What this covers

Neil Mehta of Green Oaks talks with Invest Like The Best about how his firm picks companies and what it has gotten wrong. The central tool is the Jaw-Dropping Customer Experience Framework: a great experience usually comes from breaking a trade-off, doing something operationally painful that competitors would dread, and doing it for customers who could not have described the problem themselves. Coupang is the worked example. Its first-party delivery lifted 30-day cohort retention from the 30s percent to the 60s, and Mehta says the real signal was customers who would be devastated to lose the service. The conversation also covers Green Oaks' early insurance thesis, built on an Insurance Penetration S-Curve against GDP per capita.

The range is wide. Mehta defends a Coverage vs Conviction Matrix view in which most venture firms are coverage factories, and he argues that competition for the top founders has actually eased. Several claims are contested. One is that Growth Persistence defines the best software companies. Another is that a 35% net IRR expectation for venture is too high and belongs in the teens.

The harder stories are about holding on. Carvana fell from about $300 to about $5 while Green Oaks kept buying, a bet on solvency and fixing per-unit losses, and Mehta judged the CEO by how he handled his people in the crisis. Mehta also names mistakes: following mentors who dismissed Elon Musk, underrating micromanagers, and treating crypto as if the Laws of Great Businesses had changed. He also describes how he thinks about DeepSeek and AI model releases.

Sharpest takeaway

Mehta argues that a very small number of founders who build jaw-dropping customer experiences create most long-run value, so venture investors should concentrate on finding and partnering with those founders rather than maximizing deal coverage.

  • Great businesses are built by founders who break trade-offs to delight customers at a differential rate, and this is the repeatable archetype Green Oaks screens for.
  • Coverage-driven firms dilute fidelity and insight, while concentrated, first-meeting-first investing lets a firm win the top 10-15 founders each year.
  • Investors must hold through volatility and reason from durable business laws rather than short-term momentum or hype cycles.

The claims · ranked59 claims · weighted by value

This asset isn't compiled yet

You're seeing its claims, ranked. Compile it to build the argument threads, weight them, and check each claim against your library — the full view.

0.79

Insurance penetration as a share of GDP follows an S-curve against GDP per capita, so a levered bet on a developing country's growth is to own its best consumer insurance company, a thesis that drove Green Oaks' early holding-company strategy.

causalhigh valuecontestednovelty 4/4durability 3/4· Neil Mehta

“if you believe in a country's GDP per capita growth, a levered investment is to buy the best insurance company, the PNC consumer insurance company”

0.73

Most venture firms have become coverage-optimizing factories that chase every quarter's Series A and B deals; this is the right model for roughly 90% of the industry, but for the 10-15 best founders each year it is the wrong way to work, because a founder should be able to find the firm first and get the firm's full commitment.

normativehigh valuecontestednovelty 3/4durability 3/4· Neil Mehta

“I think like 90 something plus% of our industry should work that way. But I think for like the 10 to 15 best founders each year, that's precisely the wrong way to work”

0.73

Growth is an output rather than an input, and the best technology companies show high growth persistence (the next year's growth is 80-90% of the prior year's), so very high growth is healthy for software and bits companies even if it causes some things to break.

causalhigh valuecontestednovelty 3/4durability 3/4· Neil Mehta

“Growth is an output, not an input... the best companies have extraordinary high growth persistence or growth endurance”

0.70

Great founders look approximately the same while bad founders look different, so there is a repeatable archetype for a great founder that, once learned, makes the investing process repeatable.

normativehigh valuecontestednovelty 3/4durability 3/4· Neil Mehta

“every great founder looks approximately the same. Every bad founder looks different... there's an archetype for for a great founder”

0.66

Coupang's first-party delivery capability raised 30-day cohort retention from roughly the 30s percent for the overall market to the 60s percent for Rocket customers, and the real signal of success was customers who said they would be devastated to lose the service.

normativehigh valueestablishednovelty 2/4durability 2/4· Neil Mehta

“average retention pre1p delivery was probably in the 30s... for Rocket... was in the 60s on a cohort retention basis”

0.64

Roughly 1% of S&P 500 companies account for about 90% of its value, and those companies were growth companies that reallocated free cash flow away from legacy incumbents over decades.

factualhigh valuecontestednovelty 2/4durability 3/4· Neil Mehta

“I think 1% of the S&P 500 make up like 90% of the value and most of those were growth all of those were really growth companies”

0.61

In venture, the best and worst companies at Series B often trade at roughly the same multiple within the same 12 months, so few investors can tell the difference, unlike public markets where expert analysts differ widely on earnings forecasts.

factualhigh valuespeaker onlynovelty 4/4durability 3/4· Neil Mehta

“every single one of those had some other company that traded like in its sphere of competition that traded at approximately the same turn”

0.61

Growth persistence, meaning the ability to sustain very high year-over-year growth for a long time, is the defining trait of the best technology companies, more than any single growth rate.

factualhigh valuecontestednovelty 2/4durability 3/4· Neil Mehta

“the best companies have extraordinary high growth persistence or growth endurance”

0.60

Investing is a game of reducing noise rather than complexity, and the investors who win are those with the temperament to simplify their job so that roughly 110 points of IQ suffice, since few peers pursue that discipline.

normativehigh valuespeaker onlynovelty 3/4durability 4/4· Neil Mehta

“Investing is not a game of... it's a game of reducing complexity. To me, it's a game of reducing noise”

0.60

Many great companies were built by holding ideas loosely: Mehta says he changes his mind often and has no pride in authorship, and the moment he writes an opinion in a public letter it becomes a position he feels compelled to defend.

normativehigh valuespeaker onlynovelty 3/4durability 4/4· Neil Mehta

“the moment I wrote it down, it became a perspective that I had to defend”

0.60

The Wright brothers analogy shows that business has fixed laws: great businesses follow the same laws regardless of technology wave, and Green Oaks' costliest errors (e.g., crypto with no board or auditor) came from believing the laws had changed.

normativehigh valuespeaker onlynovelty 3/4durability 4/4· Neil Mehta

“the laws of great businesses are the laws of great businesses... whenever we've screwed it up at Green Oaks, it's usually been because we ignored the laws”

0.57

The single most important metric for a company is whether its people, especially the most important ones, believe their best days are ahead of them rather than behind them, and this energy can be detected very early.

normativehigh valuespeaker onlynovelty 3/4durability 3/4· Neil Mehta

“Are your best days ahead of you or behind you... I think if you were running a company, that's the one stat you should care about more than anything else in the world”

0.57

The best way to judge a CEO in a crisis is how they handle the human side and slow decisions down; Carvana's CEO Ernie spent his time on employees' children and team morale and methodically ran A/B tests on cuts rather than making snap moves, and this was why Green Oaks trusted him.

normativehigh valuespeaker onlynovelty 3/4durability 3/4· Neil Mehta

“There's very few CEOs in the fog of war. When things speed up, people start to make snap decisions very quickly. What impressed me most about Ernie at that moment in time was how he just slowed everything down”

0.57

Masayoshi Son's SoftBank invested in Arm when a semiconductor analyst listed ten reasons it would fail, and Mehta argues the analyst missed the growing market, so investors should be judged not by momentum but by how they stand by entrepreneurs when the going gets tough.

normativehigh valuespeaker onlynovelty 3/4durability 3/4· Neil Mehta

“Sometimes you can measure investors not by figuring out what the where the momentum is going, but when the going gets tough, how they stand up for entrepreneurs.”

0.57

Mehta has learned that the firm's own internal errors, not external rivals, are the main threat to its performance, so it spends little time analyzing competitors.

normativehigh valuespeaker onlynovelty 3/4durability 3/4· Neil Mehta

“if we're going to screw it up or lose it it's usually something we're going to do internally”

0.57

Many great internet and technology businesses are hidden inside bad P&Ls, so Green Oaks is happy with volatility and follows founders through rough patches, since nearly every business it held more than five years went through a period when its fundamental premise was questioned.

causalhigh valuespeaker onlynovelty 3/4durability 3/4· Neil Mehta

“Often times in internet and technology uh good businesses are hidden in bad P&Ls... we're happy with volatility”

0.57

Mehta's public persona is intentionally low-profile, and he changes his mind frequently rather than defending past views, since public statements turn opinions into identity that must be defended.

normativehigh valuespeaker onlynovelty 3/4durability 3/4· Neil Mehta

“the moment I wrote it down, it became a perspective that I had to defend”

0.57

Carvana's stock fell from about $300 to about $5 a share, yet Green Oaks kept buying, reasoning that the company could survive and fix its unit economics (about $3,000 per unit EBITDA loss plus about $2,000 interest per unit) and that its debt was reflexive with the equity thesis, so the investment was a two-part bet on solvency and scale.

causalhigh valuespeaker onlynovelty 3/4durability 3/4· Neil Mehta

“it's a two-part investment for us. One was the company won't go bankrupt. Second is is this a business that could go from 400 500,000 units to 2 million, 3 million, 5 million units over time”

0.57

Mehta now believes Green Oaks' biggest early learning was about micromanagement: the firm wrongly followed mentors who said Elon Musk was not backable, outsourced the primary diligence, and got feedback that read worse than it was, so the firm now values micromanagers, fast firers and people in the weeds.

normativehigh valuespeaker onlynovelty 3/4durability 3/4· Neil Mehta

“We like micromanagers. We like people that are in the weeds. We like people that uh fire fast”

0.57

When a model company releases a cheaper, faster model such as DeepSeek's R1, Green Oaks does not chase the benchmark; it brings the abstraction level back to what the development means for customers and works backward from that, because a large investment with 12-month catch-up does not fit the laws of a great business model.

normativehigh valuespeaker onlynovelty 3/4durability 3/4· Neil Mehta

“We try to bring the abstraction level back up to what does this mean for customers and then we work backwards from that”

0.56

Venture capital's 35% net IRR long-run expectation is too high for any asset class; it should be in the teens, which would let the industry deploy far more capital and still attract investors.

normativehigh valuecontestednovelty 2/4durability 2/4· Neil Mehta

“why should people have earned 35 net IRS for the course of like many decades? Like that's too high for any asset class. It should be in the teens”

0.56

A jaw-dropping customer experience (JDCE) is typically built by breaking a trade-off, doing something technically or operationally very difficult that competitors would find nightmarish, and doing it from a customer-centric perspective, because customers usually cannot articulate their pain points themselves.

causalhigh valuespeaker onlynovelty 2/4durability 4/4· Neil Mehta

“the steps to creating a JDCE, a jaw-dropping customer experience, it usually starts with like breaking trade-offs”

0.56

Business success is governed by durable laws, like aerodynamics for the Wright brothers, and great businesses follow the same laws regardless of technology hype: delight customers, break trade-offs, build a competitive advantage, and operate in a large market.

normativehigh valuespeaker onlynovelty 2/4durability 4/4· Neil Mehta

“the laws of great businesses are the laws of great businesses”

0.54

Venture competition has actually become less intense for great companies even though capital supply has grown roughly 100 times since Don Valentine's Nvidia investment, because most firms are specialized (by geography, sector or stage) and the number of firms competing for the top 10 to 15 founders is very low.

causalhigh valuespeaker onlynovelty 3/4durability 3/4· Neil Mehta

“Two things are allowed to be true at the same time, which is our space has too much capital... it's actually less competitive for great companies”

0.54

Green Oaks' fund sizes have grown because they are set by the ability to write 500 million to 1 billion dollar checks to great founders, not by returns; the firm has held about 10 to 12 companies per fund (historically up to 15) and has 55 companies across about $15 billion AUM, so fund size is a function of check size, not company count.

factualhigh valuespeaker onlynovelty 3/4durability 3/4· Neil Mehta

“our largest investments are 500 million to a billion plus in size... that number has served us pretty well”

0.54

Green Oaks' frontier-insurance holding company (GGH) was among its biggest mistakes: frontier insurers often lacked audits and solvency rules, Rwandan premium economics were deeply loss-making, and a court claim revealed a political, opaque claims system that made the business unworkable.

factualhigh valuespeaker onlynovelty 3/4durability 3/4· Neil Mehta

“Nobody really audits these companies and and and if as long as you continue to write more premiums the next year, you could kind of make it work”

0.54

Coverage-optimized venture firms sacrifice fidelity and insight: a firm that tries to never miss a Series A by hiring a very large team cannot consistently identify exceptional investments, and Mehta says he has never met a firm with more than a few good investors.

causalhigh valuespeaker onlynovelty 3/4durability 3/4· Neil Mehta

“if you are driving for coverage... then what you're sacrificing is fidelity and insight. I've never met a firm that's had more than a few good investors.”

0.54

Green Oaks' winning edge is differentiated insight delivered at speed and velocity: it has built a small team flywheel that increases information velocity and asymmetry, which in the last couple of years allowed it to move on very large checks.

causalhigh valuespeaker onlynovelty 3/4durability 3/4· Neil Mehta

“Our favorite combination is when you have the same speed and the same information, but you have differential insight”

0.53

Having unreasonable expectations is a competitive advantage: after Whiz staff left for the war in Israel, Mehta refused to lower expectations and the company had one of its best quarters, and Coupang's 18% year-over-year growth was pushed back up toward high growth instead of accepting the 30% norm that Amazon set.

normativehigh valuespeaker onlynovelty 3/4durability 2/4· Neil Mehta

“having unreasonable expectations is a competitive advantage”

0.53

During the 2023 SVB collapse, Green Oaks agreed within about 30 minutes to invest $500 million in Rippling after Parker Conrad's Friday morning call, because Rippling was not in financial trouble but needed to ensure its customers were paid on Monday.

factualhigh valuespeaker onlynovelty 3/4durability 2/4· Neil Mehta

“It took us about uh 30 minutes to agree to invest 500 million”

0.53

Green Oaks prepares so deeply before a first meeting that it feels like a fifth or sixth meeting; from tens of thousands of meetings per year it picks roughly 200 where it expects to be a differentially great partner, and it prefers to visit companies rather than have founders come to it.

factualhigh valuespeaker onlynovelty 3/4durability 2/4· Neil Mehta

“There's probably 200 that happen a year where we think we be a differentially great partner to that person... Our first meeting should feel more like a fifth or sixth meeting”

0.53

Green Oaks wrote a $500 million check to Trip Actions (now Navan) within about four days in 2020, at the depth of COVID, when most investors were retrenching, and that aggression took the company from roughly number four or five in its market to the top two.

factualhigh valuespeaker onlynovelty 3/4durability 2/4· Neil Mehta

“we'll write uh you know sort of unlimited number up to 500 million... it took us 4 days or something like that”

0.52

Coupang's move to one-day, one-P delivery was extraordinarily hard: unit economics broke at first, and it took two to four years of building warehouses, routing software, localized distribution points and inventory systems before the flywheel worked, yet cohort retention rose from the 30s to the 60s percent.

factualhigh valuespeaker onlynovelty 2/4durability 3/4· Neil Mehta

“it took years, it took two to four years to really start to build that flywheel so it worked”

0.52

Only a very small number of the world's founders produce a significant proportion of the value humans enjoy, and building a jaw-dropping customer experience (JDC) is a core tenet for finding them because it requires breaking trade-offs that competitors think are impossible.

causalhigh valuespeaker onlynovelty 2/4durability 3/4· Neil Mehta

“a very small number of the world's founders are going to produce a significant proportion of the value that humans enjoy”

0.52

Focus means saying no to everything else; the best founders have the ability to identify the single most important thing and block their calendars for it, as Coupang's CEO did by spending two weeks blocking out time to negotiate cost of goods sold.

normativehigh valuespeaker onlynovelty 2/4durability 3/4· Neil Mehta

“Focus focus means saying no to everything else, everything else at the cost of doing what the single most important thing is”

0.49

Mehta bought buildings on Filmore Street in San Francisco through a nonprofit as a deliberately bad financial investment (buying at a 5.25% cap rate and leasing to restaurants at a 3% cap rate) to preserve the street's vitality, arguing that San Francisco is a unique hub for building the future that policy is at risk of killing.

normativehigh valuespeaker onlynovelty 2/4durability 2/4· Neil Mehta

“I'm buying stuff at like a five and a quarter cap... I'm putting in like a mom and pop restaurant at a threecap which like barely pays its rent”

0.49

Good founders should be judged by the intensity they bring to managing teams: Green Oaks prefers founders who micromanage, stay in the weeds, and fire quickly, because the firm itself runs a very tight, intense team.

normativehigh valuespeaker onlynovelty 2/4durability 2/4· Neil Mehta

“We like micromanagers. We like people that are in the weeds. We like people that fire fast.”

0.49

Green Oaks's closest Carvana decision involved conviction that Carvana would become the Amazon of cars, but the firm did not invest near $500 million during COVID, and that missed opportunity is one of its big mistakes; the stock later went from about $300 to near $5.

factualhigh valuespeaker onlynovelty 2/4durability 2/4· Neil Mehta

“we ended up not proceeding and investing I think it was going to be about 500 million”

0.47

Yuri Milner broke Mehta's mental model of growth investing by making a large, category-defining private investment in Facebook at about 10 billion dollars enterprise value while the company was still losing money, demonstrating that private companies can be bought at that scale outside the usual revenue-multiple approach.

factualhigh valuespeaker onlynovelty 2/4durability 3/4· Neil Mehta

“Yuri broke that mental model for me.”

0.46

Green Oaks used to show investors 92 percent Series B coverage as a badge of pride, but it now focuses on separating the vital few opportunities from the trivial many, doing fewer meetings and with the same team, a shift Mehta calls a sea change.

factualhigh valuespeaker onlynovelty 2/4durability 2/4· Neil Mehta

“We used to show this slide to our investors of here's how many series B's happened and we had 92% coverage.”

0.44

Truly remarkable founders rarely build bad businesses because they think in terms of jaw-dropping customer experiences, competitive moats and large TAMs; founders who have not figured this out are usually very young and learn fast.

causalhigh valuespeaker onlynovelty 2/4durability 2/4· Neil Mehta

“I don't think there are many truly amazing founders that are building bad businesses”

0.44

Relying on a permanent holding-company balance sheet was rejected for Green Oaks because the Frontier insurance experience demonstrated the firm did not want to be in those businesses; the 150 million holding company was the single largest mistake at Green Oaks.

causalhigh valuespeaker onlynovelty 2/4durability 2/4· Neil Mehta

“it was one of the single biggest mistakes we've made at Green Oaks”

0.31

Green Oaks was founded with a small team and early funding that depended on a specific pitch: describing a set of ideas in Indian and Southeast Asian internet companies and a 20 to 30 year opportunity, rather than promising to make money, which resonated with Henry Kravis.

factualspeaker onlynovelty 2/4durability 2/4· Neil Mehta

“this is the beginning of a 20 or 30-year opportunity ahead of us”

0.30

Successful investors must choose between aiming for the Hall of Fame of returns or the Hall of Fame of AUM; Mehta chose returns, which means limiting fund size to the point where investing does not reduce returns.

normativespeaker onlynovelty 3/4durability 2/4· Neil Mehta

“you have to decide whether you want to be in the hall of fame of returns or the hall of fame of aum”

0.28

The competitive set for top founder investing is best understood as a 2x2 of reducing noise, and the reason few investors remain serious is that most have layered mandates such as Brazil fintech or Y Combinator companies that make their work noisier.

causalspeaker onlynovelty 2/4durability 2/4· Neil Mehta

“there are firms whose only job is to do fintech in Brazil. Their firms that their only job is to do everything that comes out of Y Combinator”

0.28

A few early founders and partners (Henry Kravis, Yuri Milner) shaped Green Oaks: Kravis committed after one pitch and visited the young team, and Milner's willingness to make large investments in private category-defining internet companies broke Mehta's mental model of growth investing.

factualspeaker onlynovelty 2/4durability 2/4· Neil Mehta

“Yuri broke that mental model for me. I mean, he was the first one I remember making a large-scale sort of category defining investment”

0.26

Green Oaks's firm culture is intensely high-performing but not lavish: partners work about 80 hours a week, have no beach houses, avoid Twitter, and focus exclusively on understanding companies.

factualspeaker onlynovelty 1/4durability 2/4· Neil Mehta

“We have no beach houses. We like being in the office, you know, 80 hours a week”

0.26

San Francisco is uniquely important because it harnesses the aspirational culture of company building better than anywhere else, and progressive policies such as anti-business, high-tax, and anti-growth positions are imminently fixable threats to it.

normativespeaker onlynovelty 1/4durability 2/4· Neil Mehta

“I don't think there's anywhere else on earth that's anywhere like it.”

0.26

Mehta's cited anecdote from Coupang's video customer interviews, in which a mother cried about diapers arriving the morning of, shows that a strong JDC produces emotional response far beyond what a net promoter score of nine captures.

normativespeaker onlynovelty 2/4durability 2/4· Neil Mehta

“That is not a good that's not an MPS score of nine. That is a jaw-dropping customer experience.”

0.26

Mehta's realization about Tencent came in 2008 at the Beijing Olympics, when people had phones glowing with QQ, which was adding about 30 million subscribers a month, which made the distressed European banks he was studying uninteresting.

factualspeaker onlynovelty 2/4durability 2/4· Neil Mehta

“QQ was adding something like 30 million subscribers a month... I want to spend all my time on 10-centent”

0.26

Green Oaks values beauty in businesses and relationships, and this is why it invests in companies that make the world better and passes on attractive but trivial deals like buying a Minnesota software company at three times revenue to flip it.

normativespeaker onlynovelty 2/4durability 2/4· Neil Mehta

“we like beautiful businesses. We like we love beautiful relationships. We care about like beauty in the world”

0.26

Mehta's early investing at 27, when Green Oaks started, was driven by three things: the teams, the quality of the businesses they were building, and the returns math, which he argued was the beginning of a 20-to-30-year opportunity as mobile and cloud penetration curves were just starting.

forecastspeaker onlynovelty 2/4durability 2/4· Neil Mehta

“this is the beginning of a 20 or 30-year opportunity ahead of us”

0.24

The Benny partnership is central: Benny's fast zero-to-one thinking and his ability to extend the firm's time horizon balance Mehta's short-term tendencies, and the two talk for hours nightly about investments.

factualspeaker onlynovelty 1/4durability 2/4· Neil Mehta

“Benny is so good at stepping back and reminding everybody, including myself, especially myself... what we're trying to optimize for over the fullness of time”

0.22

Being down 95 percent versus 97.5 percent is only a half-percent difference in the drawdown math, so buying a falling stock at 50, 30, 20, and then 5 dollars feels bad but the loss is not meaningfully different from the earlier entries.

factualspeaker onlynovelty 2/4durability 1/4· Neil Mehta

“what's the difference between being down 95% and 97 and a half% just half”

0.22

Carvana's stock fell from roughly 100 dollars to 5 dollars, which Mehta notes is rare: almost no companies that fell from a 70 billion dollar market cap recovered without being a fraud.

factualspeaker onlynovelty 2/4durability 1/4· Neil Mehta

“how many companies can you name that went from $70 billion market cap to one and weren't like a fraud”

0.21

Green Oaks was founded by Mehta and Benny in 2011 after they left DE Shaw, and the firm's first pitch was built around specific businesses and penetration curves in cloud and mobile rather than a generic promise of future success.

factualspeaker onlynovelty 1/4durability 2/4· Neil Mehta

“I was very much in the latter camp. I was describing what we were seeing at companies”

0.21

Green Oaks's greatest relationships are with founders like Bom Kim rather than just companies, and the firm's mission is to keep searching for the next potential S&P 500 company.

factualspeaker onlynovelty 1/4durability 2/4· Neil Mehta

“fundamentally what you've done at Green Oaks is always search for the next person”

0.18

Early Green Oaks used to ask 'What are the most negative things you could say about us' and Mehta accepts most criticisms, such as being too hard on the team, while arguing that the firm's early success was not being judged on what it did ten years later.

factualspeaker onlynovelty 1/4durability 1/4· Neil Mehta

“Green Oaks wildly successful early but as of late what have they done”

0.18

Green Oaks invested a little under a billion dollars in Coupang across about 10 years, leading five to eight rounds, and kept buying shares after the IPO, which exemplifies its concentrated, long-term approach.

factualspeaker onlynovelty 1/4durability 1/4· Neil Mehta

“We invested a little under a billion in total capital across 10 years. Uh we invested um led 5 to 8 rounds.”