was around a $700 million market cap when it went public and later became a roughly 10x example; Shopify was also once a roughly $2 billion public company.
His most anticipated opportunity is “a really bad downturn” sometime within the next ten years because markets and economies do not rise forever. Combined with AI’s productivity boom, that could resemble avoiding internet 1.0 casualties and buying companies founded around 2003–06. The prerequisite is dry powder: “If you don’t have money, you’re out of the game.”

Albert on Optimizing the Probability of Success, Not the Payoff Episode 175 - March 6, 2026