If You Hate Your Job, This is How to Start Over - Bill Gurley
In a Nutshell
Bill Gurley argues that 60-70% of people regret their career choices due to an education "conveyor belt" pushing safe jobs without passion, leading to burnout and open-loop "boldness regrets" from inaction—use Jeff Bezos' regret minimization framework to pivot early. Success stories of late bloomers (e.g., Tito's founder at 40+), financial flexibility, dream job folders, peer learning groups, and loving the grind reveal paths forward; build bravery via side hustles, battle cards for scenarios, and continuous learning, especially with AI as a jetpack for proactive crafters. Life speeds by—prioritize purpose over grinding unloved jobs, as stories stick more than advice, and pivots amplify latent potential for fulfillment and impact.
These notes were generated by AI and may contain inaccuracies.
Bill Gurley spent 25 years as a venture capitalist and 4 years before that as a sellside analyst on Wall Street. He started writing to differentiate himself as an early blogger, jotting down ideas, developing some into blog posts. While reading biographies, he noticed a through line after the third one across wildly different fields, jotted notes, and let it simmer. Got asked by the dean of the University of Texas business school in Austin to talk to the MBA class, developed it into a PowerPoint presentation posted on YouTube. James Clear noticed, posted it on his website, prodded him to develop it as a book. A few years ago, began retirement as a VC, drawn to this idea over VC book, investment book, or Uber tell-all. Felt more authentic, bigger impact. Spent six years with co-writer and researcher developing the book.
Launched SurveyMonkey survey: if you could go back and start over, would you do a different career? Seven out of 10 said yes. Took it to Wharton people analytics for scientific version with broader audience: six out of 10 said yes.
Talked to Daniel Pink about regret; in his book, biggest regrets toward end of life are regrets of inaction, boldness regrets—what you didn't do. Humans forgive mistakes (learn, won't do again) but ruminate on what they didn't try.
Current education path is a conveyor belt pushing children into a meat grinder toward safe jobs (pre-AI). Teaches grinding and perseverance (Angela Duckworth's grit), but without love, it turns into burnout. Book's purpose: give people permission to do what they want. If subgroup reads it, gains conviction to succeed in what they love, huge impact. Those who love what they do are more successful, radiate positive energy, become charismatic role models.
Life is a use it or lose it proposition.
Brains ruminate, replay things, build what-if stories (positive ones), compare to current life, creates anxiety. Zeigarnik effect: open-loop closed-loop bias. Study on waiting staff: unbelievable recall while tables open (no pen/pad/iPad), nothing after closing. Hotel room example: remember current room, forget two hotels ago. Cliffhangers work because human mind abhors open loops. Boldness regret leaves an open loop. Mind abhors uncertainty/ambiguity, prefers imagined catastrophe for certainty (anxiety). Regret minimization framework: Jeff Bezos at DE Shaw, imagined 80-year-old self for advice on starting Amazon. Gurley used it twice before VC (dream job): after couple years in jobs, reflected "do I want to do this for 30 years?" Unhappy reflection prompted next step.
Regret minimization akin to Stephen Covey's "begin with the end in mind" from Seven Habits of Highly Effective People.
Conveyor belt (Jonathan Haidt: resume arms race): intense pipeline makes first job feel like sunk investment, tweaking feels like throwing it away (loss aversion). Studies: 5 years after college, 40% no longer in major field; higher at 10 years, but feeling trapped may suppress changes. Young people feel obligation to decisions locked in at 17. Colleges shifted major declaration from end of sophomore year to end of junior year in high school (3 years earlier, less exploration time). Smart people say not enough time for kids to explore (Haidt chapter on loss of play; Rick Rubin). Jean Twenge's Generations: extended adolescence—kids forced into adult decisions at 17 (career tracks for 30 years), yet move out later, driver's licenses later, employment later, relationships lower rates, less alcohol/casual sex. Socially immature, professionally expedited.
Angela Duckworth 10 years after Grit: wished positioned as 50/50 passion/perseverance, overemphasized perseverance teaching generation to grind without passion.
Perseverance seductive because passion harder to engineer, feels egalitarian: "hold hard enough, get through." Finding passion hardest; many 17-year-olds truthfully say "I don't know."
Life moves faster than realized, then done. Bezos regret minimization holds accountable. Get locked in sooner by spending to limits (freeze in job). Wall Street example: high salaries but Hamptons place, summer lease, membership club—high burn rate requires constant fuel. Advice to young adults with decent salary: don't spend against it for flexibility to move cities, change jobs.
Flexibility hard to show off (not like plane/holiday photos). Trade hidden metrics (sleep quality, Saturday peace) for observable (pay packet, position). Trades everywhere.
Career fulfillment not luxury; respect those thriving in non-occupation (religious community, volunteer), might even be career path. Book for unhappy at work (impacting life) or those with inkling but lacking permission/confidence.
Steve Harvey story (on Oprah?): Age 7, teacher exercise to write what want to be; called last, made to read aloud (had stutter): "I want to be on TV." Teacher condemned, called mother, got in trouble, sent to room fearing dad's riot act. Dad sat, heard story, said put paper in bedstand, read every day.
Hard for parents (Gurley has 3 kids): DNA to prioritize economic welfare, associate with money. Matthew McConaughey: switched law school to film school, dad said "don't half-ass it"—gave push.
Book unlocks latent potential for those with inkling world says incapable/shouldn't do. Observable: financially free; hidden: passionate/happy at 35.
Financial: overspending or hand-to-mouth. Jen Atkin profile: most successful hair stylist ever, launched beauty products, sold company; moved to LA with $300. Mail room starts: David Geffen, Barry Diller in Hollywood.
Family pressure: McConaughey thought dad wanted lawyer (actually supporting stated goal). Cultures: Russia academics, family traditions (doctors).
Fear: "what if doesn't work?" Book embeds stories/narrative (Morgan Housel on power; structured interleaving every other chapter Atlantic-size article about someone who started
The book is divided into two parts that interleave: every other chapter is an Atlantic-size article about someone who started on the bottom rung with intention and made it to the top. These profiles stick in the brain more than textbook-style career books that say do A, do B, do C, lacking spirituality. > Facts don't care about your feelings is as backward as it could be; feelings don't give a single fuck about the facts.
There are eight, nine, or 10 profiles alternated with principles, like candy and vegetables, bringing ideas to life. All profiles are from fields parents would tell you not to go into, intentionally skipping investment bankers and doctors with nice linear progressions, to help people get past fear by providing motivation and method.
Frameworks and justifications don't stick; stories do. Optimizing for memorability: the human brain retains stories better than well-written insights. Aphorisms are like the lead melody of a song. Long-form non-fiction articles from Long Reads, Atlantic, New York magazine (e.g., 1993-94 cover with Suge Knight, Snoop, Biggie, 20 pages) led to books like The New Journalism (Tom Wolfe, Truman Capote) and The New New Journalism (Krakauer, Lewis). This pivots to Substack, where journalists like Barry Weiss (now at CBS) and Matt Taibbi (Twitter files) built careers.
Risks include sunk cost fallacy, starting at the bottom again at 37, what people will think.
Advice:
- Keep a dream job folder (Google Doc): notes on what you'd do one day, people to talk to, making it feel real before leaping (from Shark Proof by Harvey Mackay).
- If curiosity occupies downtime, it's a tell (e.g., as engineer at Compaq Computers in Houston, traded stocks after reading Liar's Poker, using stock analysis books).
- Chapter Never Too Late: lists ~20 post-40 pivots, deep on four. Favorite: Bert Beverage (Tito), seismology degree, oil/gas in South America (quit due to fear, boom-bust), mortgage broker. Watched PBS special: draw line down paper, loves on left (hanging at bars), good at on right (chemistry), started spirits company. Texas had no licenses then; Tito's now bestselling spirit in North America, 100% owned, bootstrapped (19 credit cards).
Now gives back via foundations. Profiles show people touch more lives than imagined, serving leveraged way (e.g., Danny Meyer, restaurant tour, built Shake Shack, Setting the Table; his restaurants meaningful on resumes, like a church; ethos of reading the room, understated hospitality, hyper curious, excessive learner—took chefs to Europe for notes).
I don't want to chase my passion. I want to be purpose driven. (Response: noble but not always luxury with high burn rate/family; some jobs suck; climbers from bottom touch hundreds of lives.)
Continuous learning (second principle): best passion test is if learning feels free, like you'd do it over TV.
Joe Hudson: > Enjoyment is efficiency.
Discipline has friction; motivation reduces it (playlist/caffeine); obsession inverts friction—pulls you. If pointed right, looks superhuman from outside, feels unchosen inside (e.g., podcast studio in every house/Airbnb for 10 years, 1000 episodes).
Single ordinating principle: Bezos (customer experience); Musk (closer to Mars). Both grew innovative companies to hundreds of thousands employees without bureaucracy (rare; Elon deep in details via heuristics).
Test: Do you see yourself doing this 30 years from now? If no, > get busy living or get busy dying—start planning now, don't quit tomorrow.
- Dave Evans exercise: create 3-5 scenarios, battle card them (AI helps: 20-pager per path—what like/not like, 3 things people love/hate; ask people in fields, role-play).
- Side hustles: ask employer permission for own-time projects (Ben Gilbert/Acquired: at Microsoft created Garage for founder relevance; at Madrona VC started podcast—both happy).
- Spare time signals (shower thoughts, e.g., pickleball obsession).
Don't quit without knowing where going.
Story: Real estate lawyer loved football offensive plays/diagrams; launched website/software/podcast, now 900k followers, talks plays with experts—known in industry. Best stories in parental-no-go jobs. Support jobs abound around passions (Hollywood/music).
You can fail at hated job/underperform unwanted industry; better odds doing okay at loved one. Life is use-it-or-lose-it; humans forgive failures well if proud of bravery.
Psychology: people happier after decisions (e.g., moving cities)—novelty/variety fun, even if objectively worse.
Example: 25 years Bay Area to Austin high-rise—invigorating (new coffee, walks).
Plateau vs. bored: 30-year test; peer relationships (outside org, same path/different company) for split-testing (shitty boss? talent lack? transient?).
Deliberately upgrade without transactional feel: avoid climber/sharp-elbowed/zero-sum finite games mindset. Careers have many winners (podcasting, Austin).
Great test: trust to share best ideas learned.
Story: Chris Delcante, athletic director at University of Texas, when he broke into sports administration, he was in the
A man in the development office went to a conference, met peers at the beginning of their careers, exchanged numbers, started a text group that grew to eight people. All eight are now D1 athletic directors. They started at the bottom, shared problems via text, elevated learning each year. They did weekends away with wives, hosted their own conferences, brought external guests to educate and help them rise together.
MrBeast at 17 found three others hacking YouTube on 16-hour daily Skype calls. He said if there were a fifth person, they would have made a million dollars too because the insights were usable and deterministic, unknown to others. > "We got 40,000 hours because we were sharing it all."
Peers provide co-learning, support, reflections, vulnerability. Mentors serve different purposes. Do both.
Idyllic mentor view is broken; young people cold call too high with 98% rejection. Divide into aspirational mentors (study like Star Wars characters via YouTube, podcasts, AI) and guidance mentors (go two levels below; they're thrilled to give advice, 10x hit rate).
Flattery drives podcast invites. Trick: ask "put me in touch with whoever handles your podcast adverts." Flatters if it's them; delegates if not. For mentors: "put me in touch with whoever handles Bill's inquiries for coffee." Pay attention to delegates; learn through them, be generous. David Goggins story: ex-Navy SEAL, endurance athlete, rare podcast guest (only Rogan and this show in six years). After 9 months coordinating with team member, nearly canceled, but persisted; Goggins credited her as the reason he appeared.
Loving the grind is non-negotiable; can't learn it, it's innate. Relates to flow state where it's not work. Some wired for it; others fake it via meat grinder. Met top robotics honors students (Eagle Scout, 1580 SAT) programmed to grind, but not path to greatness. Brute forcing creativity via white-knuckling miserable successes provides activation energy but not sustainable.
Some have winning as meta-passion (Michael Jordan invented rivalries/slights, took shots in Hall of Fame speech). Novak Djokovic: loves hitting the ball. Counter: Shaq's grateful All-Star monologue on Inside the NBA, thanking influencers, making Ernie cry. Pro-social, thankful energy better than bitterness.
Sports need sharp elbows; most careers (e.g., author) don't. Always give back: appreciate helpers when advancing rungs; builds process, network, supporters, weathers bad times.
College grinders burn out, want break from studying. Best like Danny Meyer study constantly. Fields like medicine require it; ballet 16-hour days applauded, engineering not. Continuous spare-time learning tests obsession.
Elon Musk's X post: seeking people for hardest problems, unrelenting pace. Critics called abusive; some see dream (18 hours on stimulants/Chick-fil-A, smartest people). Young AI founders embrace China's 996 (9am-9pm, 6 days/week); say "Silicon Valley got lazy." SF offices empty, but AI kids want it. Not illegal; workload legalization like drugs. Compete with those mindset; do what you love or lose to passionate learner.
Grinders in unloved certified paths (accountant/engineer) fear AI crushing jobs. Proactive craft-builders/learners see AI as jetpack to do more faster.
Guidance doesn't sculpt us into something new. It exaggerates who we were already. The patent is almost cruel. The ones who least need the medicine are the ones most likely to overdose on it, while the ones who need it desperately are immune.
Advice amplifies predispositions: "don't be pushy" makes conscientious guys timid, boundary-pushers ignore. "Work harder" feeds insecure overachievers, lazy coast. "Take responsibility" burdens self-blamers. Filters via traits, flatters self-image, confirms fears. Advice hyper-responders absorb matching inner fears.
Develop grind capacity, but without passion, drive fast wrong direction to unfulfilling wins. At true retirement (speaker turning 60, loved VC but passing to next generation), reflective.
It's hard at 25 to see the end, but at 18 you don't see the end times, giving superpowers to play. Realizing there's an end to your career motivates frontloading fear for a different at-bat. Regret minimization exercises help. George, the housemate, does one morning a month: wakes up, lies in bed, imagines having no arms or legs for 30 minutes to feel gratitude, then reborn into this body with arms and legs.
Emotional work gets below the neck. Smart people do expected value calculations on success likelihood and utility of enjoyment, but hidden unobservable problems persist. Strategies like George's or regret minimization overload emotion, not information.
Tony Robbins workbook Unleash the Giant Within (1994 Audible, 1.5 hours, hard to find publicly): Imagine what it's cost you in the past, costing now, will cost in future. Overload the pain: rumination, grief, wistfulness. How it's affecting you now, limiting time, stopping happiness. Future: life if nothing changes for 1-3 decades—relationships, mind quality, health. Then imagine life if you changed. Oscillate between pain and pleasure.
Moved from UK at 32-33, successful career running nightclubs, one of biggest events companies in north England, dedicated 15 years. Post-COVID, clubs reopened but didn't want to go back, wanted to do podcast. Best single decision ever made, zero regrets.
Morgan's blog post on nostalgia: can't see past pains/fears well after moving on. He and wife reminisced New York 20s pre-kids: wake up Saturday, lie in bed, coffee, walk around, no responsibilities. Retrospectively, fears were wasted; sees how well life went.
Golden eras only exist in memory. Journals show fears: what if it goes wrong, laughed at, live in parents' spare room. Worst scenario unlikely; gradations between best and worst. Even obvious decisions need galactic pain/pleasure to galvanize activation energy.
Battle card notion: fully imagine other reality without comparison, every element on its own, then current one, battle them. Good at bringing future forward in brain, reverse time travel, life NPV. Made regret minimization work; useful as venture capitalist.
Key traits:
- Product instincts: Critical, especially new waves like AI where product surfaces less understood (e.g., mobile movement). Need to be ahead on human interfaces.
- Salesmanship: Sell investors, employees, culture, deals, customers. Chief salesperson.
- Determinism/grit: Top 1 percentile, chips on shoulders put chips in pockets (Josh Wolf, Lux Capital). Failed before, point to prove. Afraid of how determined they are.
Bezos: Only looks for determinism—is this person going to do this no matter what? All questions probe that: what if this, past approaches, why, alternatives.
Champions lack off switch normal people have. In zero-sum competitive founder world, go unreasonable lengths. Strong ethics needed to avoid crossing line.
Lindsay Vonn, 41-42 downhill skier in Olympics: retired, came back, competitive trials, injured ACL in Italy, braced up, competed anyway, accident, 3 surgeries. Microcosm of open loop: last shot at 40s, sending it. Better to burn out than fade away.
Founder more important than company. Bet on founder finding thing. Proof: original idea fails, pivot succeeds—Slack (game failed, internal tool won); Discord (game failed, Skype alternative for gamers won). Pivots famous in VC.
New CEO 50/50 bet at best; avoid risk, stick with founder. Founder mythology high in Silicon Valley. B2B companies higher success replacing founder than consumer (consumers fickle, product artistic like movie vs. systematic B2B; Steve Jobs: buyer ≠ user in business).
Beverage company (Newtonic): first big one not directly operated/founded. Raised 3.4M last year, raising 6M now. Co-founder Luke drove 5.5 hours Portsmouth to Leeds for one investor meeting—obsession, determinism.
Fundraising reveals perceptions. Friend replied to deck with Midwit meme: extremes (Neanderthal, sage) say "I bet on Chris"; middle overcomplicates with EV calc. People bet you'll make it work.
Anti-deck founders exist; wrote blog "In Defense of the Deck"—shows Steve Jobs, Mark Benioff, Jeff Bezos using decks. VCs limited boards (2/year), limited shots, must fall in love emotionally, not just positive IRR like public markets. Need to love category, idea, want 10 years on problem—flip zero to one.
Harder for unsexy businesses. Mid-market PE does those. Businesses shouldn't take VC: Series A→B→C dilutes ownership dramatically; exit value needed 10x higher for same money (e.g., 100% sweat equity vs. diluted).
More $20M tuck-in acquirers than $500M; large public cos do $20M quietly, boss doesn't care—lifetime wealth at 100%. Harder to sell big: fewer buyers, scrutiny, audited financials, S-8 filing.
Think market of company it sells to, not selling market. Sell first $20M (90% owned), then take big VC shot with freedom.
Liquidation preference: VCs preferred stock paid first like debt in sale. Big rounds (e.g., $500M AI) common stock doesn't participate until after. In first raise, bought back in, gained share—investors love it.
Gymshark: 2.6B pound company, entirely private, founder Ben (friend) still 75% owned. Tito's guy bootstrapped 100%.
Most excited: personal empowerment. Humans evolve with tools (plow→tractor→computer). Skeptics like Luddites with looms. Futureproof: run at AI edge in your industry. Most AI-productive human not fired—go-to person.
Hone your craft: study history (quote founding fathers), study tech edge—look like unicorn. Middle is where you die.
FOMO on new things: top 10 app store unknown triggers hives—allergic to ignorance. Hasn't done Claude yet, feels guilty. Mac Minis sold out.
You've got massive FOMO with AI. It's useful. If you're not leaning in, you're missing prompt engineering skills. Every day think of a new way to use it and test its capabilities. Being in it every day helps explore boundaries and learn new things.
Had a great experience with the concluding chapter in the book. Almost done wrapping up, feeling good. Editor says take a shot at the conclusion this weekend. Spent six to seven hours, sent it. Editor says really don't like this—first critical feedback. Asked ChatGPT pro version ($20-$30/month) to write a 20-page report on the best concluding chapters in non-fiction books of all time, with a summary of each and what made it special. Reading through, noticed eight or nine of the 10 took an orthogonal direction to the book—didn't summarize, brought in a new perspective. That gave an idea. Ideated on that concept, took it in a different direction. Editor loved it.
Concluding chapter titled It Ain't Easy. Whole book encourages you to go do this. This is a warning at the end: it ain't easy, you got to struggle, suffer, be okay with failure—in praise of the grind. Without AI, wouldn't have done that. AI didn't write it, just gave a new lens for thinking about it.
AI's impact on work: LLM (large language model) really good at text manipulation. Jobs involving searching text, summarizing text under threat.
- Paralegals already under threat. Lawyers have drastically reduced paralegal force.
- Coding is highly structured text, more structured than text itself with constraints—more freedom in language than coding. Coding under threat.
Best people generating most code in future will harness the tool like a farmer learned to use a tractor instead of plowing by hand. No hand hoers anymore. Text is a big deal.
Translation jobs dangerous. Someone needs to check translation's right—that's a different job: editor of translation, an opportunity. Move upstream, be best at exercising the tool. Old tool doesn't work.
If 21 and starting again with no knowledge or connections, definitely related to AI. Rolling around in it more, would have launched 10 Claude bots already.
Fascinated by agent idea for personal assistant like a running buddy.
Fascinated by building new CRM from scratch for small companies without CRM—easier to start with zero in the database. Can't get there from Salesforce due to huge database with fields and forms no one wants. Starting with that construct builds the wrong thing—backwards integrating something janky users don't want to see. Storage mechanism won't be visible like traditional enterprise database apps.
Range (second half impacted most)—antidote to Malcolm Gladwell (close friends, Gladwell quoted on cover). Known for generalist vs. specialist, comparing athletes. Second half: far analogies. People switching industries, careers, or academic focuses are biggest innovators. Enter through side door with different mental model than field natives. See things differently, bring patterns from elsewhere or see patterns others miss.
Touches on it in learning chapter and peer chapter: stretch peers to different industries, stretch learning to other industries for biggest nuggets. More noisy, harder, less aligned, but insights are power law step changes.
Four years ago, obsessed with cinematography after moving to America. Recreated Interstellar house on 85 ft video wall (biggest in Texas) using same technology as Star Wars. Rendered entire scene in Unreal Engine 5, set dressing with live elements: dirt on floor, cactuses, reversed air stream. Sat with McConn on 11th anniversary of Interstellar in front of the house. Started learning cinematography, got friends from industry, asked how to elevate a podcast—did cool things. Largely passion project.
Steve Jobs' famous Stanford graduation speech: if never took calligraphy class, doesn't know if builds Mac or iPhone right way.
Things borrowed from far away can be impactful, hard to know at the time. Calls it a superpower or advanced level—not easy because translating like an exchange rate. Taste, discernment needed: is this relevant?
Areas AI can't replicate: community, networking ability, taste, discernment.
AI can't build friendship or choose right guest. MrBeast approach uses quantifiable metrics (likelihood of big play guest)—local maxima everyone can optimize. Refining for taste is different.
Bill Gurley, you're awesome. Thanks for having me.
Friends said you'd love his energy—he's a legend occupying high-rises.
Check out everything on X profile (@bgurley)—historically spent most time there (early investor in Twitter). Instagram account developing for first time for the book.
Instagram's a different language from Twitter. Figuring it out.
Keep Chris Williamson in your library
Save the videos and channels worth coming back to, and find them again in one place.
