YT YouTube2Text

YouTube2Text → Transcripts → The Economics of Owning a Software Company

The Economics of Owning a Software Company — Transcript

by Cost Anatomy · 4,113 words · 631 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:00Okay, so you want to own a software
0:02company. $22.6 billion
0:06in cash for typing. This week, a French
0:09giant paid $22.6 billion for PTC, a
0:13company you've probably never heard of.
0:15What does it sell? Lines of code. That's
0:17it. Text on a screen. Now, here's the
0:20part that should scare you. You're
0:21thinking, "I could do that. Build an app
0:23once, sell it a million times. Every
0:25copy costs me zero." And you're right,
0:28you could. The margins are real. 90%
0:31money while you sleep. But here's what
0:33nobody tells you. For every PTC that
0:36sells for billions, 10,000 software
0:38companies die in silence. Not because
0:41their code was bad. Their code was fine.
0:43They died because of something that has
0:45nothing to do with code. Something the
0:47winners know and the losers never figure
0:49out. By the end of this video, you'll
0:51know exactly what it is. And you'll know
0:53whether your software company would
0:55survive or become one of the 10,000.
0:58Let's start with the dream, because the
1:00dream is what pulls everyone in. Picture
1:02it. You and two friends. A garage, a few
1:05laptops, six months of late nights and
1:08too much coffee. Then, you launch. And
1:10the math looks like magic. A car company
1:13has to build every single car. A
1:15restaurant has to cook every single
1:17meal. But you? You write the code once.
1:20Then, you sell it a million times. Copy
1:22number one costs you six months. Copy
1:25number 1 million costs you basically
1:27zero. That's not a business. That's a
1:29money printer. And the margins? A
1:32typical software company keeps 75 to 80
1:34cents of every dollar as gross profit.
1:37Compare that to a grocery store begging
1:38for two cents on the dollar. It's not
1:40even the same sport. So, you charge $20
1:43a month. You get 10,000 users. That's
1:46200 grand a month. 2.4 million a year
1:49from code you wrote once sitting in your
1:52pajamas. That's the dream. And the dream
1:54is real. It happens. You've seen the
1:57stories, the indie hacker making 30
1:59grand a month from a tiny app he built
2:01alone, the two-person company pulling in
2:04millions with no office, no investors,
2:06no employees. They post the screenshots,
2:09the revenue graphs going up and to the
2:11right. And those stories are true, every
2:13word. What's not true is the implication
2:16that it's normal, that it's easy, that
2:19you'll be next. Those screenshots are
2:21survivorship bias in its purest form.
2:23For every revenue screenshot that goes
2:25viral, there are 10,000 founders staring
2:28at a flat line, wondering what they're
2:30doing wrong. Nothing. They're doing
2:32nothing wrong. They're just living the
2:34normal outcome. And we're talking
2:35America here, the biggest software
2:37market on the planet. Silicon Valley
2:39didn't become Silicon Valley by
2:41accident. The US has the deepest
2:43pockets, the biggest companies, and the
2:45most aggressive investors on Earth. If
2:48you're going to sell software, this is
2:50the arena, which also means this is
2:52where the competition is absolutely
2:54savage. Everyone on the planet wants a
2:57piece of the American software market.
2:59You're not competing with the guy in the
3:00next garage, you're competing with the
3:02entire world, just not the way you
3:05think. Want proof the dream is real?
3:07Look at Adobe. They used to sell
3:08Photoshop in a box for $2,500
3:11once. Then they switched to
3:13subscriptions, 60 bucks a month forever.
3:16Their revenue didn't just grow, it
3:18tripled. Microsoft did the same with
3:20Office. The entire industry learned one
3:22lesson, renting software beats selling
3:25it every single time. That's the dream
3:28pulling you in, and it's not fake, it's
3:30just incomplete. Here's the thing nobody
3:33tells you, starting a software company
3:35is cheap, running one will eat you
3:37alive. But wait, you say, all I need is
3:40a laptop. Sure, and all a restaurant
3:43needs is a stove. Let's talk about the
3:45actual bill, because before you sell a
3:47single subscription, you're going to
3:49spend money that'll make your stomach
3:51drop. First, people. You can't code
3:54everything yourself forever. A decent
3:56software engineer in the US costs you
3:59150 to 250 thousand dollars a year. Not
4:02a team, one engineer. You need three of
4:05them? That's half a million a year
4:07before rent, before laptops, before
4:10anything. Fine, you say, I'll hire cheap
4:13developers overseas. People try this
4:15every single day and every single day
4:18they learn the same lesson. Cheap code
4:20is the most expensive code there is. You
4:22save 50 grand on salaries and spend a
4:25year rebuilding everything from scratch.
4:27Second, time. The average software
4:29product takes 12 to 18 months to build
4:32properly. That's 12 to 18 months of
4:35paying salaries with zero revenue coming
4:37in. Do the math. Three engineers, 18
4:40months. You're burning through nearly a
4:42million dollars before a single customer
4:44pays you a dime. Okay, you say, I'll
4:47raise money from investors. Great. Now,
4:49you don't really own the company
4:51anymore. You work for the investors and
4:53they will remind you of that at every
4:55single board meeting. And don't forget
4:57the co-founder math. You split the
4:59company three ways, then investors take
5:0120% then the next round takes 20 more.
5:04By the time you're successful, you own a
5:06slice of a slice of a slice. Founders
5:09joke about it. I own 4% of a billion
5:12dollar company. Sounds amazing until you
5:14realize that's after 10 years of 80 hour
5:17weeks and a salary half what you'd make
5:19at Google. This is the first brutal
5:21truth. Software has almost no startup
5:24costs and massive running costs. The
5:26laptop is cheap, the humans are not.
5:29Third, the hidden bill. The stuff nobody
5:32budgets for. Cloud hosting. Your app
5:34lives on Amazon's computers and Amazon
5:37sends you the bill. 20 grand a month
5:39once you have real users. Software tools
5:42your own team needs. Another 50 grand a
5:44year. Lawyers for the incorporation, the
5:46contracts, the patents you'll never
5:48enforce, accountants, insurance. Add it
5:52up and the free startup costs you a
5:54quarter million a year before you've
5:56hired anyone. The laptop was the
5:58cheapest part of this entire journey.
6:00Everything after it costs real money.
6:02And since we're talking America, add the
6:04American employment tax. Health
6:06insurance for your team? 15 to 20 grand
6:09per employee per year. Payroll taxes on
6:12top of every salary. A decent office in
6:14any real American city? 5 grand a month
6:17minimum. The cheap startup in America
6:19burns cash faster than anywhere else on
6:22Earth because here everything costs
6:24more, including the humans. Okay, deep
6:27breath. Let's open the cash register and
6:30look inside. Revenue stream number one,
6:32subscriptions. This is the big one. This
6:35is the actual business. Forget selling
6:38software once. Nobody does that anymore.
6:40The real game is SaaS. Software as a
6:43service. You don't sell the app, you
6:45rent it. $20 a month, 50, 500, forever.
6:50And this is where the math gets
6:52beautiful. 10,000 customers paying 50
6:54bucks a month. That's $500,000
6:58a month, every month. It just arrives
7:01while you sleep, while you're on
7:03vacation. Recurring revenue, the holy
7:05grail of business. Investors go
7:08absolutely insane for this stuff. A
7:10normal business selling one-time
7:11products might be worth two or three
7:13times its yearly revenue. A SaaS company
7:16with sticky subscriptions? 10 times, 15
7:19times, 20 times revenue. That's why
7:22Schneider paid $22.6 billion for PTC.
7:26We'll get to that. Stick around. But
7:27here's what the gurus don't put in their
7:29tweets. And here's the advanced move the
7:31best companies pull, expansion revenue.
7:34Your customer starts with five seats,
7:36then they add 20, then they upgrade to
7:38the premium plan. The best SaaS
7:40companies grow more than 120% a year
7:43from existing customers alone without
7:46signing a single new one. It's called
7:48negative churn. Your existing customers
7:50pay you more over time, not less. That's
7:53the real money printer, not new
7:55customers, old customers spending more.
7:58But it only works if they stay, which
8:00brings us to the leak. Subscriptions
8:02only work if people stay, and people
8:05don't stay. It's called churn. Every
8:07month a percentage of your customers
8:09cancel. 5% monthly churn sounds small,
8:12right? Wrong. 5% monthly churn means you
8:15lose nearly half your customers every
8:18single year. You're running on a
8:19treadmill that's speeding up, and you
8:21have to sprint just to stand still. Let
8:23me make it real. You have a thousand
8:25customers. You lose 50 every month. To
8:28just stay at a thousand, you need to
8:29find 50 new customers every month,
8:32forever. To grow, you need a hundred,
8:35two hundred. And finding new customers,
8:37that's where the dream goes to die. And
8:39there's a cruel detail inside churn. Not
8:42all customers are equal. Losing a
8:44hundred tiny customers who pay ten
8:45bucks, annoying. Losing one enterprise
8:48customer paying ten grand a month,
8:50devastating. Smart companies track
8:52revenue churn, not just customer churn,
8:55because one big cancellation can wipe
8:57out a hundred small wins. Either way,
8:59the bucket leaks, and plugging it is a
9:02full-time job most founders never budget
9:04for. CAC, customer acquisition cost.
9:08Three letters that murder more software
9:10companies than bad code ever will.
9:12Here's how it works. You spend money on
9:14ads, on sales people, on marketing to
9:17win one customer. Let's say it costs you
9:19$300 to land a customer who pays 50
9:22bucks a month, six months to break even.
9:24Fine. But what if they cancel after four
9:26months? You just lost money on every
9:29single customer. You're growing straight
9:31into bankruptcy. This is the math that
9:34kills. The best software in the world
9:36means nothing if it costs you more to
9:38acquire a customer than that customer
9:40ever pays you. And here's the sick joke.
9:42The better your product category, the
9:44more expensive customers are because
9:46you're bidding against every competitor
9:48for the same eyeballs. Google Ads for
9:50software keywords, $10 a click, 20, 50
9:54for one click, not a customer, a click.
9:57So, you built the perfect app.
10:00Congratulations. Now you need to
10:02outspend giants just to tell anyone it
10:04exists. Investors have a golden rule for
10:06this, LTV to CAC, lifetime value versus
10:10acquisition cost. For every dollar you
10:13spend acquiring a customer, that
10:15customer should pay you at least $3 over
10:17their lifetime. Three to one, that's the
10:19bar. Below three to one, you're lighting
10:22money on fire. Above it, you have a
10:24machine. Pour money in, get more money
10:27out. And the payback period, how fast
10:29you earn back that acquisition cost,
10:31better be under 12 months because if it
10:33takes two years to break even on a
10:35customer and the average customer leaves
10:37after 18 months, congratulations, every
10:40customer loses you money. The founders
10:42who survive obsess over one question,
10:45where do customers come from cheap?
10:47Content, word of mouth, virality,
10:50anything but ads because ads are an
10:52auction and you're bidding against
10:54companies with deeper pockets. Okay, you
10:57say, "Forget small customers. I'll sell
10:59to big companies, million-dollar
11:01contracts." Welcome to enterprise sales.
11:04The revenue stream that looks like a
11:06cheat code and fights like a final boss.
11:08Yes, the contracts are enormous. A
11:11single enterprise deal can be worth
11:13hundreds of thousands, millions. PTC,
11:16the company that just got bought for
11:1722.6 billion dollars, sells to giant
11:21manufacturers, the kind of companies
11:23that build airplanes. That's why it's
11:25worth billions. But here's what nobody
11:27tells you about enterprise, the sales
11:29cycle is nine to 18 months. You don't
11:32sell to a big company, you survive a
11:35gauntlet. Demos, pilots, security
11:38reviews, legal reviews, procurement, a
11:40committee of 15 people who all need to
11:43say yes, and one person who can say no.
11:46You'll hire sales people who cost you
11:47200 grand a year each, and they'll close
11:51maybe four deals a year, if they're
11:53good. And the customer? They'll demand
11:55custom features. "We love your product,
11:58but can it also do this one thing only
12:01we need?" You say yes, because it's a
12:03million dollars. Now your clean,
12:06beautiful product has a weird tentacle
12:08growing out of it. Do that 10 times, and
12:10your software is a monster. Enterprise
12:13is where the big money lives. It's also
12:15where software companies go to become
12:17consulting firms wearing a software
12:18costume. And here's the part that breaks
12:20founders. You need a champion. One
12:23insider at that big company who sells
12:25your product when you're not in the
12:27room. Without a champion, your deal dies
12:29in committee. With one, it survives.
12:32Finding that person takes months. Then
12:35they get reorganized, or quit, and
12:37you're back to zero. Oh, and the free
12:39trial? The proof of concept? Big
12:42companies will run you through a 6-month
12:44pilot, consume all your engineers' time
12:46with customer requests, and then choose
12:49the competitor. It happens constantly.
12:51Enterprise deals are beautiful. They're
12:53also where quarters go to die. Okay,
12:56remember the mystery from the beginning?
12:58The thing that kills software companies
13:00that has nothing to do with code? This
13:02is it. Distribution. Nobody tells you
13:05this, but code is a commodity. There are
13:07millions of developers on this planet.
13:09Your app can be rebuilt by a team in
13:11another country in 3 months. What can't
13:14be copied? Your distribution, your
13:16audience, your brand, your sales
13:19machine. The winners don't win because
13:21they have the best product. They win
13:23because they figured out how to get
13:24customers cheaper than everyone else.
13:26Think about it. The best project
13:28management software ever built means
13:30nothing if nobody's heard of it. The
13:3210th best one with a killer sales team
13:34and a famous brand, worth billions. This
13:37is the brutal truth nobody tweets about.
13:39Building the software is maybe 10% of
13:42the business. Getting people to buy it,
13:44use it, and stay, that's the other 90.
13:47Every failed founder says, "We just need
13:49to finish the product." Every successful
13:51founder says, "We should have started
13:53selling on day one." Your code is not
13:56the business. Your ability to sell the
13:58code is the business. And this is where
14:01the famous lie dies. Build it and they
14:04will come. No, they won't. They never
14:07come. There are millions of apps sitting
14:09in app stores right now with zero
14:11downloads. Perfectly good software.
14:13Nobody knows it exists. "But ours is
14:16different." Every founder says that. The
14:18App Store doesn't care. Here's the
14:20reframe that separates winners from the
14:2210,000. Stop asking, "How do I build the
14:25best product?" Start asking, "How do I
14:28get 1,000 people to pay?" The first
14:30question leads to a beautiful product
14:32nobody uses. The second question leads
14:35to a business. History is full of proof.
14:38VHS beat Betamax. Worst product, better
14:41distribution. MySpace had the users
14:44first. Facebook won on distribution
14:46mechanics, not features. In software,
14:49Slack didn't win because it had the most
14:51features. It won because teams invited
14:53other teams. Bottom-up distribution that
14:56spread like a virus inside companies.
14:58The pattern never changes. The best
15:01product doesn't win. The best
15:02distributed product wins. So, when a
15:05founder tells you, "Our tech is 10 times
15:07better." Smile politely. Then ask, "How
15:10do customers find you?" Watch the
15:12silence. "But wait," you say, "what
15:14about moats? Patents? Secret
15:17technology?" Cute. Here's the thing
15:19about software. It has almost no moat. A
15:22factory costs billions to copy. A
15:24software product costs a few developers
15:27and some time. Someone will copy you. If
15:30you're winning, it's not a question of
15:32if, it's when. But they'll never copy it
15:34as well as us. Doesn't matter. They
15:36don't need to be better, they need to be
15:38cheaper or free. There is always someone
15:41willing to build a worse version of your
15:43product and give it away to steal your
15:45customers. The only real moats in
15:47software, network effects, where the
15:49product gets better the more people use
15:51it. Think social networks, marketplaces,
15:54and switching costs, where leaving is so
15:56painful that customers stay even when
15:59they're unhappy. This is why PTC is
16:01worth 22.6 billion dollars. We'll get
16:04there. Stick around. Everything else,
16:07marketing, speed, luck. If your entire
16:10business plan is, we have better code,
16:13you don't have a business plan. You have
16:15a hope. And now there's a new threat
16:17eating moats alive, AI. Code is getting
16:20cheaper every single month. What took a
16:22team of 10 a year, now takes two people
16:25a few weeks with AI tools. Think about
16:27what that means. If code keeps getting
16:29cheaper, then code is worth less, and
16:32distribution is worth more. The winners
16:35of the AI era won't be the best coders.
16:37They'll be the ones who own the customer
16:39relationship. Oh, and one more moat
16:41killer, open source. Someone, somewhere,
16:44is right now building a free version of
16:47your product. Not to make money, for
16:49fun, for reputation. And free is a price
16:52you cannot beat. Your moat isn't your
16:54code. It was never your code. All right,
16:57let's answer the question from the
16:58beginning. Why did a French company just
17:00pay 22.6 billion dollars in cash for
17:04lines of code? Because PTC doesn't sell
17:07code, it sells something much more
17:09valuable, lock-in. PTC makes industrial
17:12software, the kind giant manufacturers
17:15use to design airplanes, cars, machines.
17:18When a company builds its entire
17:20engineering process on PTC's software,
17:22switching isn't just annoying, it's
17:24terrifying. Imagine you're building
17:26airplanes. Every design, every part,
17:29every process lives inside this
17:31software. Switching to a competitor
17:33means retraining thousands of engineers,
17:36migrating decades of data, and risking
17:38mistakes on airplanes. Nobody switches,
17:42ever. That's a switching cost moat so
17:44deep it's basically a fortress. PTC's
17:47customers pay every year, year after
17:50year, because leaving would be corporate
17:52suicide. And that is what $22.6 billion
17:56buys. Not code, certainty. Recurring
17:59revenue from customers who physically
18:01cannot leave. Schneider didn't buy
18:03software, it bought a money printer with
18:05a padlock on it. And notice who paid,
18:08not a tech company, an industrial giant,
18:10a 200-year-old French energy company.
18:13Because the most valuable software in
18:15the world right now isn't social media
18:17or games. It's the boring, invisible
18:20software that runs factories, power
18:22grids, and airplanes. The market just
18:24told you exactly where the money is. The
18:27question is whether you're listening.
18:29And don't miss the symbolism. A French
18:31giant just paid a fortune for American
18:33software. The best industrial software
18:36on Earth is built right here, Boston in
18:38PTC's case. The whole world comes to
18:41America to buy software. That's the
18:43prize. And that's also why everyone
18:45wants to kill you for it. The details
18:48tell the story. $205 a share, a 42%
18:52premium over the market price, meaning
18:54Schneider paid nearly half again what
18:56the stock was worth. The deal closes in
18:58the third quarter of 2027. And here's
19:01the funny part, Schneider's own stock
19:03fell 8% on the news. Investors hated the
19:06price, but Schneider paid it anyway.
19:09Because when you find a fortress,
19:10customers who can never leave, revenue
19:13that never stops, you overpaid gladly.
19:16That's the signal for the whole
19:17industry. In the age of AI, the most
19:20valuable thing in software isn't the
19:21technology, it's the locked-in customer.
19:24Let's talk about where software
19:25companies go to die. It's called the
19:27Valley of Death, and almost everyone
19:30visits. Here's how it works. You raise
19:32money, say $2 million. That gives you
19:35about 18 months of runway. 18 months
19:37before the money runs out. Month 1 to 6,
19:40building, exciting. Month 6 to 12,
19:43launching, nobody comes, concerning.
19:46Month 12 to 15, pivoting, panic. Month
19:4915 to 18, begging investors for more
19:52money with a slideshow full of
19:54learnings. 90% of startups die. You've
19:57heard that stat, but here's what it
19:59actually means. They don't die because
20:01the idea was bad, they die because they
20:03ran out of money before they figured out
20:05distribution. The product was fine, the
20:07timing was fine, the bank account hit
20:10zero. This is why the laptop and a dream
20:12story is a lie. The laptop is cheap.
20:15Surviving 18 months with no revenue
20:17while paying humans? That's the real
20:19price of admission. And the founders who
20:21make it? Most of them will tell you they
20:23were weeks away from dying. The
20:25difference between a billion-dollar
20:27company and a dead one is often just one
20:30more month of cash. The Valley has
20:32levels, by the way. First, you burn the
20:34seed money. Then, you raise a bridge
20:36round, emergency cash to survive a few
20:38more months. Then comes the down round,
20:41raising money at a lower valuation,
20:43which wipes out everyone's ownership and
20:45morale. And the most common ending? Not
20:47bankruptcy, the acquihire. A bigger
20:50company buys your dying startup, not for
20:52the product, which they'll shut down,
20:54but for the engineers. Your dream ends
20:57as someone else's recruiting deal.
20:59Nobody tweets about the acquihire, but
21:01it's how most software stories actually
21:03end. So, who actually wins this game?
21:05It's not who you think. It's not the
21:07genius with the revolutionary app. It's
21:09not the team with the cleanest code. The
21:11winners are boring. They sell unsexy
21:14software to unsexy industries.
21:16Scheduling software for dentists,
21:18inventory tools for plumbers, compliance
21:21software for factories. Nobody writes
21:23tweets about them. Nobody makes
21:25documentaries about them. But you know
21:26what? Dentists pay every month. Plumbers
21:29don't churn. Factories can't switch. The
21:32winners pick a niche so small the giants
21:35ignore it and so painful the customers
21:37can't leave. They charge real money.
21:39They answer the phone. They survive.
21:41Meanwhile, the brilliant founders
21:43building the next big social app are on
21:45their third pivot, burning investor
21:47money, competing with a thousand clones.
21:50Boring wins. Exciting dies. If you want
21:53to own a software company, don't ask
21:55what's the coolest thing I can build.
21:57Ask what's the most painful problem for
21:59a customer who can actually pay. Look at
22:01the pattern. Toast, software for
22:03restaurants. Procore, software for
22:05construction sites. Veeva, software for
22:08pharma companies. Boring industries,
22:10boring products, enormous companies.
22:13Nobody at a dinner party brags about
22:15building compliance software for
22:16factories. But the founders of those
22:18companies are quietly rich, while the
22:21Uber for dog walking founders are
22:22quietly updating their resumes. The
22:25formula is brutally simple. Find
22:27customers with painful problems and real
22:29budgets in an industry the cool kids
22:31ignore. Then, never let them leave.
22:34Boring isn't compromise. Boring is the
22:36strategy. Take Toast. Started in Boston.
22:39Software for restaurants. Not exactly
22:41glamorous. Today, it's a multi-billion
22:44dollar American company. Or Procore,
22:47construction software from California.
22:49The founders didn't chase the shiny
22:51consumer app dream. They went where the
22:53pain was. The unsexy industries that
22:55America actually runs on. The ones
22:58Silicon Valley ignores. That's the
23:00American software playbook nobody talks
23:02about. Not the next social network, the
23:04next boring fortress. So, should you own
23:08a software company? Here's the honest
23:10math. The dream is real. 90% margins,
23:13infinite scale, billion-dollar exits.
23:16PTC just proved it again. But the dream
23:19has a price, and the price isn't code,
23:21it's distribution. It's surviving 18
23:24months of burning cash. It's outselling,
23:27outlasting, and outhustling 10,000
23:30competitors who can all code just as
23:31well as you. You're not in the software
23:33business, you're in the customer
23:35business wearing a software costume. If
23:38that excites you, if you love selling,
23:40if you can survive the valley of death,
23:42if you can find a boring niche with
23:44painful problems and paying customers,
23:47then yes, the economics are the best in
23:49any business we've covered. But if you
23:51just want to write code and have money
23:53appear, buy someone else's stock
23:54instead. So, here's your checklist.
23:57Before you start, do you have a way to
23:59get customers that doesn't depend on
24:00outspending giants? Can you survive 18
24:03months with no revenue? Is your niche
24:05boring enough that nobody wants it, but
24:08painful enough that customers will pay?
24:10Three yeses? You might just make it. Two
24:12yeses? You've got a fighting chance, but
24:15fix the third one before you quit your
24:16job. One or zero? Don't start yet. Go
24:19learn the game from inside a software
24:21company first, then come back. Because
24:24in software, the code is the easy part,
24:26and nobody tells you that until it's too
24:28late. If this video opened your eyes,
24:31hit like. It tells the algorithm to put
24:33this in front of the next future founder
24:34who needs to hear it. And subscribe,
24:36because this channel is just getting
24:38started. Every single video breaks down
24:41the real economics of a business, the
24:43dream, the bill, and the brutal truth
24:45nobody tells you. Your subscription is
24:47literally what helps this channel keep
24:49growing. So, join the ride, and I'll see
24:52you in the next one.

About this transcript

This page contains the full transcript of The Economics of Owning a Software Company by Cost Anatomy, generated from the public captions YouTube serves with the video. The transcript has 4,113 words across 631 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.