Marc Randolph and the idea that survived the mail

A compact disc goes into an envelope in Santa Cruz. It travels through the mail, arrives at Reed Hastings’s house, and emerges unbroken. In the story Marc Randolph tells about the beginning of Netflix, this is the experiment that matters. There is no audience, no launch announcement, no screen full of movies. There is a piece of plastic and a practical question. Can you send one of these things without destroying it?

It is an appealingly modest opening for a business that would become part of the furniture of everyday life. Randolph and Hastings were considering renting movies over the internet in 1997. Videocassettes made the arithmetic troublesome: bulky objects, expensive journeys. DVDs offered a smaller package. Since the two couldn’t find a DVD for their test, a music CD stood in. The postal service supplied the verdict.

Everything after that would require rather more persuasion. A disc surviving an envelope proved something about postage. It left the awkward questions about customers, prices, competition, and repeat purchases entirely open. Randolph’s life is interesting in that space between a promising experiment and a business people will use. He has spent years explaining how much work fits inside it.

The education of a mail-order enthusiast

Born in Chappaqua, New York, in 1958, Randolph studied geology at Hamilton College. When he returned to address the class of 2020 at its delayed commencement celebration, he offered a record few university prospectuses would put on the cover: a 2.7 GPA, a failed English literature course, and Ds in both microeconomics and macroeconomics. The future entrepreneur had made a reasonably thorough inspection of academic disappointment.

He also described an assortment of early jobs: a ski shop in Memphis, a bar in Colorado, real estate in New York, and work as a gofer for the head of a sheet music company. It was a useful speech for graduates expected to have their futures arranged before their gowns went back on the hanger. His career supplied an argument for taking a less tidy route.

The sheet music company was Cherry Lane Music. Randolph learned direct marketing through its mail-order operation, experimenting with ways to sell its catalogue to consumers. That work gave him an education a transcript could not capture. A mailing invites a response. Change the offer and you can watch the response change. The customer, rather inconveniently for anyone attached to a theory, gets a vote.

Computer mail-order businesses and marketing at Borland followed. By the time Netflix was under discussion, he had already spent years thinking about selling from a distance. A website could offer something a printed catalogue could barely attempt: a page adapted to the individual looking at it. The internet gave an experienced marketer a new set of tools, and a familiar question about what would persuade someone to order.

Highway 17 was the first meeting room

Randolph met Hastings through the software business. Integrity QA, where Randolph worked, was acquired by Pure Atria, which Hastings led. When another acquisition put their next jobs in question, their commute between Santa Cruz and Silicon Valley became a place to consider new businesses. Randolph pitched ideas; Hastings helped interrogate them. A carpool has the useful property that everyone must remain for the entire presentation.

The possibilities included personalized shampoo, customized baseball bats, and dog food tailored to individual dogs. In retrospect, these can sound like comic supporting characters surrounding the inevitable winner. At the time, they were candidates. Netflix had no privileged claim on the future. It had to earn that claim through a sequence of decisions, including the simple decision to find out whether a disc could travel cheaply.

Randolph is impatient with the polished origin story that makes a business spring fully formed from an epiphany. The famous account of Hastings’s Blockbuster late fee makes the beginning easy to remember. Randolph’s account gives the commute, the rejected ideas, and the experiments more room. It is messier, and it leaves aspiring founders with something they can actually try. Renting a car and brainstorming would be excessive. Mailing the disc was enough.

A working website, an unfinished business

Netflix opened its website on April 14, 1998. Randolph was its first CEO. The launch offered DVD sales and individual rentals; the subscription arrangement people remember came later. This detail puts a welcome dent in hindsight. Even a founder who wanted recurring revenue began with a model that did not provide it.

Randolph remembers spending six months and a million dollars preparing the first website, then discovering how little preparation could guarantee. The site crashed. Customers found it confusing. Their behavior refused to follow the plan. Careful presentation had brought the company into contact with reality, which had neglected to read the presentation.

His response gradually changed. Elaborate tests became faster tests. Photography, copy, and checking absorbed less time; learning absorbed more attention. The point was to discover which offer mattered before spending another stretch perfecting an offer customers would ignore. For someone trained to persuade, there was a discipline here: listen to the answer, especially when it is unwelcome.

In 1999, subscriptions, a queue of selected films, and automatic replacement of returned discs came together. Customers could keep movies without due dates or late fees. A rental website was becoming a service with a place in the household routine. The queue also turned the next choice into something already made, waiting for the previous disc to come home.

The transformation is easy to flatten into a single brilliant decision. Its texture is more useful. Shipping had to work. The offer had to make sense. People needed a reason to return. Each answer restricted the next question. Randolph’s contribution belonged to this early period, when the company was still discovering what kind of company it could be.

The title he could let go

There was also a question about who should run the business. Hastings told Randolph he wanted to become CEO. Randolph has described the conversation as painful. He sat in the dark as the office closed around him, then went home and talked it through with his wife. Having an argument for change does not make being the subject of that change pleasant.

He concluded that the company would be stronger with Hastings taking the role. The choice forced him to separate two ambitions that had previously traveled together: being CEO and building a successful company. Employees were giving the venture their time and effort. His preference about the title had to sit alongside an obligation to them.

“Culture is not what you say; it’s what you do.”Marc Randolph

That sentence has a particular weight when the speaker has had to act on it. Randolph and Hastings continued working together. He credits their collaboration with the changes that made the rental model succeed. A founder giving up the chief executive’s chair can be presented as defeat or enlightenment. His description allows for disappointment, adjustment, and a practical judgment about the work ahead.

By 2003, he was ready to leave Netflix. The company had gone public the previous year and was hiring people with abilities suited to its expanding operation. He loved the early stages: questions around a table, problems without settled answers, the pleasure of getting something started. As Netflix matured, the daily job became less like the work he wanted.

A buyer declines, a founder carries on

Before that departure, there had been a much less comfortable exit possibility. In 2000, Netflix approached Blockbuster about being acquired. The proposed price was $50 million. Randolph recalls the meeting in Dallas ending without the rescue the founders hoped for. Today the episode often arrives as a joke with Blockbuster at the receiving end. For the people flying home, it was a financing problem still unsolved.

Randolph’s account preserves an excellent wardrobe detail. The invitation came during a company retreat; he had brought casual clothes. Hastings’s Hawaiian shirt at least had buttons. Even a consequential business meeting must accommodate the small indignities of packing. The anecdote is funny because the stakes and the clothes belong to entirely different occasions.

The rejected proposal also complicates the idea that founders always know they should remain independent. These founders were looking for help. They could see an argument for combining Netflix’s online work with Blockbuster’s stores. That argument failed to convince the buyer. Carrying on was a response to the situation they faced, with the ending still unavailable.

Back at the beginning, by choice

Leaving Netflix did not end Randolph’s involvement with startups. Looker, the analytics software company he helped co-found, became another part of his career. Google announced its acquisition agreement in 2019 and announced completion in February 2020. The $2.6 billion transaction was company news, rather than a measure of Randolph’s personal fortune. The distinction matters when a founder’s biography becomes a string of large numbers.

His 2019 memoir, That Will Never Work, brought the early Netflix years to readers. A podcast with the same name makes room for another sort of story: an entrepreneur brings a business problem, and Randolph helps work through it. He also appears as a judge and investor on Entrepreneur Elevator Pitch. The format changes; his interest in the uncertain beginning persists.

He remains fond of paper notebooks. They are available, portable, and socially legible: someone writing a note looks like someone writing a note. An electronic device leaves room for suspicions about Facebook. For an entrepreneur associated with a digital service, it is a pleasingly ordinary preference. The idea needs somewhere to land before anyone can test it.

There is a life beyond the table, too. Randolph has spoken about climbing, skiing, surfing, and mountain biking, and about protecting time with his family. His current biography lists board work with NOLS and the Truckee Donner Land Trust. The outdoors has remained part of the life around the companies, rather than an activity postponed until every business question is settled.

The customer still gets a vote

In a January 2026 essay, Randolph returned to recruiting. He argued that interviews should leave candidates wanting to join, even when they will not receive an offer. A rejected applicant still has friends, colleagues, and a memory of the encounter. Reputation accumulates through ordinary treatment. Here the marketer’s concern with a response becomes a concern with how people leave a room.

His stated mission now is to help people get started. The invitation is modest enough to be useful. An idea can remain comfortable in a notebook indefinitely; acting on it supplies an answer, and usually another question. Randolph’s disc arrived in one piece. The business required years of revision. The envelope gave him permission to begin.