- General Overview
- The Central Thesis
- No recipe: complex, dynamic situations—layoffs, crises, entitlement—have no formula
- Easy versus hard: goals, hiring, and org charts are easy; survival decisions and cold-sweat nights are hard
- Experience over doctrine: personal experience, not conventional wisdom, is the only real knowledge
- Emotion fights logic: hard things are hard because feeling overwhelms reasoning; stories offer clues, not answers
- Origins: From Communist to Venture Capitalist
- Red-diaper childhood: grandfather lost his job in the McCarthy era; father edited the radical magazine Ramparts
- Wagon lesson: asking a favor despite terror won a lifelong friend; surfaces mislead, only experience teaches
- Perception matters: an alternate plausible reading of events can keep a frightened workforce hopeful
- Family first: daughter Mariah's autism diagnosis and a failing home life forced quitting NetLabs
- Founders over hired managers: NetLabs' professional managers knew no technology—founders should run companies
- Loudcloud to Opsware: Surviving and Winning
- Free-capital question: Rachleff's prompt drove unconstrained ambition—200 employees and $10M in contracts in seven months
- Crash: forecast fell from $100M to $37M and thirty investors refused, yet one yes raised $120M
- Unwanted IPO: sold at $6 with three weeks of cash while the press called it "IPO from hell"
- Eat shit once: Conte's rule—absorb all the pain at once instead of nibbling away credibility
- Selling Loudcloud: EDS paid $63.5M cash for the cloud business, freeing Opsware to sell software
- Darwin project: six months of seven-day weeks beat BladeLogic; Opsware eventually sold for $1.65B
- When Things Fall Apart
- The Struggle: the dream becomes a nightmare; self-doubt turns to self-hatred and isolation
- Calculus, not statistics: the CEO's job is the same at one-in-a-thousand odds as at nine-in-ten
- Share the burden: more brains on the problem; an honest all-hands can rally the team to save you
- Tell it like it is: required communication is inversely proportional to trust; bad news must travel fast
- Layoffs done right: get your head right, don't delay, blame the plan, let managers fire their own people
- Firing executives: treat it as a system failure; script it, brief the board, let her shape the story
- People, Products, and Profits—in That Order
- Barksdale's order: take care of the people, then the products, then the profits
- Hire for strength: Powell's rule—do the job yourself to know exactly what strength to hire
- Training is the boss's job: twelve hours of teaching can yield two hundred hours of output; make it mandatory
- Good versus bad organization: good ones focus on work; bad ones fight boundaries and broken processes
- Why goodness pays: with product-market fit, only liking the job tethers people; bad spirals rarely reverse
- Good PM, bad PM: the good PM is CEO of the product—defines "what," owns delivery, makes no excuses
- The Going Concern: Culture, Politics, and Titles
- Profanity policy: a clarity tool, acceptable unless used to intimidate or harass
- Politics starts with the CEO: refusing to play politics breeds the fiercest kind; fixed process is the antidote
- Right ambition: hire people whose success follows company success—two percent of zero is zero
- Titles degrade: the Peter Principle and Law of Crappy People; disciplined leveling is the mitigation
- Bad smart employees: the Heretic, the Flake, and the Jerk—brilliance without reliability destroys teams
- Programming culture: a few shocking mechanisms—door desks, punctuality fines—that fit your values
- Leading When You Don't Know Where You're Going
- CEO psychology: the hardest, untaught skill—no one to blame, and every company hits "we're fucked" moments
- Courage over intelligence: the hardest decisions test courage; the right path is often obvious but unpopular
- Ones and twos: ones love strategy and decisions, twos love process; CEO succession between them is hard
- Leadership triad: vision, ambition aligned with employees, and earned competence create followers
- Peacetime versus wartime: tolerating deviation versus breaking protocol; context dictates which CEO you must be
- Feedback: come from the right place, stay direct and private, and deliver it at high frequency
- No Rules: Entrepreneurship and the End of the Beginning
- No rules: when the sky turns purple, don't argue it's blue—adapt and move on
- Accountability versus creativity: punishing misses kills risk; ignoring them makes reliable performers feel like chumps
- Evaluating executives: judge today's performance against real need, not a hypothetical scaled-up job
- Selling the company: sell when the market is redefined and offers signal a local maximum
- a16z: "some experience required"—VCs who ran companies, building networks as infrastructure
- Embrace the struggle: hard things are hard because emotion fights logic; your odd background holds the keys
- The Central Thesis
- Deep Dive
- Introduction
- The Real Hard Things
- Missed by advice books: hard parts are layoffs, entitlement, communication breakdowns, and cold-sweat nights
- Contrast: goals, hiring, org charts, and dreams are easy by comparison
- No Recipe for Complexity
- Core premise: complex, dynamic situations have no formula
- Examples: company building, crisis leadership, hit songs, NFL quarterbacking, presidential runs
- Book’s approach: shares experience and story, not a recipe
- Lessons from Struggle
- Origin: blog posts read by millions, now paired with their backstories
- Hip-hop lens: artists as entrepreneurs—competing, earning, being misunderstood
- Purpose: offer clues and inspiration to builders of something from nothing
- The Real Hard Things
- Chapter 1: From Communist to Venture Capitalist
- Fear, Perspective, and Priorities (Chapter 1: From Communist to Venture Capitalist · I)
- Berkeley Roots
- Red-diaper family: grandfather lost his teaching job in the McCarthy era; father edited the New Left magazine Ramparts.
- Shy child: cried through nursery school until his mother’s patience let him enter the world.
- Diverse block: Bonita Avenue mixed hippies, strivers, and drug-using upper classes—no conventional middle class.
- The Wagon Lesson
- Roger’s dare: a troubled friend ordered Ben to demand a wagon and spit a racial slur.
- Fear and action: being terrified didn’t make him gutless; asking for a ride won a “sure” and a lifelong friend.
- No shortcuts: surfaces mislead; personal experience, not conventional wisdom, is the only real knowledge.
- Football and Leadership
- Turn your shit in: Coach Mendoza’s opening speech made accountability unforgettable.
- Leadership defined: Powell—get someone to follow you, even only out of curiosity.
- Multiple circles: football and calculus classmates gave radically different readings of the same events.
- Facts vs. perception: an alternate plausible scenario can keep a worried workforce hopeful.
- Blind Date
- Blind date: Felicia bailed after dinner was cooked; an empathy appeal brought her back.
- Fistfight: a basketball scuffle left a black eye; first impressions screamed “thug.”
- Marriage: both ignored first impressions; nearly twenty-five years and three children followed.
- Silicon Valley and Priorities
- SGI dream: invented modern computer graphics; everyone smart, products cool—a dream job.
- NetLabs mistake: professional managers knew no tech; founders, not hired execs, should run companies.
- Autism diagnosis: daughter Mariah’s needs made startup work a terrible family burden.
- Father’s joke: flowers are cheap, divorce is expensive; priorities needed resetting.
- Family first: doing everything meant failing the most important thing; become a man, not a boy.
- Berkeley Roots
- Family, Netscape, and the Cloud (Chapter 1: From Communist to Venture Capitalist · II)
- First Things First
- Family before self: quit NetLabs to stabilize home life, stopped self-focus, became who he wanted to be.
- Lotus as reset: found a job at Lotus to get home life straightened out before chasing career.
- Mosaic revelation: graphical interface to the Internet made the future obvious; working off-Internet felt like wasted time.
- Netscape attraction: after reading about Jim Clark and Marc Andreessen, Ben immediately sought an interview.
- Internet vs. Information Superhighway
- Proprietary status quo: Oracle and Microsoft built toll-based Information Superhighway; they would tax every transaction.
- Internet dismissed: most thought it too arcane, insecure, and slow; few companies ran TCP/IP.
- Gates's blind spot: The Road Ahead predicted the Information Superhighway, not the Internet, until later revised.
- Netscape's mission: Marc and Clark pivoted from video distribution to making the Internet secure and usable.
- Netscape's Rise and Microsoft's Threat
- Marc interview: no résumé questions; deep history of email/collaboration; Ben called him smartest ever.
- Web server line: Ben ran Enterprise Web Server; two products, $1,200 regular and $5,000 secure SSL, started with two engineers.
- Historic IPO: priced $28, spiked to $75, closed at $58; $3B market value only sixteen months after founding.
- Microsoft bundling: IE free with Windows 95 killed browser revenue; answer shifted to server sales.
- IIS reality check: Microsoft's server had every feature and was five times faster; five-month window to respond.
- SuiteSpot and the Marc Blowup
- Mike Homer warrior: grew stronger in crises, faced press alone during Microsoft's "embrace and extend."
- SuiteSpot counter: dirt-cheap open alternative to Microsoft BackOffice; Informix gave database access for $50 a copy.
- Marc leaks early: revealed strategy to Computer Reseller News two weeks before launch.
- Furious email: Marc blamed server product management for $3B loss; Ben feared for his job.
- Tension that lasted: SuiteSpot grew to $400M/year; Marc and Ben built an 18-year partnership on daily mutual challenge.
- Netscape's Legacy and Loudcloud
- AOL sale: Microsoft's subsidies forced Netscape into AOL; as a media company, AOL didn't match Ben and Marc's technology focus.
- Winning long-term: Netscape shifted developers from Win32 to web standards, breaking Microsoft's platform monopoly.
- Foundational tech: JavaScript, SSL, and cookies were invented at Netscape.
- Loudcloud idea: partners' sites crashed at scale; developers needed a cloud for security, scaling, and disaster recovery.
- Loudcloud team: Tim Howes (LDAP) and In Sik Rhee (Kiva) joined as cofounders.
- Cloud naming: borrowed from telecom's "smart cloud"; "cloud" as computing platform was Loudcloud's lasting remnant.
- First Things First
- Fear, Perspective, and Priorities (Chapter 1: From Communist to Venture Capitalist · I)
- Chapter 2: “I Will Survive”
- Euphoria, Terror, and Survival (Chapter 2: “I Will Survive” · I)
- Euphoric Launch and Free Capital
- Benchmark led: $15M at $45M pre; Marc Andreessen added $6M and became chairman.
- Free-capital question: Andy Rachleff's prompt pushed Loudcloud to think unconstrained about growth.
- Explosive growth: 200 employees and $10M in contracts within seven months of founding.
- Bigger bets: ninth hire was a recruiter; "the Taj" was too small before move-in.
- Crash and Fundraising Reality
- Dot-com crash: NASDAQ lost 80% from its peak, turning startups into dot-bombs.
- Cash burn: almost all $66M was deployed; 300 employees and little cash left.
- Softbank rejection: back-channel report: "they thought you were smoking crack."
- Market of one: one yes produced a $120M Series C at $700M pre; ignore the thirty no's.
- Forecast collapse: $100M predicted bookings became $37M; "the wheels came off."
- Bill Campbell, the Board's Trusted Voice
- Legendary mentor: former Intuit CEO, college coach, and trusted by Jobs, Bezos, and Schmidt.
- Two friends you need: one who is genuinely excited for you, plus one who takes your crisis call.
- Decisive in deadlock: Ben listed IPO cons; Bill said "It's not the money"—then "It's the fucking money."
- The Unwanted IPO
- Only path: private funding shut down; the public window was slightly open.
- Revenue mismatch: two-year contracts recognized monthly; S-1 showed tiny trailing revenue against a huge forecast.
- Unready company: weak sales processes, bankrupting customers, huge losses, shaky operations.
- Press hammering: Red Herring joked about burning dollar bills; BusinessWeek called it "IPO from hell."
- Cash bombshell: banker discovered $27.6M restricted cash after underwriting; only three weeks of cash.
- Reverse split fury: employees' fantasy share price halved; the all-hands meeting turned hostile.
- Road Show and Family Cost
- Brutal road show: markets crashed daily, investors looked tortured; Ben slept two hours total.
- Shifting benchmark: comparable companies lost half their value, making the $10 price roughly double.
- Felicia's crisis: wife stopped breathing from an allergic reaction while he was on tour.
- Her command: "Get the IPO done. There is no tomorrow for you and the company."
- Discombobulation: mismatched suit, lost bearings; he neglected the only thing that mattered.
- Euphoric Launch and Free Capital
- Surviving the Dot-Com Crash (Chapter 2: “I Will Survive” · II)
- The Nadir IPO
- IPO at the bottom: Loudcloud sold at $6 a share and raised $162.5M, but no celebration or closing dinner.
- Yahoo’s signal: Tim Koogle’s ouster marked the dot-com crash’s nadir and crushed morale.
- CEO’s depression: the night before the offering was a highlight of depression, not triumph.
- Kupor’s quip: “We did it!” — “Yeah, but we’re still fucked.” captured the mood.
- Resetting Guidance
- Forecast trap: company could meet the quarter but not the year; first-call reset destroys credibility.
- Conte’s rule: “If you are going to eat shit, don’t nibble” — take all the pain at once.
- Slash and layoffs: revenue guidance cut from $75M to $55M; 15% of employees sent home.
- Banks defect: Goldman and Morgan Stanley dropped research coverage; stock fell from $6 to $2.
- 9/11 close call: British government deal nearly vanished into the war chest, leaving the quarter fragile.
- Exodus bankruptcy: $50B competitor collapsed, proving the cloud business needed a backup plan.
- Wartime CEO
- Transition moment: two MBAs’ 45-slide critique earned “Did I ask for this presentation?”
- Sole accountability: survival or destruction rested on CEO decisions; no excuses could help.
- Delegation boundary: managers still owned their areas, but the survival question was Ben’s alone.
- Data Return insight: their double trouble revealed Loudcloud’s fate; easier to see in others.
- Plan B: Oxide
- New question: asked what to do after bankruptcy — buy Opsware out and start a software company.
- Opsware: automation software for provisioning, deployment, and recovery; needed to become standalone.
- Oxide project: ten engineers began the nine-month separation, kept secret from staff.
- Public framing: Oxide was called a product line; revealing exit would doom the cloud business.
- Atriax collapse: largest customer went bankrupt, owed $25M; stock fell 50% and PIPE financing died.
- Doomed plan: cash gap widened to $75M below breakeven; cloud business could not survive.
- Selling Loudcloud
- Spiral named: Bill Campbell confirmed the loop: lower cash, weaker confidence, lower sales, less cash.
- M&A focus: needs trump wants; EDS needed Loudcloud more than IBM, so John found Jeff Kelly.
- Ovitz tactics: artificial deadlines, playing bidders against each other, everything short of illegal.
- EDS deal: $63.5M cash for Loudcloud; retained Opsware and licensed software for $20M/year.
- Aftermath
- Human cost: 150 employees moved to EDS; 140 were laid off.
- Bill’s timing advice: tell every employee where they stand immediately, not a minute later.
- Trust rebuilt: fair treatment of departing employees was the foundation for those who stayed.
- The Nadir IPO
- Euphoria, Terror, and Survival (Chapter 2: “I Will Survive” · I)
- Chapter 3: This Time with Feeling
- Reselling the Story to Survive (Chapter 3: This Time with Feeling · I)
- Selling the Dream Twice
- Post-sale collapse: EDS sale left no customers or revenue; stock fell to $0.35, about half the cash in the bank.
- Off-site honesty: Ben shared everything, issued new grants, and asked undecided employees to quit that day.
- Loyalty dividend: Two quit; seventy-eight stayed, and all but two remained through the HP sale later.
- Resurrecting the Company
- Stock story: Beat delisting by telling a simple story—team, $60M bank, EDS contract, IP—and the stock crossed $1.
- Ship the wrong product: Opsware wasn't productized; entering the market required selling before knowing the right product.
- Executive rebuild: Let go CFO, sales, and marketing leaders who were great at old jobs but unqualified for new ones.
- The EDS Ultimatum
- Existential customer: EDS was 90 percent of revenue; stalled deployment meant cancel, refund, and the end of Opsware.
- Frank Johnson: The EDS executive erupted, called Opsware a piece of shit, and vowed to throw them out.
- Sixty-day clock: Anthony got Frank to grant sixty days to fix everything—the company had sixty days to live.
- The Must-Win Plan
- Remove roadblocks: Daily meetings with Jason and Anthony ensured no project delay lasted more than twenty-four hours.
- Split roles: Jason delivered expected value; Anthony hunted for unexpected, exciting value beyond contract promises.
- Tangram insight: Anthony discovered Frank loved Tangram inventory software and hated the free CA replacement being forced on him.
- Acquire Tangram: Ben bought the $6M-market-cap firm for $10M despite unanimous objections, to give Frank a free win.
- Reckoning and Humanity
- Frank converted: At day sixty, Frank called Opsware his best vendor, saving the account and the company.
- Tangram payoff: The acquisition became highly profitable; Cary engineers proved cheaper than Bangalore after recruiting and training costs.
- John Nelli: Tangram's CFO developed brain cancer between signing and close; Ben spent about $200K to qualify him for COBRA.
- Quiet compassion: Ben paid while Opsware was fighting for life; John's widow sent a handwritten letter after his death.
- Selling the Dream Twice
- Desperation, Reinvention, and the Exit (Chapter 3: This Time with Feeling · II)
- Desperation and the Human Element
- Empathy from experience: Ben helped a stranger because he knew what desperation felt like.
- Gratitude letter: she said his help saved her from total despair and enabled her to keep living.
- Running on empty: after endless crises, he felt no stories, no speeches, and no “rah-rah” left.
- Founding partner’s exit: Marc left to found Ning, leaving Opsware’s fate to the team.
- Plainspoken plea: Ben leveled with engineers: “We have one bullet left in the gun.”
- Survival of the Fittest: The Darwin Project
- The threat: BladeLogic beat Opsware in key accounts, dropping the stock to $2.90.
- All-in commitment: engineers worked seven days a week, 8 a.m.–10 p.m., for six months.
- Team response: the do-or-die mission felt hard and fun; no one quit.
- Junior leaders emerged: young people stepped up when thrown into the ocean and told to swim.
- Outcome: Opsware began winning proofs of concept; Ben cried reading Ted’s account years later.
- Innovation and Sales Discipline
- Requirements trap: product management resisted walking away from known needs for speculative wins.
- Innovator’s duty: the right product is the innovator’s job, not the customer’s job.
- Founder’s courage: Ben overrode data, demanding a reinvented product to beat BladeLogic.
- Sales overhaul: Mark Cranney retrained every rep and refused to tolerate sloppiness.
- Deal verification: a rep with no VP contact was mocked: “You do not have a deal.”
- The Offensive Question: What Are We Not Doing?
- Staff meeting agenda: asking “What are we not doing?” surfaced the neglected network automation market.
- Old code base: Jive was too weak to commercialize, forcing a build-or-buy decision.
- Efficient market myth: markets converge on conclusions, often wrong; Opsware traded at half its cash.
- Acquisition logic: best product architecture, Rendition Networks, had lowest revenue but won.
- Cisco deal: a $30 million prepaid license covered over 90% of the $33 million acquisition.
- The Ultimate Decision
- Virtualization shift: technology changes threatened Opsware’s architecture while enabling cloud.
- Pressure to sell: an activist shareholder demanded removal despite a 10x gain on her shares.
- Auction strategy: recruit Oracle as the “rabbit” to force bidders to chase.
- Internal debate: Ben argued with himself for weeks over whether the team and architecture could survive.
- Team voice: all but one direct report favored sale; Ben set $14 per share as his price.
- Emotional close: selling for $1.65B felt like death, then proved the smartest career move.
- Desperation and the Human Element
- Reselling the Story to Survive (Chapter 3: This Time with Feeling · I)
- Chapter 4: When Things Fall Apart
- Facing the Struggle with Honest Leadership (Chapter 4: When Things Fall Apart · I)
- The Struggle
- Definition: the dream becomes a nightmare; walls close in and every answer seems wrong.
- Symptoms: self-doubt becomes self-hatred, food loses taste, vacations feel worse.
- Isolation: surrounded by people, yet alone; the Struggle is land of broken promises and crushed dreams.
- Bushido rule: keeping death in mind prepares a CEO to handle all other decisions.
- Paradox: the Struggle is not failure, but it causes failure; it is also where greatness comes from.
- Believe in Calculus, Not Statistics
- Two futures: a determinate world favors calculus; an indeterminate one favors statistics and bell curves.
- CEO odds: never play the odds; the task is same whether chances are nine in ten or one in a thousand.
- Bill Campbell's warning: prepare the company for bankruptcy even while fighting for the deal.
- The dying patient: some companies are dead without knowing it; only a true friend names that truth.
- No good moves: the one skill that matters is focusing and making the best move when none exists.
- Navigating the Struggle
- Share burdens: nobody takes losses harder than the CEO; get the maximum number of brains on problems.
- Honest all-hands: telling the company you're getting your ass kicked can rally the team to save you.
- Chess, not checkers: technology is three-dimensional chess; there is always a move.
- Don't take it personally: you made the mistakes, but self-flagellation doesn't help; this challenge separates the great.
- Play long enough: technology tomorrow looks nothing like today; surviving brings luck and answers.
- Stop Being Too Positive
- Positivity delusion: projecting sunny optimism made employees doubt the CEO's grasp of reality.
- CEO feels losses most: employees can walk away; the founder is married to the company.
- Put problems where they can be solved: engineers who write code should own product worries.
- Share the setbacks: losing a big prospect is organizational intelligence, not a secret to hide.
- Tell It Like It Is
- Trust: required communication is inversely proportional to trust; transparency builds it.
- Brains: big brains cannot solve problems they don't know about; enough eyeballs make bugs shallow.
- Bad news culture: healthy companies reward people for surfacing problems; cover-ups kill.
- Beware maxims: "Don't bring me a problem without a solution" can suppress vital information.
- Final thought: resist the psychological pressure to be positive; tell it like it is.
- The Struggle
- Layoffs and Executive Firings, Done Right (Chapter 4: When Things Fall Apart · II)
- Layoffs Defy Venture Physics
- Layoffs usually break culture: survivors stop sacrificing after watching friends get cut.
- Recovering from repeated layoffs: rare; a $1.6B outcome after three layoffs violated VC “physics.”
- The differentiator: laying people off the right way preserves cultural continuity and top talent.
- Layoff Steps 1–3: Honest Framing
- Get your head right: ignore self-blame and focus on the future, not the overwhelming past.
- Don’t delay: compress decision-to-execution time; once word leaks, managers and trust get damaged.
- Be clear on cause: the company failed its plan—not the employees; say so plainly.
- Rebuild trust: admit the failure explicitly; positive spins about “performance” destroy credibility.
- Layoff Steps 4–6: Dignity in Execution
- Managers lay off their own people: never outsource; people remember every detail of that day.
- Train managers: brief company-failure explanation, nonnegotiable decision, full benefits details.
- CEO addresses the company first: message is for those staying—respect the departing, but don’t over-apologize.
- Be visible afterward: don’t disappear; talk, help carry boxes, show appreciation.
- Firing Executives: Root-Cause First
- Treat firing as a system failure: a wrongful executive hire reflects a broken interview/integration process.
- Hiring pitfalls: vague role definition, generic “great” profiles, hiring for lack of weakness.
- Hire for scale before you’re ready: overqualified executives get rejected before they can contribute.
- Ambition and integration matter: self-focused ambition or poor onboarding can derail a skilled exec.
- Special Cases: Scaling and Fast Growth
- Scaling changes every job: at quadruple size, old roles are new jobs; requalify or replace.
- Not a failure: it’s life in the big city—nobody is to blame, and resisting makes things worse.
- Fast growth needs proven speed: hire executives who have scaled quickly before, and fund them fully.
- Board, Script, and the Conversation
- Inform the board individually: alarm them now or enable department rot—the first is far better.
- Three board goals: support the plan, approve the package, preserve the fired executive’s reputation.
- Script your remarks: clear reasons, decisive language (“I have decided”), severance approved.
- Let the executive shape the story: she chooses how the news reaches the company and the world.
- Layoffs Defy Venture Physics
- Hard Calls and Lead Bullets (Chapter 4: When Things Fall Apart · III)
- Announcing an Executive Firing
- Communication order: direct reports, staff, then company—all within hours.
- Interim plan: make clear who reports to whom; CEO often steps into role.
- CEO stand-in: actually run staff meetings, one-on-ones, objectives for continuity.
- Don't trash the departed: top employees loyal to her will fear they're next.
- Treat employees as adults: they handle bad news better than being managed.
- The real test: how a CEO handles a hard firing reveals whether the company is great.
- Demoting a Loyal Friend
- Decide first: demotion is right when company needs skills your friend lacks.
- Expect emotions: embarrassment and betrayal will drive an intense discussion.
- Stay decided: enter with a firm decision or you'll leave with a mess.
- Accept risk: if you can't afford to lose him, you cannot make the change.
- Offer a move: sideways into a new area beats reporting to new boss as demotion.
- Use honest language: "I have decided," admit your own limits, acknowledge his contributions.
- Lies That Losers Tell
- Common deceptions: attrition is "performance-related," lost deals were "given away," missed milestones don't matter.
- Root cause: leaders only act on positive leading indicators, explain away negative ones.
- Andy Grove's insight: CEOs lying to investors are first lying to themselves.
- Contagion: honest employees lie to themselves; believing them makes you do the same.
- Lead Bullets vs. Silver Bullets
- No silver bullet: fundamental competitive gaps demand lead bullets—hard, direct work.
- Netscape: fixed server performance, built a $400 million business.
- Opsware: resisted pivots, rebuilt the better product, reached $1.6 billion.
- Fight, don't flee: if you're running when you should fight, ask if the company must exist.
- Nobody Cares
- Al Davis's rule: nobody cares about your excuses—just run the company.
- Reasons fail: great reasons preserve no dollars, jobs, customers, or self-respect.
- Use energy forward: spend zero time on what could have been; find the one way out.
- Announcing an Executive Firing
- Facing the Struggle with Honest Leadership (Chapter 4: When Things Fall Apart · I)
- Chapter 5: Take Care of the People, the Products, and the Profits—in That Order
- Hire for Strength, Care for People (Chapter 5: Take Care of the People, the Products, and the Profits—in That Order · I)
- Hire for Strength, Not Lack of Weakness
- Hire for strength: Colin Powell's rule; run the job yourself so you know the exact strengths needed.
- Interview test: ask candidates to describe hiring, training, and managing reps; most cannot explain training.
- Mark Cranney: square, awkward, unknown school—but a sales savant with a custom training manual and 75 references.
- Negative reference: only objection was cultural fit—ethics, hiring, deals, and management all checked out.
- Reframe flaws: his weaknesses were the price of genius—without them he'd be CEO of IBM.
- Wartime Leadership
- Peacetime vs wartime: peacetime values appropriateness and feelings; war demands killing the enemy and getting troops home.
- Wartime general: Ben saw himself at war and chose a leader who scared recruits but could win.
- Stand against consensus: board and staff voted no; Ben proceeded anyway, trusting strength over comfort.
- Outcome: the team eventually agreed Cranney was the best possible hire despite lingering discomfort.
- A Good Place to Work
- Barksdale's order: take care of people, products, profits—in that order; people is hardest and everything depends on it.
- Good vs bad organization: in good organizations people focus on work; in bad ones they fight boundaries and broken processes.
- Management training: Ben taught expectations and required one-on-ones, but found a manager skipping them for six months.
- Explain the why: authority alone wasn't enough; Ben had to articulate the deeper reason behind each practice.
- Why he works: he wants people who spend most waking hours at work to have a good life.
- Accountability: the manager's boss got a 24-hour ultimatum: hold one-on-ones or both are fired.
- Why Being Good Pays Off
- Product/market fit caveat: horribly managed companies can succeed with fit, so goodness matters most when things go wrong.
- Reasons to stay evaporate: career, wealth, and prestige vanish; liking the job is the only tether left.
- Death spiral: employees leave, value drops, best employees leave; spirals are extremely hard to reverse.
- End in itself: Bill Campbell's GO Corp failed financially, yet its alumni became industry leaders—good culture is intrinsically worthwhile.
- Hire for Strength, Not Lack of Weakness
- Training, Management, and Ethical Hiring (Chapter 5: Take Care of the People, the Products, and the Profits—in That Order · II)
- Be Like Bill: Build a Good Company
- GO's paradox: employees called it a best-ever workplace despite stalled careers, no money, and public failure.
- Bill Campbell's legacy: judged by building a good company, not returns — Doerr still recommended him for Intuit's CEO.
- Why Startups Should Train Their People
- The absurd gap: McDonald's trains line workers; tech firms expect brilliance with zero training.
- Grove's rule: "Training is the boss's job" — managers must teach, not outsource to outside firms.
- Netscape proof: one "Good Product Manager/Bad Product Manager" document transformed underperformers into a top team.
- Pay it forward: that document and Grove's High Output Management drove Loudcloud's training investment.
- Why You Should Train Your People
- Productivity: training is the highest-leverage act — 12 hours of teaching can gain 200 hours of output.
- Performance management: you can only fire with certainty if training established the expectations first.
- Product quality: untrained engineers Frankenstein an elegant architecture through duplicated, inconsistent code.
- Employee retention: people quit over bad managers and no learning; training fixes both.
- What Should You Do First?
- Functional training: teach the job's core knowledge, from role expectations to engineering boot camps.
- Management training: set explicit expectations — one-on-ones, feedback, training, objectives — then teach the skills.
- Peer expertise: let stars teach negotiation, interviewing, finance; teaching becomes an honor badge.
- Implementing Your Training Program
- Make it mandatory: withhold new-employee requisitions from managers without a training plan.
- CEO owns expectations: teach the management-expectations course yourself; top managers teach the rest.
- No time excuse: "Too busy to train is the moral equivalent of being too hungry to eat."
- Good Product Manager/Bad Product Manager
- Ownership: the good PM is CEO of the product — full responsibility and a winning plan, never excuses.
- Define "what": good PMs crisply specify the target and manage delivery; bad PMs solve "how" and put out fires.
- Focus externally: good PMs aim at revenue, customers, and the press story; bad PMs chase competitor features.
- Discipline: good PMs decompose problems, take written positions, and file status reports on time.
- Is It Okay to Hire People from Your Friend's Company?
- How it happens: the employee is already looking and comes to you; you're the last step, not the raider.
- The read on Cathy: losing a star looks like betrayal and signals company demise to her remaining team.
- The one-time exception: explain it's a one-off; she'll forgive but never forget, and the memory lingers.
- Be Like Bill: Build a Good Company
- People, Loyalty, and Executive Hiring (Chapter 5: Take Care of the People, the Products, and the Profits—in That Order · III)
- Raiding a Friend’s Company
- Social pressure trumps logic: raiding a friend’s top talent wins the hire but loses the friendship.
- Reflexive Principle: never hire from a company whose raiding would horrify you.
- No-poach policy: designate companies needing CEO approval before hiring.
- Transparency: tell candidates you’ll reference-check with their CEO before offering.
- Talk to the friend first to gauge the relationship cost and avoid bad hires.
- Why It’s Hard to Bring Big Company Execs into Little Companies
- Different jobs: big-company execs optimize an existing machine; startup execs create momentum.
- Rhythm mismatch: waiting for incoming demands leaves small-company peers suspicious.
- Skill-set mismatch: running needs process; building needs hiring, domain expertise, and creation.
- Screen with role questions: first-month plans, job differences, and motivation for joining.
- Beware interrupt-driven answers: the promised incoming work never arrives.
- Beware equity-only motivation: 1% of nothing is nothing; want creators.
- Aggressively Integrate the New Executive
- Force creation: set monthly, weekly, and daily objectives; the company is watching.
- Demand real fluency: product, technology, customers, and market are non-negotiable.
- Daily Q&A: require questions from first principles; no questions means trouble.
- Put them in the mix: assign contacts and require learning reports.
- Thirty-day bar: fire if they’re not ramping by day thirty.
- Payoff: the right experienced exec accelerates your company—but watch leading indicators.
- Hiring Executives: If You’ve Never Done the Job, How Do You Hire Somebody Good?
- Know what you want: the most-skipped step; avoid interview-only education.
- Avoid hiring traps: look-and-feel, central-casting ideals, and lack-of-weakness logic.
- Act the role to learn the function well enough to judge candidates.
- Use domain experts for input, but keep the decision yours—they don’t know your company.
- Build a rigorous process: write desired strengths and tolerable weaknesses across functional, operational, strategic, and team criteria.
- Assemble two interview groups: those who evaluate fit and those who must support the hire.
- Raiding a Friend’s Company
- People, Metrics, Debt, and HR (Chapter 5: Take Care of the People, the Products, and the Profits—in That Order · IV)
- The Final Hiring Decision
- Two finalists: limit the final round to two candidates for focus.
- Talent-matched questions: assign interviewers who understand what a good answer sounds like.
- Calibrate criteria: debrief each interview to drive a common understanding of the yardsticks.
- CEO-run references: check both backdoor and front-door references against the hiring criteria.
- Solo decision: only the CEO has the full context; consensus favors lack of weakness over strength.
- When Employees Misinterpret Managers
- Hockey stick: extreme end-of-quarter bookings made planning and public reporting harrowing.
- Misaligned incentives: early-deal bonuses shifted revenue; choose revenue versus predictability up front.
- Asked for, not wanted: all three managers delivered the measurable goal, not the real qualitative goal.
- Numbers trap: managing purely by numbers is like painting by numbers—it sacrifices qualitative goals.
- White-box management: assess how results were produced, not just the numbers, to protect the future.
- Side effects: any measured goal creates behaviors; test the goal against the behaviors it will cause.
- Management Debt
- Definition: expedient, short-term management decisions accrue interest like technical debt.
- Two in the box: splitting one role blurs authority, removes accountability, and worsens over time.
- Counteroffer leak: matching a key employee's offer teaches everyone that threatening to quit earns raises.
- No feedback system: avoiding performance reviews creates fuzzy direction and systematically poor performance.
- Hard answer: experienced CEOs pay management debt by making the painful organizational choice now.
- Management Quality Assurance
- HR as QA: good HR can't create great culture, but it exposes when management isn't delivering.
- Employee life cycle: measure recruiting, comp, integration, performance, and motivation to assess management quality.
- HR leader skills: process design, diplomacy, and industry knowledge are all required.
- CEO's trusted adviser: must have intellectual heft and super-perceptive reading of unspoken slippage.
- The Final Hiring Decision
- Hire for Strength, Care for People (Chapter 5: Take Care of the People, the Products, and the Profits—in That Order · I)
- Chapter 6: Concerning the Going Concern
- Culture, Politics, and the Right Ambition (Chapter 6: Concerning the Going Concern · I)
- Profanity as a Policy Decision
- Profanity as clarity: A CEO's curse words amplify urgency and make messages spread.
- CEO Tourette's syndrome: Stress made profanity involuntary, not just intentional.
- Talent over propriety: Banning profanity shrinks the talent pool; top engineers come from profane cultures.
- Cupcakes rule: Profanity is acceptable unless used to intimidate or harass.
- Clarity, not solution: Complaints stopped once the policy was explicit and livable.
- Growth Forces Change
- Inevitable change: At scale, your company is different from its founding days.
- Embrace evolution: A good 1,000-person company requires admitting what must change.
- Same company myth: Ten-person culture cannot survive unchanged at 100,000 employees.
- Politics Starts With the CEO
- Apolitical CEOs: Not being political does not prevent politics; it often breeds the fiercest kind.
- Compensation incentives: Raising pay on request rewards self-advocacy over business contribution.
- Squeaky wheel effect: Aggressive executives get off-cycle raises while apolitical competence is penalized.
- Career discussions: Commenting on ambitions turns into political cannon fodder once executives twist your words.
- Anti-Politics Processes
- Right ambition first: Hire executives whose success is a by-product of company success.
- Fixed processes: Cover compensation, promotions, and territory with strict rules that don't bend.
- Handling raise requests: Refer executives to the formal performance and compensation cycle.
- Organizational design: Don't tip off potential reorgs; decide quietly, then execute fast.
- Visible promotions: A formal process gives confidence that promotions are merit-based and explainable.
- He Said, She Said
- Behavior complaints: Bring both executives together; never address the issue one-on-one.
- Competency complaints old news: If you already knew, the executive lost support; let them go.
- Competency complaints new news: Shut down agreement immediately; reassess before the rumor cripples the target.
- Festering danger: Unresolved accusations become self-fulfilling prophecies.
- Right Kind of Ambition
- Global vs local: Executives should optimize company success, not personal success.
- Two percent of zero is zero: If the company wins nothing, personal equity is worthless.
- Mission over manager: Talented employees won't work hard to advance a self-interested boss.
- High-IQ isn't enough: Wrong ambition poisons even the smartest management team.
- Profanity as a Policy Decision
- Ambition, Titles, and Smart Bad Employees (Chapter 6: Concerning the Going Concern · II)
- The Right Kind of Ambition
- Right ambition makes managers radically more valuable; wrong kind is dangerous, as Yertle the Turtle shows.
- “Me” prism: candidates claim credit, call the company “my play,” and speak of building their résumé.
- “Team” prism: candidates deflect credit, own their misjudgments, and care most about how your company wins.
- Sales leadership especially needs team-first ambition; local incentives otherwise breed fraud and customer harm.
- Mark Cranney: team-first screen hire; sales rose 10x, market cap 20x, and attrition stayed low.
- Senior leaders must do right things for right reasons; careerist individuals are tolerable, wrong-minded managers are not.
- Titles: Why They Exist and How They Go Wrong
- Titles arise from employee résumé needs, organizational shorthand, and peer comparisons for compensation.
- Peter Principle: competent managers keep promoting until they reach incompetence; unavoidable, so mitigate.
- Law of Crappy People: every title level degrades to its worst incumbent as juniors benchmark upward.
- Disciplined promotion process is the mitigation: crisp level definitions and leveling across groups.
- Promotions council: reviewers compare candidates against named exemplars and current holders to keep quality.
- Without disciplined leveling, employees obsess over title inequities; with it, titles become ordinary.
- Andreessen vs. Zuckerberg on Title Size
- Andreessen: titles are the cheapest ask; hand out Chiefs and Presidents to outbid competitors.
- Zuckerberg: deliberately low titles force releveling, internalize the system, and keep status meaningful.
- Facebook’s low titles cost some hires but select out people who care too much about titles.
- Which approach wins: depends on recruiting brand; discipline in leveling matters more than title height.
- When Smart People Are Bad Employees
- Intelligence is necessary but not sufficient; companies also need reliability, effort, and team behavior.
- The Heretic: smart employee uses company flaws to prove it’s hopeless; hard to reverse after going public.
- The Flake: brilliant but unreliable; Opsware’s bipolar/cocaine genius stayed unproductive until firing.
- The Jerk: consistent asinine behavior cripples communication and is most damaging at executive levels.
- Lesson: potential means nothing unless the employee can be depended on as part of the team.
- The Right Kind of Ambition
- Tough Calls and Company Culture (Chapter 6: Concerning the Going Concern · III)
- Brilliant Jerks Break Communication
- Attacks silence issues: if a brilliant exec attacks anyone raising problems, nobody brings up topics in her presence.
- Bite only works from a big dog: mediocre jerks are ignored; brilliant jerks poison executive communication.
- Hold the bus for one exceptional player: Madden’s rule: bus leaves on time; only a rare talent justifies delay.
- One Dennis Rodman per team: Jackson’s limit; you may personally mitigate one star’s negatives, not multiple.
- Destructive big dogs go to the pound: when a star destroys staff communication, removal is the remedy.
- Senior Hires Are a Race Against Time
- Hire for time: startups race; experience compresses the journey from start to success.
- Acquire specific knowledge: hire senior people for know-how, not for “adult supervision” or company legitimacy.
- Internal vs external knowledge test: engineering managers need inside code/team knowledge; sales leaders need customer/world knowledge.
- Beware the PED effect: senior hires can produce incredible heights or internal degeneration if mismanaged.
- Managing New Senior Executives
- Three senior-hire risks: imported culture, political navigation skills, and your ignorance of their job.
- Demand cultural compliance: their culture is not your culture; adopt new ideas explicitly, never drift.
- Forbid political tactics: watch for big-company maneuvering and don’t tolerate it.
- Set a high performance bar: don’t lower standards because you can’t do the job; judge by world-class output.
- Measure four dimensions: results, management, innovation, and working with peers—not just goal achievement.
- Learn standards from greats: interview top performers to calibrate what world-class looks like.
- One-on-Ones Are Employee Meetings
- Communication architecture: CEO designs how information flows; absent design, company stagnates.
- Employee owns the meeting: agenda set by employee; can cancel if nothing pressing; manager listens 90%, talks 10%.
- Draw out issues deliberately: especially for introverts; status reports, email, and Asana can’t carry hard personal or half-formed topics.
- Powerful prompts: ask “What’s the number-one problem?” “What’s not fun?” “If you were me, what would you change?”
- Time-tested channel: one-on-ones are not magic; better designs are welcome if they still surface big issues.
- Programming Your Culture
- Startup imperatives: build a product 10x better and take the market before rivals do.
- Culture is secondary: bankrupt firms have world-class cultures; culture only matters after achieving those two goals.
- Culture preserves and attracts: designed way of working differentiates you, keeps operating values alive, and filters for mission fit.
- Built to Last gets it backwards: cultlike appearance is a retrospective result, not the design goal.
- Creating Behavioral Mechanisms
- Design small, shocking points: one mechanism should trigger daily behavior change, not just weirdness.
- Use shock value: Godfather’s horse-head prank shows disturbing symbols get action.
- Door desks at Amazon: Bezos embedded frugality with cheap doors and nailed legs, no long auditing campaign.
- Let culture evolve: most of what looks cultural grows from founders’ behavior; only a few design points are installed deliberately.
- Brilliant Jerks Break Communication
- Shock Therapy, Culture, and Scaling (Chapter 6: Concerning the Going Concern · IV)
- Shock Therapy as Culture
- Door desks at Amazon: employees internalize frugality through a physical, daily reminder of the mission.
- Ten-dollar-per-minute fine: Andreessen Horowitz enforced punctuality to prove entrepreneurs matter more than VCs.
- Move fast and break things: Zuckerberg's shocking motto forces reflection—innovation demands accepting short-term breakage.
- Mechanism must match values: Square's design culture would never use door desks; shock only works if it fits your ethos.
- Perks Are Not Culture
- Yoga and dogs are perks: they improve satisfaction but don't establish a core value that drives the business.
- Culture connects to mission: true culture is specific to what the company aims to achieve and promotes it in perpetuity.
- Culture's purpose: design it to get the company to do what you want, for a very long time.
- The Black Art of Scaling
- Scale is inevitable: building an important company means learning the black art of scaling a human organization.
- Give ground grudgingly: like an offensive lineman, grow by losing ground slowly, not by holding firm or collapsing.
- Specialization first: when onboarding new engineers costs more than doing the work, dedicate teams to build, test, operate.
- Organizational design: all designs are bad; choose the least evil by optimizing the most important communication paths.
- Process is communication: formal processes replace ad hoc talk when a company outgrows point-to-point contact.
- Process design: start with the output, engineer accountability, and formalize early—old people resist new processes.
- The Scale Anticipation Fallacy
- Don't prejudge executives: evaluating people against theoretical future needs is counterproductive and unfair.
- Scale is learned: nobody is born managing thousands; advance judgment retards development.
- Fit over rank: Zuckerberg for Facebook, Ellison for Oracle—greatness is specific to company and time.
- Judge relative, not absolute: ask whether a hireable executive would outperform the current one today.
- Evaluate holistically quarterly: separating scale from performance leads to swapping good executives for worse ones.
- Shock Therapy as Culture
- Culture, Politics, and the Right Ambition (Chapter 6: Concerning the Going Concern · I)
- Chapter 7: How to Lead Even When You Don’t Know Where You Are Going
- Courage, Psychology, and Crisis Leadership (Chapter 7: How to Lead Even When You Don’t Know Where You Are Going · I)
- The Opsware Penny-Stock Crisis
- Herb Allen: classy investor who backed Opsware when stock sank to $0.35.
- Three options: reverse split, accept penny-stock delisting, or hit the road to find buyers.
- Decision: Ben chose the road; reverse split felt like capitulation to the market.
- Allen's logic: he knew nothing about the industry but bet on determined founders.
- Outcome: Allen & Company buying lifted the stock from $0.35 to $3 in months.
- Managing CEO Psychology
- CEO psychology: the hardest CEO skill; harder than org design, metrics, hiring, firing.
- No training: you learn CEO by being CEO; the grading curve mean is 22 out of 100.
- No one to blame: founding CEO owns every hire, miss, and failure; no prior regime to blame.
- Two failure modes: taking problems too personally or not personally enough; both damage company.
- Lonely choices: Jason Rosenthal faced a cash crisis alone; no adviser had the answer.
- WFIO: every company hits multiple "we're fucked, it's over" moments; they feel worse than they are.
- Techniques to Steady Yourself
- Make friends: talk with people who survived similar ordeals; advice matters less than solidarity.
- Write it down: writing the logic externalizes the decision and calms the mind.
- Focus on road: at 200 mph, steer toward the road, not the wall; same for company.
- Don't punk out: great CEOs face sleepless nights; they never quit, they endure the torture.
- Personal toolkit: psychological fixes are individual; adapt techniques to your own mind.
- Fear, Courage, and Decision-Making
- Investor lens: Andreessen Horowitz looks for brilliance and courage in entrepreneurs.
- Courage > intelligence: the hardest decisions test courage, not smarts; right path often obvious.
- Co-CEO problem: dodging who's in charge burdens every employee with double approvals.
- 54% preference: you may barely favor one choice while experienced advisers oppose; courage still required.
- Hamlet: rejecting a lucrative offer against the board became his defining moment; team rallied.
- Fear and heroism: cowards and heroes feel the same fear; discipline determines action.
- The Opsware Penny-Stock Crisis
- Courage, Leadership, and Wartime Management (Chapter 7: How to Lead Even When You Don’t Know Where You Are Going · II)
- Courage Against the Crowd
- Social credit trap: the crowd's approval warps a 70-30 call into 51-49, making the crowd's path seem safer.
- Blame matrix: deciding against the crowd gets full blame if wrong; deciding with it minimizes blame but the company suffers.
- Hamlet's misread: the team's sell advice mirrored his ambivalence, not a deep analysis of the deal.
- Courage is trainable: he never felt brave, but practice taught him to ignore the fear.
- Courage compounds: each hard correct decision builds courage; each easy wrong one builds cowardice, for the CEO and the company.
- Excuse inflation: "close call," "team against me," and "best practice" dress up the easy wrong as unavoidable.
- Ones and Twos
- Two core skills: knowing what to do, and getting the company to do what you know.
- Ones: love broad information, strategy, and decisions; may underinvest in execution until chaos results.
- Twos: love clear goals, process, and execution; can overcomplicate decisions and stall necessary pivots.
- Hierarchy design: organizations exist for decision efficiency, so CEOs are usually Ones; Functional Ones decide locally and follow central direction.
- Succession traps: promoting a Two can slow decisions; leapfrogging a One can trigger executive turnover.
- No easy answer: internal candidates beat outsiders, but One-Two dynamics make CEO transitions inherently hard.
- Leadership: Vision, Ambition, Competence
- Leadership measure: quality, quantity, and diversity of people who choose to follow the leader.
- Steve Jobs attribute: articulate a compelling vision, especially when the company is weeks from death.
- Bill Campbell attribute: employees feel the CEO cares more about them than herself; they say "my company."
- Andy Grove attribute: earned competence makes followers trust he can lead them out of the jungle.
- Born or made: vision and competence can be learned; aligning ambition with employees' interests is nearly impossible to teach.
- Virtuous spiral: trust earns patience with vision; competence earns attention; vision earns leeway.
- Peacetime vs. Wartime CEO
- Peacetime: large advantage over a growing market; focus on expanding opportunity and strengths.
- Wartime: existential threat from competition or change; the company has one bullet and must hit the target.
- Style shift: peacetime empowers broad creativity; wartime demands strict adherence and precision.
- Contrast list: peacetime follows protocol and builds culture; wartime breaks protocol, is paranoid, and lets war define culture.
- Competitor mindset: peacetime sees distant ships; wartime sees intruders trying to kidnap the children.
- Mission focus: peacetime CEO aims to expand the market; wartime CEO aims to win it.
- Courage Against the Crowd
- The Unnatural Art of CEO Leadership (Chapter 7: How to Lead Even When You Don’t Know Where You Are Going · III)
- Peacetime vs. Wartime Leadership
- Peacetime CEO: tolerates deviations, minimizes conflict, seeks broad buy-in, sets big, hairy, audacious goals.
- Wartime CEO: intolerant, heightens contradictions, skips consensus, trains employees to survive the fight.
- Wartime CEO has no categories: with no #1 or #2 businesses, peacetime exit rules don't apply.
- Context dictates style: Jobs left in peace, returned in war; Chambers struggled; Google needed war.
- Mastering both is hard but possible: know management rules and when to follow or violate them.
- Management books teach peacetime: consultants rarely know war; Andy Grove is the rare wartime exception.
- CEO Is an Unnatural Job
- CEOs are made, not born: unnatural motions, like boxing footwork, need practice until they feel natural.
- Natural desire to be liked is dangerous: good CEOs upset people short-term to be respected long-term.
- Constant feedback is the atomic building block: without it, reviews, comp, politics, and firing go poorly.
- Feeling like a fraud is normal: every founder CEO has felt awkward; that is how CEOs get made.
- Giving Feedback Effectively
- Shit Sandwich: compliments bracket criticism; works for juniors, but senior staff see through The One Minute Manager trick.
- Come from the right place: authentically want her success; make her feel you're in her corner.
- Don't get personal or public: prepare people to succeed; never embarrass them in front of peers.
- Tailor tone and be direct: match the employee's personality; "I couldn't follow it" beats watered-down deception.
- Feedback is a dialogue: employees may know more; invite challenge and argue the point to conclusion.
- High-frequency feedback: have an opinion on everything; it depersonalizes critique and lets bad news travel fast.
- Evaluating CEOs: Does She Know What to Do?
- Evaluate CEOs by three questions: know what to do, get company to do it, deliver against right objectives.
- Strategy becomes the story: CEO keeps a clear, compelling why that unites employees, partners, investors, press.
- Story is not a mission statement: it can take many pages, as Bezos's 1997 letter did for Amazon.
- CEO output is decisions: measured by speed and quality on incomplete information and high stakes.
- Courage is decisive: the best choices are often deeply unpopular; selling Loudcloud to EDS was right.
- Gather intelligence constantly: embed knowledge quest into staff, customer, and one-on-one meetings.
- Evaluating CEOs: Execution and Results
- Execution needs leadership plus operations: capacity, motivation, clear context, and minimal politics.
- CEO owns team quality: hiring, screening, and matching talent to the company's current challenges.
- Test contribution ease: individuals should get work done without fighting politics or broken processes; Reed Hastings designed Netflix for this.
- Set objectives correctly: avoid board-managed low bars or unrealistically high expectations.
- Compare against opportunity: hardware won't be capital-light like internet; Yelp can't grow like Twitter.
- Results are lagging: white-box questions predict future better; share the test so CEOs perform better.
- Peacetime vs. Wartime Leadership
- Courage, Psychology, and Crisis Leadership (Chapter 7: How to Lead Even When You Don’t Know Where You Are Going · I)
- Chapter 8: First Rule of Entrepreneurship: There Are No Rules
- Chaos, Accountability, and Empathy (Chapter 8: First Rule of Entrepreneurship: There Are No Rules · I)
- When the Sky Turns Purple
- No rules: the first rule of entrepreneurship is that there are no rules; expect the unexpected.
- Sky is purple: when business assumptions flip overnight, don't argue the sky is blue—adapt and move on.
- E&Y betrayal: the auditor's national office demanded restatement or contract amendments, nearly killing Opsware's $1.6B sale.
- CA clause: a legacy contract ambiguity let E&Y impose a new accounting interpretation after years of approval.
- Saving the Opsware Deal
- All-night scramble: used board connections, sales reps, and legal rewrites to get three banks to amend contracts within 24 hours.
- Buyer fallout: BMC withdrew; HP lowered its bid to $13.75 because of the deal's "taint."
- Gold-standard stance: Ben refused any price below $14.25, arguing the deal rested on Opsware being the premium company.
- Board tension: the board advised accepting the lower offer, but Ben held firm and HP finally agreed.
- Accountability vs. Creativity Paradox
- Paradox: punishing missed commitments discourages risk; ignoring misses makes reliable performers feel like chumps.
- Core assumption: believe employees are creative and motivated unless proven otherwise; cynicism kills innovation.
- Effort: world-class effort is nonnegotiable; people who don't give it must be checked.
- Promises: hold people to commitments, but account for the difficulty of hard engineering promises.
- Results: judge misses by seniority, degree of difficulty, and whether the risk was stupid or excellent.
- Resolving the Paradox
- Seniority: experienced people must forecast accurately; junior misses deserve teaching moments.
- Difficulty: if scaling was a miracle, thank the engineer; if the project was trivial, address the slip.
- Risk quality: if the risk was right and the product would have failed, don't second-guess the choice.
- Creative magic: in tech, the line between mediocre and magical is letting people take wise creative risks.
- Freaky Friday Management Technique
- Warring teams: Customer Support and Sales Engineering escalated complaints despite both having strong people and managers.
- Role swap: have hostile teams switch places to experience each other's daily pressures firsthand.
- Film lesson: Freaky Friday shows that living another person's challenges builds understanding and defuses conflict.
- When the Sky Turns Purple
- Unconventional Management, Standards, and Selling (Chapter 8: First Rule of Entrepreneurship: There Are No Rules · II)
- Freaky Friday Management
- Role swap: Permanently switch leaders of warring teams to break entrenched conflict.
- Fresh eyes: Each executive diagnosed core issues within one week of walking in the other's shoes.
- Simple fixes: New processes cleared up combat; the two teams became the company's best collaborators.
- Freaky Friday: A surprisingly insightful management model for forcing empathy and structural change.
- Keeping Executives World-Class
- Hiring bias: CEOs assume new hires are world-class; time, scale, and markets erode that fast.
- Standards are dynamic: Raise the bar as you learn what the company actually needs.
- What you do matters: Real culture judges performance, not reputation.
- CEO can't develop reports: Executives must be 99 percent ready; training happens elsewhere.
- Evaluate today: Judge current performance, not a hypothetical job two years away.
- Expectations and Loyalty
- State the future plainly: Tell executives their job will change as the company doubles in size.
- New-job trap: Executives fail by doing their old job instead of mastering the new one.
- Loyalty goes downward: Employees doing the work deserve world-class management, even if it means replacing old team members.
- Should You Sell Your Company?
- Logical rule: Stay stand-alone if you're early in a huge market and can be number one.
- Market definition: Google was in search, not portals; misjudging the market changes the decision.
- Acquisition types: Talent/tech deals run $5–50M; product deals $25–250M; business deals value whole operations.
- Opsware case: Sold when the market was redefined, virtualization shifted R&D, and offers signaled a local maxima.
- Muting the Emotion
- Pay the CEO at market: Once the company is a real business, avoid sale decisions driven by personal finances.
- Be clear with employees: Describe the independence logic so investors' and employees' interests align.
- Prepare both sides: No easy answer exists, but intellectual and emotional readiness makes the call manageable.
- Freaky Friday Management
- Chaos, Accountability, and Empathy (Chapter 8: First Rule of Entrepreneurship: There Are No Rules · I)
- Chapter 9: The End of the Beginning
- From CEO to Investor: The Origin
- Catalyst: after Opsware and HP, Ben saw entrepreneurship as a black art — nobody wrote down the hard lessons.
- Flashpoint: a top investor asked, “When are you going to get a real CEO?” — a wound that became the firm’s mission.
- Founding motto: “some experience required” — VCs should have founded and run companies, not just funded them.
- Breaking the VC Catch-22
- Concentration: of 800+ VC firms, only about six delivered great returns.
- Self-reinforcing loop: best entrepreneurs choose best VCs, so incumbents’ track records make newcomers unbeatable.
- Opening: a real entrepreneur community now exists, so word-of-mouth can build a new VC brand.
- A Firm Built for Founder CEOs
- Core belief: technical founders are the best CEOs; enduring tech giants were all founder-run.
- Two gaps: founder CEOs lack the professional CEO skill set and the CEO-grade network.
- Mentorship over classrooms: you learn CEO-ing by being CEO; general partners must accelerate that pain.
- Alternative path: when founders shouldn’t be CEO, help them pick and integrate a professional CEO.
- The CAA Model: Network as Infrastructure
- Ovitz’s insight: an integrated firm-wide network beats isolated agents; CAA shifted Hollywood’s power to talent.
- Applied networks: large companies, executives, engineers, press and analysts, investors and acquirers.
- Name and brand: “a16z” honored Marc’s reputation because the firm itself had no track record.
- Counterprogramming: while old banks avoided PR, a16z launched with fanfare — a 2009 Fortune cover.
- Building the Team and the Final Lesson
- Hiring for strength: smart is not enough; hire people great where you need greatness and aligned with the mission.
- Scott Kupor: Opsware’s frustrated finance director found his dream job as COO — doing strategy, operations, and deals.
- Perception lag: Ben was a strong CEO by 2003–2007, yet the press called Opsware dead until the HP sale proved otherwise.
- Freedom of honesty: CEOs must show unrelenting confidence; VCs can finally speak the hard truth.
- Embrace the struggle: hard things are hard because emotion fights logic; your weird background holds the keys.
- From CEO to Investor: The Origin
- Appendix: Questions for Head of Enterprise Sales Force
- Strategic Acumen
- Smart enough: assess her ability to pitch her current company and grasp your opportunity.
- Strategic contribution: will she meaningfully shape company direction, not just execute?
- Industry wiring: knows competition, active deals, and has mapped your organization.
- Diagnostic speed: can quickly size up the market, international expansion, and your business.
- Hiring and Team Building
- Hiring model: knows her rep profile, recruiting time, and interview tests for talent.
- Bad hires: ask her to describe a recent one and what it taught her.
- Validating references: does her current team want to follow her, and can they be referenced?
- Sales managers: can she define the job, test for skills, and hire managers systematically?
- Sales Process and Training
- Process ownership: distinguish those who write the game plan from those who follow it.
- Technical sales fluency: understands benchmarking, lockout documents, proof of concepts, demos.
- Enforcement and CRM: has clear expectations for process compliance and CRM use.
- Training program: can describe the balance of product versus process training in detail.
- Rep Evaluation and Deal-Making
- Evaluation model: moves beyond basic performance to transactional versus enterprise rep differences.
- Comp design: understands accelerators, spiffs, and plan mechanics.
- Big-deal skills: has enlarged or accelerated large deals with customer references.
- Marketing and channels: articulates brand, lead generation, enablement, and channel conflict.
- Operational Excellence: People and Decisions
- Managing direct reports: knows what she looks for, how she trains, and how she evaluates.
- Decision-making: uses systematic information gathering and disciplined staff meetings with action tracking.
- Core processes: can explain design of interviewing, performance management, integration, and planning.
- Confrontation: handles territory requests, promotion, firing, and chronic bad behavior directly.
- Metrics, Org Design, and Character
- Metric pairing: leading and lagging indicators matched to avoid valuing speed without quality.
- Org design: knows why chosen strengths matter and how conflicts get resolved.
- Systematic thinker: works from systems rather than one-off fixes.
- Honesty and homework: direct, articulate, asks incisive questions, and has prepared.
- Strategic Acumen
- Acknowledgments
- Not distilled: acknowledgments are front matter and carry no core ideas to map.
- Introduction
- Core Conclusion and Practical Takeaways
- The Central Truth
- Hard things are hard: no formula exists for crisis leadership—only experience, story, and clues
- The Struggle is normal: self-doubt and isolation are not failure but the forge where greatness comes
- No silver bullets: fundamental competitive gaps demand lead bullets—hard, direct work
- Peacetime vs. wartime: the same company needs opposite leadership styles depending on existential threat
- Nobody cares: excuses preserve no dollars or jobs; waste no energy on what could have been
- Mindset Shifts
- Believe in calculus, not statistics: play the move, not the odds—the task is identical at one-in-a-thousand
- Tell it like it is: transparency buys trust and brains; sunny optimism makes employees doubt your grip
- Courage over intelligence: hard decisions test discipline, not smarts; the right path is often obvious
- Stop self-flagellation: you made the mistakes, but the CEO must absorb the struggle and keep moving
- Play long enough: technology shifts constantly; surviving eventually brings luck and answers
- Managing People
- Hire for strength, not lack of weakness: run the job yourself so you know the strengths needed
- Train relentlessly: training is the boss's job; twelve hours of teaching can gain two hundred hours of output
- One-on-ones are employee meetings: they set the agenda, you listen 90%; ask the number-one problem
- Tell every employee where they stand: Bill Campbell's rule—immediately, not a minute later
- Pay management debt: fix two-in-a-box roles, counteroffer leaks, and missing feedback before interest compounds
- Leading Through Crisis
- Get your head right: layoffs stem from the company failing its plan, never from employee failure
- Don't delay: once word leaks, managers and trust are damaged beyond repair
- Managers fire their own people: never outsource the hardest conversation; stay visible afterward
- Put problems where they can be solved: engineers who write code should own product worries
- Focus on the road, not the wall: at 200 mph you steer toward where you want to go
- Building the Company
- People, products, profits—in that order: goodness matters most precisely when things go wrong
- Culture is designed mechanisms: door desks or punctuality fines—shock points that change daily behavior
- Process replaces ad hoc communication: start from the output, engineer accountability, formalize early
- Design for the least evil: all org designs are bad; optimize the most important communication paths
- Resolve the accountability paradox: weigh seniority, difficulty, and risk quality before punishing a miss
- The Central Truth
opening map…