· Micky Malka

Micky Malka, Founder of Ribbit Capital

Durable advantage comes from playing an infinite game — no winning or losing, only ahead or behind — where written conviction, compounding relationships, and reputation as the only asset let you re-run the same decision pattern across every macro environment.

infinite-gamefintechventure-capitalreputationcompoundingaicryptotastefounder-dna0% confidence

Why this is in the corpus

Malka is a five-time founder turned Ribbit Capital founder who operationalizes infinite-game doctrine: year-long written theses compressed to napkins, founder-DNA reading built on decade-long trust, the Dee Hock chaordic apprenticeship, and the Lemon Bank to Walmart One Pay 12-year compounding arc. Dense, non-obvious operator doctrine on conviction, reputation, taste, and the coming agent-money infrastructure wave.

Summary for skimmers

Micky Malka (Ribbit Capital) on refusing labels, writing year-long theses that fit on a napkin, Buffett's pattern-consistency since age 13, the infinite game (only ahead or behind — and he'd rather be behind), Dee Hock's chaordic organizations, reading founder DNA via the sleepless-night call, the ikigai-of-the-tiger checklist, Lemon Bank reincarnated as Walmart One Pay, token factories, agent money on 24/7 rails, and why Silicon Valley must recover beauty and taste.

Briefing

What survives the editorial filter

This page should feel like a smart colleague already listened for you and left only the operating logic worth keeping. Not everything said in the episode makes it through.

Trust signal

Direct episode extraction

Best used for

Decision-grade retrieval metadata not yet added for this episode.

Hold lightly

No explicit downgrade reason stored yet for this episode.

Principles

Durable claims that survive beyond the speaker's biography — each with explicit limits, transferability judgment, and evidence.

Principle

The infinite game: you are never winning or losing, only ahead or behind

Replace win/lose framing with ahead/behind framing and the game — and your motivation — never ends.

Before a decision, ask 'does this put me ahead or behind in the infinite game' rather than 'will I win'.

Principle

The best entrepreneurs never have to sell their company

An exit — even a successful one — is a design failure if your goal is compounding forever.

Design the structure, culture, and capital base so you never have to sell — permanence is the point.

Principle

Earn the sleepless-night phone call — then never answer the question

The measure of an investor relationship is whether the founder calls you at the sleepless-night moment — and your job then is to help them answer, never to answer.

Track whether your portfolio founders call you first at crisis moments; if not, the relationship is transactional.

Principle

Written conviction is what lets you be wrong for a long time

Conviction built through writing is the only thing that lets you stay wrong-looking long enough to be right.

If you cannot put the thesis in writing, you do not yet have conviction — you have a mood.

Principle

Compounding is the master equation — and it applies to relationships

Relationships, ethics, work, and knowledge compound exactly like capital — consistency is the interest rate.

Invest in relationships and reputation daily the way you would invest capital — the payoff arrives decades later at scale.

Principle

Study the pattern, not the decisions (the Buffett letters lesson)

When studying great operators, extract the invariant decision pattern, not the individual decisions.

When you study a great operator, write down the process they repeat across environments, not the moves they made.

Principle

Refuse labels — labels stop you from seeing

Accepting labels — on yourself or on opportunities — blinds you to what is actually there.

Audit where you have accepted a label for yourself or your market, and ask what that label is hiding.

Principle

Entrepreneur at heart, investor by design

Define yourself by disposition and learnable craft, not by role, so identity never constrains what you can do next.

Write your own identity line as disposition + craft instead of a job title.

Principle

Reputation is the only asset

Reputation is the one non-recoverable asset — it closes or forecloses opportunities invisibly, so it outranks capital.

Price every decision by its reputation impact first; treat money lost as recoverable and reputation lost as permanent.

Principle

When you get ahead, change the rules

Deliberately reset the game you are winning so you can innovate from behind instead of defending from ahead.

If you have been comfortably ahead for a while, that is the signal to change the game, not to optimize it.

Principle

Prefer being behind — you are a better operator when losing

Behind is the higher-performance state: you learn more, work harder, and see more clearly than when ahead.

Treat being behind as an asset; treat being ahead as the warning sign.

Principle

Founder DNA must be visible in the offices, not just the founder

Diligence the culture for the founder's DNA — the bet is only durable if the DNA propagates beyond the founder.

Walk the offices: if the founder's DNA is not detectable in the team without the founder present, the advantage is not durable.

Principle

The rebel stays a rebel by never trading questions for preaching

The breaking point of every operator is the day they stop asking questions and start preaching — defer that day to your last.

Monitor your own ask-to-preach ratio; when it inverts, your rebel time is ending.

Frameworks

Reusable systems and operating models — including when they help and when they break.

Framework

The napkin test: a year-long thesis that compresses to a napkin

A thesis is only done when a year of written work compresses onto a napkin.

Run the pipeline: write long, test with people who know, then compress to a napkin — if you cannot, keep studying.

Framework

Ikigai of the tiger: the five-trait founder checklist

Back founders only when all five are present: energy of a scientist, conviction of a missionary, heart of a partner, dreams of an athlete, obsession of an owner.

Score prospective partners against all five traits during time spent together — absence of any one is disqualifying.

Framework

Every service reduces to identity, value, and intelligence

Decompose any AI-era company into the three primitives it must master: identity, value, intelligence.

Map your product onto the identity/value/intelligence triad and identify which primitive is your moat.

Framework

Chaordic organization: operate at the intersection of chaos and order

Design organizations at the edge of chaos and order — too much of either kills innovation or communication.

When designing an org, ask which side you are over-weighted on — order that blocks innovation or chaos that blocks communication — and move toward the edge.

Signals

What appears to be shifting, for whom it matters, and what happens if you ignore it.

Signal

Every company is becoming a token factory

The industrial-revolution analogy for the AI era: companies are converging on a factory model whose raw material is machine-readable tokens.

Position your company as a token factory: identify your token supply and what new thing you manufacture from it.

Signal

Agents need 24/7 money rails — prediction markets are the early tell

Read prediction markets as rails-building for agent money, not gambling — agents require 24/7 settlement infrastructure.

Track 24/7 settlement infrastructure as the leading indicator of the agent-money wave.

Signal

The youngest generation finds software alone boring — they want physical

The generational preference is flipping from capital-light software to physical building — expect atoms-plus-bits companies from the best young founders.

If you are betting on young founders, expect physical-product ambitions and mixed-age team structures, not pure software.

Opportunities

Only included where there is a buyer, a real wedge, and a plausible revenue path — not vague idea theater.

Opportunity

Money is weirdly behind knowledge in the AI era

The gap between AI's penetration of knowledge and its absence from money is the largest open opportunity in fintech.

Build or invest where AI meets money movement — the catch-up trade from knowledge to money is still early.

Opportunity

Silicon Valley lost beauty — founders who bring taste back will win

Taste and beauty are the abandoned axis of competition in technology — the next winners will compete on how products make people feel.

Ask of your product what Malka asks: what is authentic, what is beautifully designed, and what does it make you feel — then compete there.

Lessons still worth keeping

Useful takeaways that did not fully clear the bar for durable principle status.

Lesson

Five exits taught him the vehicle was the failure, not the companies

If you keep having to exit, the defect is in vehicle design — learn to design the structure that lets you play forever.

Before starting the next thing, design the structure for forever — otherwise you are pre-committing to another forced exit.

Lesson

Macro matters — a lesson you only truly learn with all your money on the table

No amount of execution beats hostile macro — and the lesson only fully lands when your own capital is fully exposed to it.

Size your exposure so a macro stall you cannot control does not kill you before your execution can matter.

Lesson

The grandfather loan: buy the asset, pay the interest, learn the system

A small, real position with real cost of capital teaches more than any amount of passive reading.

To learn a domain, take a small real position and let the ownership force the study.

Lesson

You only need to get rich once — time is the asset (Munger at Happy Hollow)

Time remaining is the true balance sheet — protect the ability to compound for decades and never need to get rich twice.

Optimize for staying in the game for decades; getting rich once, kept, beats getting rich twice.

The Plays

Try these this week

Verb-first executable actions — each one tied to a stated outcome in the episode.

The six-words interview question inherited from Dee Hock

Outcome: Ask candidates to rank six words and judge the walk-through, not the ranking.

Context: Malka calls this his go-to interview question and attributes it directly to Dee Hock, founder of Visa, with whom he spent ten to twelve years. The diagnostic value sits in the narration: the order itself is not scored, the way the candidate walks through the order is. This is the operating counterpart to Hock's chaordic theory, which Malka also inherited.

“I give people six words and I tell 'em to organize him in a certain order. And, and then the way they walked through that order says a lot about, about the person.”
— Micky Malka

5-10 minutes inside a normal interview; the calibration base compounds over years of reuse. per

Stop or pivot when

→The ordering itself is not the signal; the reasoning path is.
→Reject the question if the candidate can produce a rehearsed answer - that means the six words are too conventional.

Scripts

Before you start

· Clarity on your own organisational values, or you cannot choose the six words.
· Discipline to never change the words, or the comparison base is destroyed.
· An interviewer trained to listen to reasoning rather than wait for an answer.

Run the ikigai-of-the-tiger screen in person before you commit

Outcome: Screen partners against five behavioural traits observed in person, not against a deck.

Context: Malka names the five as the energy of a scientist, the conviction of a missionary, the heart of a partner, the dreams of an athlete, and the obsession of an owner. He is explicit that the list had to fit on a napkin before it was trusted, and that fitting on a napkin is itself the proof of sufficient study. The screen governs who Ribbit will build a long-term relationship with, not just who it will fund.

“When we meet them and we spend time with them, we wanna make sure those things are present.”
— Micky Malka

~10 years to develop the list; it was compressed to a napkin about 5 years ago. Per founder, months of relationship time. per

Stop or pivot when

→All five traits should be present for a long-term partnership.
→The list must fit on a napkin or it has not been thought about long enough.

Scripts

Before you start

· Enough deal history that patterns are real rather than imagined.
· A partnership model that allows months of pre-investment relationship time.
· Willingness to pass on a good business when the traits are absent.

Verify founder DNA by touring the offices, not by re-interviewing the founder

Outcome: Diligence the offices in the far-flung countries to find out whether the founder's DNA actually replicated.

Context: Malka describes visiting Revolut offices in the Middle East, Barcelona and London within an eighteen-month window and reporting that the DNA is still breathable in each. He frames this as more important than understanding the founder alone: he wants to see whether the DNA reflects in the offices, in the team, in the people, and in the way they talk and move. He also notes every org is different, so the read has to be founder-specific.

“And you still go to the railroad offices, which is like in now in, I don't know, 15, 20 countries. And I've been to the, the one in the Middle East in Barcelona in London in the last year and a half.”
— Micky Malka

Rolling - three sites within an 18-month window in the Revolut example. per

Stop or pivot when

→DNA must be perceptible in the team and the way they talk, not just in the founder.
→Every org is different, so calibrate against that founder's DNA, not a generic culture rubric.

Scripts

Before you start

· An existing read on the founder, or there is nothing to compare the offices against.
· Access and travel budget to reach non-headquarters sites.
· A company large enough to have distant offices - not applicable pre-scale.

Walk the stores, use every product, then propose the JV the incumbent cannot prioritise

Outcome: Field-test the incumbent's product until you have a defect list, then pitch a co-owned JV framed around their priority stack.

Context: The sequence ran from a 2019 dinner arranged by Sarah Friar with then-CEO Doug McMillon, to a cold follow-up in January before COVID, to store visits where the team bought remittances and prepaid cards and found bugs and 1990s-era fintech, to a meeting with John Furner. Malka's pitch was explicit about the priority stack. The result was One Pay, a jointly owned company with a new brand, two acquisitions for infrastructure, and every Walmart financial product routed through it.

“we went as a team to couple of Walmart stores. We tried every financial product we can buy, we could, we did Remittances prepaid car, and we found bugs and we found all these things that looked like 1990s FinTech.”
— Micky Malka

Dinner in 2019, follow-up letter January 2020, JV built through the following years into One Pay. per

Stop or pivot when

→Only viable when the category is genuinely not the incumbent's priority one, two or three.
→Requires the incumbent to be founder-cultured or long-tenured leadership - Furner and McMillon had ~30 years each.
→You must have already run the equivalent business somewhere else, or you have no credibility.

Scripts

Before you start

· Prior operating proof in the same category - Lemon Bank served 15 million previously unbanked Brazilians.
· Capital to co-own and to buy infrastructure companies.
· A CEO relationship built on reputation, not on a deal process.

Build the relationship until you get the sleepless-night call, then refuse to answer the question

Outcome: Earn the 3am call by not talking about business, then help them answer rather than answering.

Context: Malka says the phone call at the sleepless moment means the firm did a hundred things right to deserve the right to be there. His explicit rule once he has it is never answer the question, just help him or her answer the question. The prerequisite is conversation about life, family, society and politics, because that is the only route to the question that actually matters.

“You gotta spend time with them and you gotta ask them all the questions and you gotta build a level of trust where honestly you don't have to talk about business that much.”
— Micky Malka

Years before the first sleepless-night call; decades for the full relationship. per

Stop or pivot when

→The relationship is working when the founder calls at the moment they cannot sleep.
→Answering the question directly is the failure mode, not the success mode.
→If you can only talk about business, the trust is not there yet.

Scripts

Before you start

· A reputation that survives reference checks - the founder checks you before they open up.
· A time horizon that tolerates years of unmonetised relationship investment.
· Genuine interest in the person; the play collapses if the non-business conversation is instrumental.

Write the thesis for a year, then publish it to founders and LPs as an accountability contract

Outcome: A written and published thesis is what lets you be wrong for a long time without abandoning the position.

Context: Malka describes essays that begin at roughly 20 pages and take four to five months, or up to a year and a half for the full cycle of think, articulate, test, share, come back, work it. The finished essay is shared with portfolio founders, close collaborators, and LPs explicitly so they can hold the firm accountable or walk away. The firm then goes back to the founders it met while writing, most of whom it has not invested in, and the essay becomes the relationship opener.

“We share it with our founders, people that we work with closely and we allow them to and, and our LPs, our investors. Why? Because we want 'em to hold us accountable.”
— Micky Malka

4-5 months minimum per essay; roughly a year to a year and a half for the full think-articulate-test-share-return-rework cycle. per

Stop or pivot when

→Napkin test: the thesis must be explainable on a napkin before publication.
→Time horizon: the thesis must describe the world in 10-15 years, not today.
→LPs are given a genuine exit - if they do not believe the thesis they can walk.

Scripts

Before you start

· An investor base or audience willing to be told the thesis rather than sold it.
· Willingness to be publicly wrong for a long time.
· A team that can hold a research thread across a year without shipping pressure.

Decision Moments

Actual decisions, real outcomes

Specific decisions narrated in the episode with their outcomes and transferable lessons.

In 2003, at 25 or 26, Malka and his partner had just sold their first .com company. Brazil had roughly half its population unbanked and no branchless banking infrastructure existed; the smartphone did not yet exist either. They had no outside investors and no institutional backing.

Did: Put every dollar of the .com proceeds back on the table to found Lemon Bank in Brazil, with no investors - just Malka and his business partner. Moved with his wife to Brazil and lived there three years. Built three physical distribution models in parallel: points inside existing stores at checkout, kiosks inside stores, and franchised micro-stores running only their payment system. Scaled to 7,000 physical locations serving 15 million customers.Outcome: Nearly lost everything - the economy would not take off no matter what they did, and they almost killed the business twice. Survived, became the first to bank 15 million previously unbanked Brazilians, and sold to the largest bank in Brazil in 2008. The insight resurfaced twelve years later as the basis for the Walmart JV.

Macro matters, and it is a different thing to learn it with all your money on the table. Malka had already seen hyperinflation and devaluation growing up in Venezuela, but says learning it with everything at risk is another matter entirely. The compensating lesson is that an insight earned this expensively persists and can be redeployed a decade later at far greater scale.

Part of an emerging decision pattern across multiple episodes

January 2020, before COVID. Malka had met Walmart CEO Doug McMillon at a small San Francisco dinner in 2019 organised by Sarah Friar, and had been given a vague open invitation to visit Bentonville. Walmart's financial products ran through a mix of vendors, partners and its own card, and Malka had a twelve-year-old thesis from Lemon Bank about serving exactly this customer.

Did: Took the open invitation literally and wrote to McMillon asking to be mentored. Before the meeting, sent the Ribbit team into Walmart stores to buy every financial product available - remittances, prepaid cards - finding bugs and infrastructure that looked like 1990s fintech. Then pitched a co-owned joint venture, framed explicitly as: this is not your priority number one, two or three, and it is ours.Outcome: Walmart created a jointly owned company with Ribbit - something it had never done. They built a new brand and team, bought two companies for infrastructure, and launched One Pay. Every Walmart financial product now routes through it: phone payments, debit, Klarna loans, installment loans, credit cards, to millions of American consumers.

The credible partnership pitch to an incumbent concedes their attention constraint rather than attacking their competence. The field-bought defect list is what makes the concession land as diagnosis rather than flattery - and the co-owned JV structure, rather than a vendor contract, is what makes the alignment durable.

Part of an emerging decision pattern across multiple episodes

Around 2012, after founding five companies starting at age 17 and exiting all of them - including Lemon Bank, sold in 2008. The exits were financially successful and had changed many lives, but Malka had concluded he was a failed entrepreneur because he had never found a vehicle he could run forever.

Did: Founded Ribbit as company number six, explicitly designed never to be sold. Refused to call it an investment firm - built it as a startup with its own technology stack, a small team, Star Wars vocabulary (meetings called tattooing, backing Jedi), radical internal transparency with shared calendars and inboxes, and group rather than individual decision-making. Spent 20 years first learning how to design the structure, culture, team, knowledge, aperture and brand for something permanent.Outcome: Fourteen years in and still running, with 10-15 year founder relationships (Revolut from Series A to bank licence, Robinhood from pre-product), the Walmart JV, and the Node digital art studio. Malka insists it is still super early and they have a long way to go.

The best entrepreneurs never have to sell their company - so if you have exited repeatedly, the failure may be the vehicle rather than the businesses. The correct response is not another company but a structure designed from the outset to be held forever, which takes as long to design as the previous companies took to build.

Part of an emerging decision pattern across multiple episodes

Tensions surfaced

Contradictions and trade-offs the episode raises — judgment calls a thoughtful operator has to navigate.

Tension

Authenticity wins — yet fake-it founders build massive companies

Authenticity is a filter for who deserves your decades, not a law of who builds big companies.

Use authenticity to choose long-term partners, but do not mistake it for a predictor of financial outcomes.

Corpus connection

Where this episode fits for retrieval

What kinds of decisions this briefing is best pulled into.

Primary decisions

• strategic-bet
• invest
• partner