#41 The Hard Thing About Hard Things: Building a Business When There Are No Easy Answers
What this covers
This episode has David Senra working through Ben Horowitz's book on running a company, with other founders' stories brought in alongside it. Horowitz's own history anchors the discussion. His company raised money at a $700 million valuation in its first nine months, then a few months later could not raise privately, which left an IPO as the only financing route. He also bought Tangram for about $10 million because a $20 million annual contract depended on keeping its customer happy.
The middle of the episode is about management and culture. It covers Andy Grove's arithmetic on why training is high-leverage, and how big-company executives live in interrupt-driven work while founders have no inertia to rely on. It also sets Marc Andreessen's view that titles are free and should be generous against Mark Zuckerberg's choice to relevel hires to lower titles. Jeff Bezos's door desks illustrate shock as a way to change behavior. Andreessen Horowitz's borrowing of Michael Ovitz's firm-wide network at Creative Artists Agency is a further example.
The last stretch concerns judgment under pressure. Founders explain away bad leading indicators, which is how Cisco, Siebel, and HP missed quarters in 2001. The episode also covers asking what you would do if you went bankrupt rather than what the worst outcome is, and believing there is an answer without computing the odds. The most contested claim is that a company early in a large market should stay standalone, as Google did when it declined offers above $1 billion. Other points include the idea that no product is liked by more people than dislike it, and that free capital produces less creative solutions.
Horowitz's book argues that hard business problems have no recipe, so founders must learn from distilled experience, staying persistent and reframing questions rather than following formulas. The host frames this as a case for learning from founders' stories.
- Founders learn by doing, since no training prepares a person to run a company except running one
- Determination matters more than intelligence, and asking different questions can unlock seemingly impossible paths
- Founders systematically misread leading indicators and need to focus on what they can actually control
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.
Andy Grove calculated that 12 hours of preparing four lectures for 10 students yields about 200 hours of work gained through a 1 percent performance improvement across their roughly 20,000 combined hours, so training is among the highest-leverage activities a manager can perform.
“Training is quite simply one of the highest leverage activities a manager can perform.”
Michael Ovitz founded Creative Artists Agency by building an integrated network giving every agent access to a firm-wide grid of opportunities, making the firm far more powerful than any single agent, and within 15 years it represented 90 percent of Hollywood's top talent.
“within 15 years, CAA represented 90% of the top talent in Hollywood and had rewritten the rules”
If a company is very early in a large market and has a good chance of being number one, it should remain standalone because no buyer can pay for the forward credit of that future value, as Google declined over-$1 billion offers early on.
“you were very early on in a large market; and B, you have a good chance of being #1 in that market, then you should remain stand-alone.”
Horowitz's company raised money at a $700 million valuation in its first nine months, but a few months later could not raise any money privately, so going public through an IPO became the only financing route.
“In the first 9 months, we were able to raise at a $700 million valuation. And just a few short months later, we couldn't raise any money at all.”
Founders act urgently on positive leading indicators but explain away negative ones with alternative explanations, so companies overestimate good news and miss downturns; this is why big tech firms like Cisco, Siebel, and HP missed quarters during the 2001 crash.
“our hypothetical founder, like almost every other founder, only took action on the positive indicator and only looked for alternative explanations on the negative leading indicator.”
When a situation seems hopeless, asking what the worst outcome is produces only dread, but asking what you would do if you went bankrupt reveals a viable path such as buying the software and starting a new company.
“Then one day, I ask myself a different question. What would I do if we went bankrupt? The answer that came up would surprise me.”
Founders who are unclear how to build their product must enter the broader market and ship and sell an imperfect product, because only by selling the wrong product fast can they learn what the right product is.
“Paradoxically, the only way to do that was to ship and try to sell the wrong product. We would fall on our faces, but we would learn fast”
When building a company you must believe there is an answer and not calculate your odds of finding it, because the task is the same whether the chance is 9 in 10 or 1 in 1,000; the future is different, so you must survive long enough to see it.
“Founders should not play the odds. When you're building a company, you must believe there is an answer, and you cannot pay attention to your odds of finding it.”
Marc Andreessen argues titles cost nothing compared to salary, bonus, stock, or span of control, so companies should give the highest titles possible, while Mark Zuckerberg deliberately sets titles below industry standard to keep leveling fair.
“Of those, title is by far the cheapest. So it makes sense to give the highest titles possible.”
Shock, such as a deliberately unusual or uncomfortable company practice, is a great mechanism for behavioral change because it quickly shows newly arrived employees how different the organization is.
“Shock is a great mechanism for behavioral change.”
Zuckerberg's Facebook releveled every new employee to lower titles so new hires do not outrank better-performing existing employees, and it forces managers to internalize the leveling system and keeps product people and engineers at the cultural core.
“First, he guarantees that every new employee gets releveled as they enter his company.”
Andreessen Horowitz copied Michael Ovitz's Creative Artists Agency model of an integrated firm-wide network, because a firm-wide grid of opportunities makes the firm far more powerful than any individual agent.
“Ovitz's breakthrough idea was to build an integrated network that would allow any of the firm's agents to connect their clients to a firm-wide grid of new opportunities.”
No training as a manager, general manager, or any other job prepares a person to run a company; the only thing that prepares you is running a company, so founders must face many unfamiliar tasks while being expected to know how to do them.
“everybody learns to be a founder by being a founder. No training as a manager, general manager or in any other job actually prepares you to run a company.”
Jeff Bezos built frugality into Amazon's culture by making all desks from cheap Home Depot doors, a visible shock that signals the company's focus on saving money to offer lower prices, rather than auditing every expense.
“All desks at Amazon would be built by buying cheap doors from Home Depot and nailing legs to them.”
No product has ever been invented that more people like than dislike; even Facebook's two billion users and Apple's devices leave most people not buying, so founders should focus on the people who like their product and find more of them.
“there's never been a product ever invented that you're going to have more people that like it than don't like it.”
Founders should learn from other founders' experiences by reading books and listening to interviews, because learning from people who have done the work is the fastest way to avoid relearning everything from scratch.
“if they didn't write this down, we would have to learn everything scratch.”
Early Internet entrepreneurs did not talk to each other because no entrepreneurial community existed, but a real community now exists, and entrepreneurs learn more by talking to others who have done what they are doing.
“No one else can actually understand what you're going through unless somebody else is actually doing what you're doing.”
Paul Graham's Y Combinator found that determination predicts startup success more than intelligence: a brilliant but non-determined person becomes ineffectual, while a determined person with less intelligence can still succeed.
“for our own purposes, just remember the formula determination is greater than intelligence.”
Asking an entrepreneur what they would do if capital were free is dangerous, because abundant money leads to less creative solutions than the constraints of having little money.
“what would you do if capital were free is a dangerous question to ask an entrepreneur. The thinking this question leads to can be extremely dangerous.”
Big-company executives are interrupt-driven because most of their work is incoming demands, so they optimize existing businesses, while founders have no inertia and the company stays at rest without their direct input.
“In the early days of a company, there is no inertia that's putting the company in motion without massive input from you, the company will stay at rest.”
Almost no one, including major tech leaders like Oracle and Microsoft, thought the Internet would be significant beyond the scientific community in its early days, showing that transformative things can hide in plain sight.
“Almost nobody thought the Internet would be significant beyond the scientific community”
Mental energy spent elaborating misery should instead go toward finding one seemingly impossible way out, so founders should spend zero time on what they could have done and devote their time to what they might do.
“Spend zero time on what you could have done and devote all of your time on what you might do”
Horowitz bought Tangram for about $10 million after learning that EDS's customer valued Tangram's product and a $20 million annual contract depended on keeping it happy; Tangram had been valued at only $6 million, and the purchase preserved revenue.
“Ben buys the company for $10 million, rolls their product into the Opsware offering. And by buying a $10 million company, he saves a $20 million a year contract.”
When raising money privately, you only need one investor to say yes, so founders should ignore the many who decline, just as products always have more non-buyers than buyers.
“Look for a market of one. You only need one investor to say yes. It's best to ignore the other 30 who say no.”
Horowitz's main advice to entrepreneurs is to embrace their unusual background and instincts, because the keys to their unique approach are inside them and advisors can only help them find those keys.
“embrace your weirdness, your background, your instinct. If the keys are not in there, they do not exist.”
The founder of Front, Mathilde Collin, said that the most helpful part of business school was meeting founders who had built companies, a finding the host says matched his own experience of learning more from successful founders' talks than from two years of coursework.
“the thing in business school that was most helpful was meeting entrepreneurs that had built companies.”
Frugality is a key founder lesson, and it should be built into culture through simple mechanisms rather than enforced by auditing expenses, because a company that delivers value to customers must be both the price and service leader.
“he decided to build frugality into his culture. He did it with an incredibly simple mechanism.”
Founders are often required to prove beyond doubt that they can run the company they created, whereas investors are not asked to prove they can run a company, which Horowitz and Andreessen saw as a paradox to flip.
“why as founders we had to prove to our investors beyond a shadow of a doubt that we could run the company”
A company run by an interim CEO is a sign that its board may be willing to sell, because most boards would rather sell than gamble on hiring a new permanent CEO.
“Tangram was run by an interim CEO, which is a great sign that they'd be willing to sell the company because most boards would rather sell a company than roll the dice on hiring a new CEO.”
Founders who cope with stress through drinking, checking out, or quitting always offer marvelous rationalizations, while great founders consistently say they did not quit, so persistence is the distinguishing trait.
“The great founders tend to be remarkably consistent in their answers. They all say, I didn't quit.”
Founders should copy proven strategies from successful people, such as the CAA network model, then add their own unique background, which in turn becomes a new idea others can copy and extend, increasing diversity of ideas instead of centralization.
“embracing the unusual parts of my background would be key to making it through.”
Most business relationships either become too tense or too complacent, but Horowitz and Andreessen sustained 18 years of productive partnership because each regularly finds fault in the other's thinking.
“With Marc and me, even after 18 years, he upsets me almost every day by finding something wrong in my thinking, and I do the same for him. It works.”
Loudcloud was conceived when Horowitz learned that AOL partner sites crashed under traffic because scaling software to millions of users is far more complex than to thousands, suggesting a company could be built to solve this for everyone.
“There ought to be a company that does that all for them.”
Horowitz reports that the company he built, and Google's pre-IPO experience, both show that large early offers can be rejected when the eventual market is far larger than any buyer's own business.
“it didn't make sense for Google to sell to any suitor at any price that the buyer would have paid.”
Of roughly 600,000 podcasts in the Apple directory, about 75 percent no longer add episodes, and podcast networks help new shows grow because each network show promotes the others.
“600,000 podcasts in the Apple podcast directory. 75% of those podcasts are no longer adding episodes.”
Jim Clark, co-founder of Netscape, was constantly paranoid that Microsoft would destroy Netscape by bundling a free browser with its operating system monopoly.
“he was constantly paranoid of Microsoft coming in and basically destroying Netscape.”
Pixar began selling expensive computers with no success, then decided that if it was going to fail it might as well fail doing what it really wanted, making computer animated films; the host sees this as the same question-reframing pattern Loudcloud used.
“if we're going to fail, we might as well fail doing what we really want to do”
Richard Branson's definition of a business as an idea that makes someone's life better, when understood as solving a problem for someone, implies an infinite number of possible businesses.
“a business is just an idea that will make someone's life better.”
Walt Disney kept two inept writers on staff because seeing what the wrong way looks like helped his team find the right way, a lesson the host connects to Horowitz's approach.
“He kept two inept writers on staff because he said they'd always do it the wrong way.”
A hyper-capable engineer hired at Opsware became the company's best employee in one quarter, then deteriorated because of untreated bipolar disorder, medication refusal, and cocaine addiction, showing that raw talent does not guarantee reliability.
“Some brilliant people can be totally unreliable. At Opsware, we once hired an unequivocal genius.”
Marc Andreessen revealed Netscape's launch strategy to a trade publication two weeks early without telling Horowitz, prompting an angry response that shows how the two partners' early relationship was forged through conflict.
“Two weeks before the launch, Marc, without telling Mike or me, reveal the entire strategy to the publication, Computer Reseller News.”
The most important entrepreneurial lesson is to focus on what you need to get right and stop worrying about what you did wrong or might do wrong.
“Perhaps the most important thing I learned as an entrepreneur was to focus on what I needed to get right and stop worrying about all the things that I did wrong”
Horowitz's book notes that Pixar's founders, Jobs, and others made decisions that defied conventional expectations, and the host connects the pattern that breaking dominant convention can be an advantage.
“Pixar was first started, they were selling a really expensive computer.”
Books that try to provide a recipe for hard business challenges are misguided, because complicated dynamic situations like building a high-tech company have no formula; the best one can offer is advice, experience, and clues.
“There is no formula for dealing with them. Nonetheless, there are many bits of advice and experience that can help with the hard things.”
Negativity toward what you are building is inevitable and you cannot let it bother you, so focus on satisfied users rather than critics.
“negativity towards what you're doing is inevitable. You can't let it bother you.”
Marc Andreessen joked that startups produce only two emotions, euphoria and terror, and that lack of sleep intensifies both, capturing the emotional volatility founders face.
“You only ever experienced two emotions, euphoria and terror, and I find that lack of sleep enhances them both.”
A venture firm's advisors should have experience founding and running companies, since advice to founders is only credible from people who have done it.
“some experience in founding and running companies is required to advice people who are founding and running companies.”
Horowitz's advice is to focus on developing strengths rather than avoiding weaknesses, which he presents as the concise principle behind his hiring and self-improvement.
“This is another great concise thought, and it's about hiring for strength, not for a lack of weakness.”
In the early days of a company there is no inertia moving it forward, so without massive input from the founder the company stays at rest.
“In the early days of a company, there is no inertia that's putting the company in motion without massive input from you”
Horowitz's book is unusually candid about mental health, describing unhealthy stress, crying, not sleeping and sweating, which the host argues founders rarely discuss but need to learn to manage especially when in charge.
“that's like how do you manage your own mental health, especially when you're in charge.”
Almost every college now has entrepreneurship programs, which the host calls oxymoronic, since over a decade ago the pilot entrepreneurship program was distinctive, while today nearly all colleges offer one.
“almost every college has entrepreneurship programs, which is kind of oxymoronic in a way.”
Valve, described as the most profitable company per employee in the United States, has a new-employee handbook that reveals a philosophy for designing companies given modern technology.
“it's on the most profitable company in the United States per employee, and that's the company Valve.”
Horowitz's advice on consistency, that great founders say they did not quit, aligns with Paul Graham's finding that determination, not intelligence, is what predicts success.
“So this is another example of determination or another way, I guess, to put that is perseverance.”
Marc Andreessen was 22 when Horowitz interviewed with him, and Horowitz found his knowledge of technology history unusually deep, suggesting that studying the history of a field is a durable advantage.
“I was shocked by how much a 22-year-old kid knew about the history of the computer business.”
Founders who abandon an idea after negative feedback make the mistake of focusing on critics rather than the people whose problem the product solved for them.
“I think that's the exact wrong conclusion that he should have focused on.”
Horowitz reflected that selling Opsware, which took eight years and all his life force, felt like the smartest move despite physical sickness, and he realized it was time to start a new kind of venture capital firm.
“I realized that it was the smartest thing I'd ever done in my career.”
Taking extensive notes and reviewing them is the fastest way to retain information from books and interviews, which is why the host created a weekly service of founder interview highlights.
“it's the fastest way to get the information.”
If Founders can sustain a project over a decade or more, covering 500 to 600 founder books, it would document the history of entrepreneurship and learn from hundreds of company builders.
“we would kind of be documenting the history of entrepreneurship in a way.”