LLenny's Podcast
← All episodes
Eric Ries (creator of the Lean Startup methodology)29 October 2023

Reflections on a movement

7Frameworks
15Insights

Frameworks in this episode

Innovation4 steps

Leap-of-Faith MVP Scoping

Scope the smallest build that tests your riskiest assumption, not a stripped-down product

Read →
Mindset3 steps

Robust-Across-Futures Action

When the future is unknowable, choose actions that are ethical across a wide range of scenarios

Read →
Communication2 steps

The Best-Practice Inversion Interview

Ask a candidate for a best practice, then for when it would not apply

Read →
Leadership3 steps

The Chief Engineer Trade-off Owner

Give one person moral authority over a whole product to keep design coherence in one head

Read →
Strategy4 steps

The Fixed-Window Pivot Decision

Time-box a full-focus test on the one thing that matters, then let the team name the next bet

Read →
Marketing4 steps

The Sticky Engine of Growth

Model retention by the three, and only three, reasons a customer returns to your product

Read →
Leadership4 steps

Trust-by-Structure Governance

Embed your promises into the company's structure so they hold even when you're gone

Read →

Insights & moments

The myth-busts, hot takes, explainers, and tools worth keeping.

Myth Buster· 1

Myth Buster28:00

An MVP Isn't a Crappy, Bare-Bones Product

Eric corrects the most common misconception about the MVP: that it means a stripped-down, low-quality thing that will crash your computer. He argues the MVP is simply the most efficient way to test whether your hypothesis is true — quality is defined by the customer, and if you don't know who the customer is, you don't even know what 'quality' means.

  • MVP is not a specific tactic or a bare-bones build — it's whatever most efficiently validates the hypothesis
  • In genuinely high-stakes industries (jet engines, data centers, oil drilling) the bar is far higher than iOS apps, yet MVP logic still applies
  • If the quality bar is truly high, you can only afford one critical feature at that level — then test it
  • Quality is defined by the customer; without knowing the customer, 'quality' is meaningless
  • Building a high-quality product on the wrong value prop is nothing to be proud of

“people think that mvp is about a specific tactic so like an MVP is like a bare bone stripped down thing that will crash your…”

Eric Ries · 28:00

“quality is defined by the customer and if we don't know who our customer is we literally don't know what the word quality means”

Eric Ries · 32:00
#mvp#product#quality#validation

Hot Take· 3

Hot Take00:00

A Failed Startup Isn't the Worst Outcome — a Zombie Company Is

Eric pushes back on the idea that a startup failing is catastrophic. He argues the far worse fate is being trapped in a company that won't die but you hate, or building something successful that becomes a malign force you're complicit in. He ties this to an under-discussed mental health crisis among founders, including so-called successes.

  • A startup failing isn't even in the top 10 worst things that can happen to a founder
  • Worse is a zombie, undead company you hate but can't leave
  • Even lucrative exits can leave founders miserable or worse — he cites a founder who exited for ~$300M and later died by suicide
  • Building a company that becomes a malign force you feel complicit in is the real harm
  • Founders should take this risk seriously early, not late

“people act like having a startup fail is the worst thing that can happen to you and like man that's not even in the top…”

Eric Ries · 00:00

“far worse is to be in a company that won't die a zombie Undead company that you hate but you can't leave”

Eric Ries · 00:00
#founder-mental-health#startups#failure#values
Hot Take49:00

When 'Craft' Is Just an Excuse to Avoid Feedback

Eric respects genuine craft — he'd rather see one thing done extraordinarily well than twenty things half-done. But he distinguishes real craft (keeping options open, good engineering, actual taste) from craft used as a cover to avoid customer feedback. If your artistic ethos is non-negotiable, fine — just don't pretend to test it. The danger is people who fake taste they don't have.

  • He'd rather see one thing done extraordinarily well than twenty things half-assed
  • Real craft (right primitives, abstractions, high-quality people) makes you faster, not slower
  • Some use craft as a catch-all excuse to avoid getting feedback
  • If an aesthetic is a true core value, treat it as a non-negotiable — don't pretend to test it
  • Caring deeply about craft raises the need for experimentation on everything else
  • Many people faking taste have no real design ethos — experimentation can build it

“I would much rather people do one thing extraordinarily well than like 20 things half ass”

Eric Ries · 49:00

“another part of craft is just an excuse for just avoiding feedback like making things pretty”

Eric Ries · 51:30
#craft#design#feedback#product
Hot Take1:37:30

We've Got the Definition of 'For-Profit' Backwards

Eric argues that being 'for-profit' should mean maximizing human flourishing — creating net-new value rather than extracting it. By current categories, Philip Morris is for-profit and the Smithsonian is not-for-profit, which he calls backwards, since the Smithsonian does far more good than 'the merchants of death.' There are exploitative ways to make money and truly additive ways, and we've conflated them.

  • Real for-profit activity should mean maximizing human flourishing and net-new value
  • Current labels call Philip Morris for-profit and the Smithsonian not-for-profit — backwards
  • There's exploitative, extractive money-making and truly additive money-making
  • He calls ESG off the rails — Philip Morris has great ESG scores, which is absurd
  • Trustworthy promises must be encoded into a company's structure, not just its rhetoric

“making a profit is actually about maximizing human flourishing that's what it means to be for-profit”

Eric Ries · 1:37:30

“our current category says that Philip Morris is a for-profit company and the Smithsonian is a not for-profit but that's just backwards”

Eric Ries · 1:37:30
#governance#capitalism#values#esg

Explainer· 3

Explainer1:20:30

Why AI Agents Endlessly Pass the Buck

Eric describes AI as fundamentally a management technology — it manages intelligence. Watching AI agents work, he found they behave like bad middle managers: given a task, they spawn a new agent to do it, which spawns another, passing the buck into infinite regress rather than actually doing the work. Anyone who's managed people recognizes the pattern.

  • AI is at its core a management technology — it manages other intelligences
  • AI agents are programmed to never say no, so they accept any task
  • Instead of doing a task, an agent often spawns a sub-agent to do it — which spawns another
  • The result is passing the buck or infinite regress, e.g. researching 10 things that each need 10 more
  • He couldn't get an AI to fill out the other 28 days of a 30-day course — it kept delegating
  • Any former middle manager reading the transcripts would recognize the behavior

“at the end of the day AI is a management technology it like the thing it does is manage intelligence and other intelligences”

Eric Ries · 1:22:30

“everyone will just SP pass the buck to somebody else or it'll be like it'll go into an infinite regress”

Eric Ries · 1:21:30
#ai#agents#management#product-development
Explainer1:50:30

The Open Secret: Foundation-Controlled Companies Outperform

Eric explains that alternative governance structures — foundation-controlled and family-controlled companies — are an open secret that quietly outperform. He points to a Danish study where foundation-controlled companies beat level-matched public companies financially, plus Toyota, Hershey, Warren Buffett's companies, and OpenAI's nonprofit-with-for-profit-subsidiary structure. In some European countries, 20-25% of public companies are structured this way.

  • A study (he believes Danish) found foundation-controlled companies outperformed matched public companies financially
  • Toyota's 'terrible' keiretsu governance actually enables its long-term thinking
  • Hershey is controlled by a foundation that runs a school; family-run companies show similar outcomes
  • OpenAI is a nonprofit foundation with a wholly owned for-profit subsidiary — and Anthropic has a long-term benefit trust
  • In some European countries 20-25% of public companies use this structure — it's normal there
  • The shareholder-primacy theory is wrong even if all you care about is shareholder returns

“open AI is a nonprofit foundation with a wholly owned for-profit subsidiary”

Eric Ries · 1:52:00

“the foundation controlled companies outperformed from a financial performance perspective”

Eric Ries · 1:54:00
#governance#corporate-structure#long-term-thinking#openai
Explainer2:03:30

Toyota's Chief Engineer: Moral Authority Over the Whole Car

Eric describes a Toyota role he wishes more founders knew about: the chief engineer, who has moral authority over an entire product line from start to finish but no direct reports. The most important product work is making trade-off decisions, and if everyone makes their own, you get incoherent design. The chief engineer is the single person who can hold the whole product in their head and make those calls.

  • Toyota's chief engineer has moral authority over an entire car line, start to finish
  • They have no direct reports and only a small staff — just a 'big room'
  • Inside Toyota the car isn't called a Camry; it's called 'so-and-so's car'
  • The most important product job is making trade-off decisions
  • Letting everyone make their own trade-offs destroys coherent design
  • You need one person who can keep the essential trade-offs in their head

“the chief engineer is the person with the moral authority over an entire line like a car from start to finish”

Eric Ries · 2:03:30

“the most important thing you have to do in a product is make trade-off decisions”

Eric Ries · 2:04:00
#toyota#product-leadership#trade-offs#organization

Story· 4

Story35:30

The Ugly IMVU Teleport Hack Customers Loved

IMVU couldn't afford inverse kinematics, so their 3D avatars were stuck sitting in place and customers complained it felt claustrophobic. As an embarrassing hack, Eric made avatars teleport between spots instead of walking. He was ashamed to ship it, but tracking surveys suddenly showed customers rating IMVU the most advanced avatar product they'd used — because teleporting beat waiting for a Sims character to walk.

  • They lacked inverse kinematics, so avatars couldn't stand and walk without looking horrible (uncanny valley)
  • Customers said the product felt claustrophobic compared to The Sims
  • As a hack, Eric made the avatar disappear and reappear in a new spot — teleporting — which he thought was below the quality bar
  • After shipping, tracking surveys rated IMVU the most advanced avatar product customers had ever used
  • A customer explained: teleporting was better than waiting for a Sims character to get up and walk

“in the tracking surveys people started to say that imbu was the most advanced Avatar product they'd ever used”

Eric Ries · 36:30

“in mvw when you want to go somewhere you teleport The Sims you got to wait for the guy to get up and walk around…”

Eric Ries · 37:00
#imvu#mvp#product#3d-avatars
Story39:30

The $18M Efficiency Product Customers Laughed Out of the Room

A team of PhDs was going to spend three years and $18M and possibly win a Nobel Prize in materials science to make a data-center product more efficient. Eric convinced them to take a brochure to customers first. The customers laughed them out of the room — they don't pay for efficiency, because operating costs are somebody else's problem. The whole premise was wrong.

  • The team planned three years and $18M of research to win on efficiency
  • Eric had them test a brochure with customers before building anything
  • Customers laughed them out of the room; efficiency was irrelevant to what they cared about
  • Data-center buyers don't pay for efficiency because operating costs fall on someone else
  • These insights seem obvious in retrospect, which is why everyone assumes it won't happen to them

“customers like laughed them out of the room and it was like nothing to do with anything they cared about”

Eric Ries · 40:00

“we don't pay for efficiency we build data centers how much they cost to operate to somebody else's problem”

Eric Ries · 40:30
#mvp#customer-development#validation#enterprise
Story44:30

Two Years of LTSE Work Wiped Out — and Why That Helped

Building the Long-Term Stock Exchange, Eric spent roughly two years on a partnership agreement, got SEC staff approval, then got ambushed when the SEC withdrew approval and the partnership collapsed. Every clause he'd fought over turned out to be a waste — but the team that lived through that failure became dramatically faster on the next attempt that ultimately got approved.

  • LTSE took ~10 years; one early iteration involved a two-year partnership + SEC approval process
  • After SEC staff approval, the SEC withdrew it and the whole thing died
  • Every hard-negotiated clause was moot because the partnership was never consummated
  • He calls it the worst thing that happened to him and, in retrospect, the best — it enabled the pivot that worked
  • Teammates who survived that failure were far faster on the next thing

“every bit of work we did that went into that partnership agreement for almost two years turned out to be a waste of time”

Eric Ries · 46:30

“the people who went through that experience with me were so much faster on the next thing that we did”

Eric Ries · 47:00
#ltse#pivot#failure#resilience
Story1:04:00

The Whiteboard That Proved His Own Memory Wrong

Years after an early planning meeting, Eric found the original whiteboard shoved in a storage closet, still in his own handwriting. But many ideas he'd written were bad ones he was sure he'd never believed — and things he credited to himself were actually his co-founder's. He had no memory of ever changing his mind, and had to admit the physical evidence beat his own recollection. Founders discover their vision through building, and human memory is deeply buggy.

  • An old whiteboard from a five-person planning meeting resurfaced years later in a closet
  • It was recognizably his handwriting, but full of ideas he was certain he'd never held
  • He credited good ideas to himself that were actually his co-founder's
  • He has no psychological memory of ever changing his mind — the original memory persists
  • The physical evidence had to win; a founder's vision is discovered, not fixed from day one
  • He ties this to why science makes you write hypotheses down — human psychology is buggy

“it's recognizably in my own handwriting but many of the things that I'm writing there are horrible ideas”

Eric Ries · 1:05:00

“I even to this day have no psychological memory of having changed my mind about it”

Eric Ries · 1:05:30
#vision#memory#self-discovery#founders

Q&A· 1

Q&A1:13:30

How to Know It's Time to Pivot

Asked how founders should decide whether to pivot, Eric says if you're even asking the question, you probably already know the answer — because if you had product-market fit, you wouldn't have time to wonder. When you admit it's not working, give yourself a fixed period to go 100% heads-down on the one thing that matters and see if you can move the needle.

  • If you're asking whether to pivot, you likely already know the answer
  • Product-market fit is an all-consuming whirlwind — there's no time for navel-gazing when it's working
  • Give yourself a fixed window (six weeks, or even one weekend) to try to make it work
  • Have one full-time person heads-down on the single most important thing, not a part-time committee
  • Stuck startups get trapped in diminishing returns — find a paradigm where experiments are productive
  • Go around the room: 'if I could start over, the thing I wish we were doing was...'

“if you're asking whether you should pivot or not you probably know the answer already”

Eric Ries · 1:13:30

“if you had product Market fit you would not have time for this”

Eric Ries · 1:14:00
#pivot#product-market-fit#decision-making#founders

Takeaway· 3

Takeaway14:30

What Winning a Movement Actually Looks Like

Eric expected the Lean Startup movement to be a pitched battle against the old guard, but no opposing army ever showed up — it won by default and became the default vocabulary. He realized the real victory isn't a famous company adopting it, but the quiet stories of readers who found the book helpful, which stripped the hype and drama out of the whole thing.

  • He imagined leading hordes into an ideological battle; instead the old way had almost no defenders
  • The movement went from 'insurgent revival' to the obvious default that even critics must use to criticize it
  • Real victory is the private stories of the book being helpful, not high-profile endorsements
  • Treating it as a scientific theory means whatever works is Lean Startup — new practices aren't heresy

“we charge onto the field and no other army ever came to charge us back like it just we won by default”

Eric Ries · 15:00

“the victory actually is in the stories that people tell me about how the book was helpful to them”

Eric Ries · 17:00
#lean-startup#movements#ego#legacy
Takeaway47:30

If You Can't Fail, You Can't Learn

Eric frames the essence of MVP as the number one lesson of the scientific method: if you can't fail, you can't learn. Founders have a psychological defense mechanism that lets them deny failure ever happened, and that denial is exactly what blocks learning. The point of testing is to find the catastrophic problem early enough to still do something about it.

  • The scientific method's core lesson: no failure means no learning
  • Founders' psychological defense is to insist they never failed or pivoted
  • If you set things up so you can't admit something didn't work, you've disabled learning
  • MVP exists to surface the company-killing problem while there's still time to act

“the number one lesson of the scientific method is if you can't fail you can't learn”

Eric Ries · 47:30
#mvp#scientific-method#failure#learning
Takeaway53:30

Being Willing to Get Fired Is a Career Accelerator

Eric advises building the company and product you actually want to work on, and being willing to get fired for doing it your way. Being known as the person who stood for a principle so hard they got fired over it magnetizes the right people to you. He was ready to be fired for his unconventional product beliefs before Lean Startup even had a name.

  • Build the company you'd be excited to work on — if it fails, at least you tried on your terms
  • Being willing to get fired for doing it right is a fast career accelerator
  • Standing for a principle attracts people who want the same thing and makes you happier
  • He went into a board meeting expecting to be fired over his conviction and was at peace with it

“just get fired for doing it the way you think is right”

Eric Ries · 54:00

“there's nothing there a faster career accelerator than being known in the industry as the person who stands for that principle so much that you…”

Eric Ries · 54:00
#career#conviction#founders#principles