The Netflix success story is not really one story. It is three, stacked on top of each other, each one representing a company that had to be willing to undermine its own current business to build the next one.
The first Netflix killed the video store. The second Netflix killed physical media. The third Netflix is killing linear television.
Most companies manage one reinvention in a lifetime. The ones that manage three usually do it because someone at the top understood, years early, that the current thing was going to stop working and refused to pretend otherwise.
Reed Hastings spent most of Netflix’s first decade preparing for a future that hadn’t arrived yet and building a culture disciplined enough to get there when it did.
The CD in the Envelope
The founding story has been mythologized in the way all good founding stories are. The version Reed Hastings told for years was that he got hit with a $40 late fee for returning Apollo 13 late to Blockbuster and decided to build something better. Marc Randolph, the co-founder and first CEO, has since called this a convenient fiction. Blockbuster couldn’t even find the transaction in its records.
The real story is less dramatic and more interesting. Randolph and Hastings were carpooling together from Santa Cruz to Sunnyvale, where they both worked at Pure Software. Randolph was obsessed with Amazon and wanted to find a category of products that could be sold by mail the same way Amazon was selling books. They went through hundreds of ideas. VHS tapes were too bulky and too expensive. DVDs, which had just been introduced in the US in early 1997, were thin, light, cheap enough to stock, and durable enough to ship.
To test the shipping durability, they put a CD in a greeting card envelope and mailed it to Hastings’s house in Santa Cruz. When it arrived intact, the decision was made.
Hastings invested $2.5 million from the sale of Pure Software to fund the company. They launched April 14, 1998 with 925 titles, essentially every DVD that had been published at that point. The first DVD ever mailed to a Netflix customer was Beetlejuice. No one involved thought they were building a $300 billion company. They thought they were building an online DVD rental service.
What they actually built, over the next few years, was something more important than that: a subscription model for entertainment.
The Subscription Model That Changed Everything
The original Netflix was a pay-per-rental model, the same basic structure as Blockbuster but delivered by mail. You ordered a DVD online, they mailed it to you, you mailed it back, you paid for the rental.
In September 1999, two years after founding, they introduced the subscription model. Unlimited rentals, flat monthly fee, no late fees, no due dates. You kept a DVD as long as you wanted. When you mailed it back, they sent the next one on your queue automatically.
Barry McCarthy, Netflix’s CFO from the beginning through 2010, described what happened next: the business went from $1 million in revenue in 1998 to $5 million in 1999, then $35 million, then $75 million, then $150 million, then nearly $300 million. He credited the subscription model specifically. It was Reed Hastings’s insight that it would resonate with consumers in a compelling way. Once it existed, the old pay-per-rental model was gone within months.
The subscription model changed everything about the business. It changed the customer relationship from transactional to ongoing. It changed the unit economics because customers were paying monthly whether they watched a lot or a little. It changed the incentives because Netflix now benefited from customers watching more rather than fewer, which meant they needed to invest in selection and recommendation rather than just logistics.
Most importantly, it created the financial foundation for everything that came after. Without the subscription model locking in recurring revenue, Netflix would not have had the stability to survive the dot-com crash, survive the approach to Blockbuster, survive the transition to streaming, or fund the original content era.
Blockbuster Laughed Them Out of the Room
September 2000. Netflix was unprofitable, hemorrhaging cash, and the dot-com bubble was collapsing around them. Hastings and Randolph chartered a plane to Dallas. They had finally gotten a meeting with John Antioco, CEO of Blockbuster.
The pitch: a merger. Netflix would run Blockbuster’s online operations. Blockbuster would keep its 7,700 physical stores. Together they would own both the future of home entertainment and its present.
Hastings asked for $50 million. Antioco laughed. Other executives in the room laughed with him. The dot-com hysteria was completely overblown, Antioco said. Netflix was a niche player. The meeting ended. Hastings and Randolph flew home in silence.
Antioco was not wrong that the dot-com hysteria was overblown. Most of the companies in Hastings and Randolph’s cohort were indeed going to zero. He was catastrophically wrong that Netflix was one of them.
Blockbuster had something extraordinary and did not know what it was. Every day, millions of people walked into its stores and made decisions about what to watch. That behavioral data, what people looked for, what they rented, what they returned early, what they couldn’t find, was a map of American entertainment taste in real time. Instead of building a technology infrastructure to capture and use that data, Blockbuster protected the late fee business model that was generating roughly $800 million a year in revenue from customers who were already annoyed at the company.
Netflix was building the database of what people watched and the algorithm to predict what they would watch next. By the time Blockbuster understood what it had lost, Netflix was already winning.
The Cinematch recommendation algorithm was the moat that never got talked about in the early years because it looked like a feature rather than a strategy. Every rating a customer gave trained the system. Every DVD returned with a five-star review was a data point. Every queue built by a subscriber refined the model. Netflix got better at predicting taste with every interaction, which meant customers got better recommendations, which meant they rented more, which meant they generated more data. The flywheel ran on its own momentum.
The IPO Nobody Thought Would Work
Netflix went public on May 23, 2002, on the NASDAQ at $15 per share, selling 5.5 million shares. The timing was terrible by conventional standards. The dot-com crash was still fresh. Online businesses were toxic to institutional investors. The company was not profitable.
None of that turned out to matter because the business was actually working. DVD players were becoming mainstream holiday gifts. The subscription model was generating sticky, recurring revenue. The customer base was growing steadily, not despite the flat fee model but because of it. Customers who no longer worried about late fees rented more and stayed longer.
The stock did not immediately rocket. Netflix spent several years as a small, unglamorous public company that analysts mostly ignored. Then in 2003 they hit profitability for the first time. Then Hastings started doing something unusual for the CEO of a profitable, growing company: he publicly committed to streaming before streaming was technically viable.
In a 2005 interview with Inc., he said he thought Netflix had at least a decade of DVD dominance ahead of it, but movies over the internet were coming and they were already investing 1-2% of revenue annually in downloading technology to be ready when it happened. That is why, he said, the company was called Netflix and not DVD-by-Mail.
He was building the next company before the current one needed to change.
The Pivot That Destroyed the Current Business to Build the Future One
January 2007. Netflix launched Watch Now, a feature allowing subscribers to stream a selection of movies and TV shows directly to their computers. The initial catalog was around 1,000 titles. The streaming quality was limited by the broadband speeds of the era. It was not the product that would eventually be worth hundreds of billions of dollars.
It was the strategic declaration that mattered. Netflix was publicly announcing that physical media was not the destination. It was the vehicle they were riding until the highway was ready.
Here is what makes this unusual: at the time of the streaming launch, the DVD-by-mail business was growing, profitable, and had no obvious competitive threat. They were not pivoting away from a failing business. They were investing heavily in a new model that would, if it succeeded, cannibalize the existing one. Most companies in this situation delay the cannibalization for as long as possible. Netflix accelerated it.
By 2012, roughly one-third of Netflix’s revenue still came from the DVD business. In 2023, they shipped the last red envelope. The physical business that had been the foundation of the company was gone completely, replaced by something worth infinitely more.
The 2011 Qwikster fiasco is the most honest look at how hard this transition actually was. Hastings announced he would split Netflix into two separate services: the streaming product would remain Netflix, and the DVD service would be renamed Qwikster and operate separately. Customers hated it. The stock dropped 77% in four months. Hastings reversed the decision within weeks.
What the Qwikster disaster revealed was how deeply customers valued the integrated experience, even as streaming was clearly becoming the dominant product. The lesson absorbed was not to protect the DVD business. It was to not be stupid about how you manage the transition. The destination had not changed. The path needed adjusting.
House of Cards and the Original Content Bet
By 2011 and 2012, Netflix was winning the streaming wars. The competition was weak, broadband had gotten fast enough to make streaming genuinely good, and subscriber numbers were climbing steadily. The problem was invisible to most observers and urgent to Netflix’s leadership: they did not own any of what they were streaming.
Every piece of content on Netflix was licensed from a studio. Those studios were watching Netflix grow and understanding, correctly, that Netflix was the distribution layer eating their business. The obvious response was to license less content to Netflix at higher prices, or to pull content entirely and launch competing services.
Netflix was building its audience on someone else’s library. When the content owners eventually took their libraries back, the product would hollow out.
The solution was original content. If Netflix owned the shows and movies, no one could take them away.
In February 2013, Netflix released all 13 episodes of House of Cards simultaneously. This was a deliberate break from how television had always worked. No weekly episodes. No appointment viewing. No waiting. The entire first season at once, available to anyone with a subscription. Binge watching as a product feature rather than a viewer’s bad habit.
House of Cards worked on two levels. The creative level, it was a genuinely good political thriller that won Emmy nominations and gave Netflix credibility as a producer. The strategic level, it was proof that the original content model could generate the kind of cultural conversation that drove subscriber growth.
The data that informed the decision to greenlight House of Cards has become one of the famous anecdotes in tech strategy. Netflix knew, from subscriber viewing data, that the original British House of Cards series had been popular with users who also watched films directed by David Fincher and films starring Kevin Spacey. The American remake had both. Netflix bid $100 million for two seasons without seeing a pilot. The analytics gave them confidence the audience was there.
This was a different kind of content bet than Hollywood had ever made. Not based on gut instinct, development executive relationships, or the star system. Based on what the data said the audience would watch.
Stranger Things, The Crown, Bridgerton, Squid Game. The originals strategy built a library that Netflix owned and that no studio could take back. Content spending reached $15.4 billion in 2024 alone.
The 2022 Crisis and the Third Reinvention
In April 2022, Netflix reported its first quarterly subscriber loss in a decade. Then another one in Q2. The stock dropped more than 70% from its peak. Analysts questioned whether the streaming model was fundamentally broken. Competitors including Disney+, HBO Max, and Apple TV+ had launched with significant libraries and were taking share.
What actually happened is more interesting than the crisis narrative. Subscriber growth had been pulled forward by the pandemic. Tens of millions of people who would have subscribed over the following two or three years subscribed in 2020 and 2021 because they were stuck at home with nothing to do. When the pandemic lifted, the natural growth curve resumed, but it looked like a crash against the inflated pandemic baseline.
Netflix’s response was the third reinvention.
They cracked down on password sharing, which had been quietly tolerated for years. An estimated 100 million households were using shared passwords. Converting even a fraction of them into paying subscribers was a substantial revenue opportunity. The crackdown drove significant subscriber additions in 2023 and 2024.
They launched an ad-supported tier in November 2022. This was a philosophical reversal for a company that had defined itself for years as the premium, ad-free alternative to traditional television. The reasoning was the same as every other strategic shift: the audience at the price point Netflix wanted to charge had a ceiling, and there was a much larger audience willing to pay less in exchange for ads. The ad tier created a new revenue stream and opened Netflix to advertisers who had watched linear TV audiences age and shrink for a decade.
They moved into live events. WWE Raw. NFL Christmas Day games. A Jake Paul and Mike Tyson boxing exhibition that drew 60 million concurrent viewers. Live sports and events drive the appointment viewing that pure on-demand libraries cannot, and they support the advertising business with predictable, high-value inventory.
By the end of 2024 Netflix had surpassed 300 million paid subscribers. Revenue reached roughly $39 billion. Operating margins hit 30% in 2025. The “broken” company of 2022 looked very different in 2025.
The Culture That Made It Possible
The reinventions did not happen by accident. They happened because of a specific organizational design that Reed Hastings built, documented in an internal culture deck that became widely circulated, and eventually published as a book.
The core principle is that Netflix operates with high talent density and high freedom. Not just freedom in the casual sense but genuine latitude to make consequential decisions without approval chains. The tradeoff is that adequate performance results in a generous severance package. Netflix does not manage people toward improvement. It identifies whether managers would fight to keep an employee, and if the answer is no, the employee leaves.
This is brutal in the way that the most honest management philosophies are. It requires managers to make hard assessments regularly rather than deferring difficult conversations indefinitely. It means the average Netflix employee is better than the average employee at most large companies, which creates a self-reinforcing dynamic where talented people want to work there because the people around them are excellent.
The freedom side of the equation is the part that enabled the reinventions. If mid-level employees can make significant product decisions without six levels of approval, the organization can move faster than competitors who require consensus at every step. When Hastings decided to bet on streaming before streaming was viable, or original content before the library was empty, or live events before the ad tier existed, those bets required fast execution by people who were empowered to act.
The Warner Deal and What Comes Next
In December 2025, Netflix agreed to acquire the studio and streaming divisions of Warner Bros. Discovery in a cash-and-stock deal valued at roughly $82.7 billion. The deal would bring HBO, HBO Max, the Warner Bros. studio infrastructure, and the company’s extensive film and television catalog under Netflix’s ownership.
At the time this article was written, the deal was under active competition from a rival Paramount Skydance bid, and was still subject to regulatory approval. Whether it closes as structured or not, the intent is clear. Netflix is not content to be the largest streaming service. It is building toward being the dominant entertainment company on the planet, owning not just distribution but studio infrastructure, franchise IP, and the most prestigious library of serialized drama ever produced.
The arc from a mailed CD in a greeting card envelope in 1997 to a $82.7 billion studio acquisition in 2025 is an absurd distance to travel. It happened because a company was willing to kill what was working to build what would work next, three times in a row, and build the culture capable of executing each reinvention without losing the core.
Marc Randolph, the co-founder who left in 2002 and watched everything that followed from the outside, said the Warner deal blew his mind. He never saw it coming.
Neither did Blockbuster.
Leave a Reply