Tech Brew Ride Home - (IHP) Amazon's Founding Part 1

Episode Date: July 3, 2023

Originally published February 2015. Finally, the long-promised foray into e-commerce, starting with… not the first… but practically the first… player in the space… and ironically enough, the 800 pound gorilla in the space to this day. Amazon. Dot com. We examine Jeff Bezos, the man. We consider Amazon, the idea. We look at e-commerce, the concept. It’s interesting. It’s groundbreaking. It’s available with free 2-day shipping for Prime members. Just kidding. Bibliography:  The Everything Store: Jeff Bezos and the Age of Amazon  The Playboy Interview: Moguls  Amazon.com: Get Big Fast  One Click: Jeff Bezos and the Rise of Amazon.com  http://archive.wired.com/wired/archive/7.03/bezos_pr.html  http://archive.fortune.com/magazines/fortune/fortune_archive/1999/01/11/253770/index.htm  http://partners.nytimes.com/library/tech/99/03/biztech/articles/14amazon.html  http://www.retireat21.com/blog/10-companies-started-garages  http://davidsheff.com/article/jeff-bezos/ Learn more about your ad choices. Visit megaphone.fm/adchoices

Welcome to the Internet History podcast. I'm your host, Brian McCombe. This finally is Chapter 7, Part 1, the chapter where we'll dive into the birth of Amazon.com and the rise of e-commerce on the web. So let's begin by doing this. If you were to do a simple thought experiment and put yourself back in 1994

when the web first began to go mainstream, and if you were someone looking to launch a business on this new medium, be honest with yourself. What would be the logical business models that you would start to experiment with? Well, the web, first and foremost, seems to be a new publishing medium, right? So launching a content site would probably be an obvious way to go. Maybe you just port some real-world media content onto the web, or maybe you launch some form of web-native content site.

And as we've seen in previous chapters, this is exactly what a lot of the early large-scale websites attempted to do. Sites like CNet, ZDNet, Hotwired, and Pathfinder, to name just a few. Alternatively, of course, you could maybe launch a service that would leverage the power of the web itself. In other words, you could monetize the web itself or monetize the tools of the web, and the unique features that make the internet so wonderful. Obviously, the search engines like Yahoo and Excite went this route, as did early bulletin board sites and web-based email sites like Hotmail, Rocket Mail, etc.

And I'm thinking here of early online classified sites as well, like Match.com and Monster. More on those in later episodes. So in this thought experiment, if you went with either of those two routes, content site or service, how then would you generate revenue? Well, as we've seen, the entire web seemed to fall backwards into advertising as a revenue model. Sure, AOL was making money by charging subscribers to interact in chat rooms, but very few other players on the early web were able to find any kind of success doing the same. Aside from porn sites, of course, subscriptions simply didn't seem to work on the early web.

But maybe you're saying to yourself right now, hey, what about e-commerce? Why depend on advertising and third-party commercial partners just to make money? Why not cut out the middlemen and just sell stuff directly to your audience? And I'd agree with you. With the benefit of hindsight, commerce certainly seems like the third obvious revenue play on the web. Just set up a virtual storefront and start selling stuff. and certainly when the dot-com era, as we remember it in our popular imagination rolled around, it was to a large degree just about taking a certain commercial leech and porting it to the

internet, a toy store on the web, a pet store on the web, coupons on the web, that seemed to be what everyone was doing. But while e-commerce today looms so large in the ecosystem of the web, it's important to remember that e-commerce represents one of the biggest conceptual leaps that the early web made. Put simply, it was not at all obvious to people at the time that the web would be a viable platform for commerce at all. The main reason? Well, consumers didn't trust the web. In the previous chapter, we mentioned briefly how hesitant people were to transact business on the early internet. I could quote from hundreds of skeptical cautionary articles of the time,

from any major publication you could name, that would describe how nervous people were about whipping their credit cards out online in the early 1990s. And in fact, I think it's really important to keep in mind this chronology for context. If the web explodes in 1994 and 1995, it's not really until late 1998 and really maybe 1990s, 1999 that e-commerce itself gains widespread acceptance. So there's about a three or four-year gap there from the time that the web enters the

mainstream until e-commerce is mainstream as well. Quite simply, e-commerce didn't have an easy evolution on the web. It wasn't by any means what you would call a no-brainer, no matter how obvious it might seem to us with the benefit of hindsight. Another thing to keep in mind is that prior to 1994, the internet had absolutely zero history with commerce. As a quasi-government system, commercial activity on the net had been outlawed by statute for most of the web's existence. In fact, the barriers to commercialization were only fully brought down in early 1994. Thus, while things like community websites, message boards, chat rooms, content, media websites,

Even search engines could draw on previous technologies and net experiments. When entrepreneurs first stepped forward to try to sell things on the web, there were almost no previous examples to build off of. There were very few early e-commerce pioneers. And so we come to the first major e-commerce pioneer. Not technically the first, but truly the first effort that attempted e-commerce in any large-scale way. So far in this podcast, we've been profiling plenty of pioneering web efforts that helped build the world we're familiar with today. But have you noticed that very

few of the early companies we've profiled are still standing even? Those that are still with us often struggle for relevance in the modern world. It's interesting then that two of the earliest and most important e-commerce efforts, Amazon and eBay, are not only still with us today, but are still absolutely dominant in their respective spheres of e-commerce influence. As I say, there were plenty of smaller e-commerce pioneers, and we'll touch on them when merited, but the next few chapters will mostly be about Amazon and eBay,

not just because they're the names everyone knows, but because, and again, this is a rare thing, thing in our narrative, believe me, when it comes to e-commerce, the guys who got there first, and the guys who were responsible for the greatest early innovations, are actually the same guys who ended up winning the e-commerce game by and large. So let's begin at the beginning with Amazon and Jeff Bezos. Jeff Bezos was born January 12, 1964, in Albuquerque, New Mexico. I don't know why I think it's important to put this in generational context, but I actually do think it is.

So Bezos would roughly be 10 years younger than Steve Jobs and Bill Gates, who were both born in 1955, and seven years older than Mark Andreessen. So he roughly fits in a mini-generation between the personal computer pioneers and the cohort that made up the web pioneers. For further context, we can note that. that Mark Zuckerberg was born in 1984, so by my estimation we could measure the Web 2.0 generation as being born roughly between 1984 and, let's say, 1990. Jeff Bezos was actually born Jeffrey Preston Jorgensen in Albuquerque

to Jacqueline and Ted Jorgensen. His maternal grandfather was the regional director of the U.S. Atomic Energy Commission in Albuquerque, Kirkie, and coincidentally worked under the auspices of DARPA in the 1950s. Bezos's father was literally a circus performer at the time of Bezos's birth, and was apparently one of the best unicyclists in the country in the 1960s, at least according to the journalist Brad Stone, who recently uncovered all of this, for his excellent recent book The Everything Store.

But Jeff's mother was a teenager at the time of his birth. So the marriage between his parents didn't last very long. When Jeff was four, his mother remarried to a Mike Bezos, also known as Miguel, who was a Cuban immigrant to the United States, who came here when he was 15 years old, thanks to a South Florida Catholic priest, who rescued teenagers from the Castro regime during the 1960s. Mike Bezos, who legally adopted his new stepson, soon got a job working for Exxon as a petroleum engineer. And so the new family moved to Houston. There is every indication that Jeff Bezos was an extremely gifted and intelligent child.

In fact, his intelligence is on record, because in the early 1970s, a book was written that took a look at the early gifted programs that were being developed for. for gifted education, called Turning on Bright Minds, a parent looks at gifted education in Texas. In that book, the author, Julie Ray, profiled a 12-year-old in a gifted enrichment program at River Oaks Elementary School, West of Houston. She named the 12-year-old Tim in order to protect his privacy

and described him as, quote, a student of general intellectual excellence, slight of build, friendly, but not serious, end quote. Coincidentally, some of his teachers described this same Tim as, quote, not particularly gifted in leadership, end quote. Those others admitted, quote, there is probably no limit to what he can do given a little guidance, end quote. Of course, the 12-year-old Tim that we're referring to here was very, really Jeff Bezos. Like Bill von Meister, young Jeff Bezos showed an affinity for gadgetry, science, and general tinkering, said his mother, quote, I think single-handedly we kept many radio shacks in business,

end quote. This inquisitive nature was nurtured by his maternal grandfather, Preston Gies, he of the U.S. Atomic Energy and DARPA fame. And according to his mother, a lot of young Jeff's formative lessons were learned on his grandfather's ranch in Kotula, Texas, near the Mexican border, where Jeff would spend his summers. Young Jeff learned how to do everything from fixing windmills to branding cattle, vaccinating cattle, even castrating cattle. Said his mother, Jackie, quote, one of the things Jeff learned is that there really aren't any problems without solutions. Obstacles are only obstacles if you think they're obstacles.

Otherwise, they're opportunities, end quote. The Bezos family eventually moved to Pensacola, Florida, and later to Miami, Florida, where Jeff spent his high school years. At Palmetto High School in Miami, Jeff was valedictorian in a class of 680 students. He won the school's best science. student award, his sophomore, junior, and senior years. He won the best math student award, his junior and senior years. And he attended the student science training program at my alma mater,

the University of Florida, while he was still a high schooler. He was also a national merit scholar. Outside of school, young Jeff worked the usual menial jobs average to high school students at this time. For example, an introduction to the world of retail came when Bezos was a fry cook at McDonald's. Also in his free time, famously, Bezos became a rabid Star Trek fan. And so another interesting story about him is that he won a student trip to NASA's Marshall Space Flight Center in Huntsville, Alabama, for writing a student paper entitled, The Effect of Zero Gravity on the Aging Rates of the Common House Flood. When his hometown paper, the Miami Herald, interviewed local valedictorians in 1982 for a feature article, Bezos spoke quite seriously about wanting to someday help build a commercial space station because he believed the future of the human race was not on Earth, what with the chance of a meteor crashing into the planet and wiping us all out.

he told the paper, quote, The whole idea is to preserve the earth, to get all the people off the earth, and to see it turned into a huge national park, end quote. His high school friend, Ursula Werner, actually Bezos's high school girlfriend, and valedictorian of the class before his, was later quoted as saying that she believed Bezos's ultimate goal

for his entrepreneurial endeavors, was to amass enough of a personal fortune that he could build his own space station. That's as good a theory as any, as Bezos seems to be in a race to space, along with the likes of Elon Musk and the Google Guys. When later confronted with this idea by Wired Magazine, Bezos originally laughed, but then said earnestly, quote, I wouldn't mind helping in some way. I do think we have all our eggs in one basket, end quote. Jeff enrolled in Princeton in the fall of 1982, and there he seems to have encountered the limits of his amazing intellect for the first time. Taking an honors physics course, Bezos briefly considered becoming a theoretical physicist.

And yet, as he later told Wired Magazine, despite being among the top 25 students in the class, quote, it was clear to me that there were three people in the class who were much, much better at it than I was, and it was much, much easier for them. It was really sort of a startling insight that there were these people whose brains were wired differently, end quote. Bezos instead would graduate from Princeton,

summa cum laude, with a, BSE in electrical engineering and also computer science. He graduated with a GPA of 3.9. His thesis project was to develop and build a special purpose computer for calculating DNA edit distances. After college, Bezos was reportedly recruited by the likes of Intel, Bell Labs, and Anderson Consulting, among others. But he chose to take a job.

at a sort of tech startup called FitTel that was looking to build a computer network that would clear and settle equity trades for Wall Street firms. The network was actually built on top of Jeannie, that early online network that was an early peer of CompuServe and Prodigy. Bezos spent two years at FitTel, where he was a programmer and software engineer, eventually rising to become head of development and director of customer service. He soon jumped to a more traditional finance firm, Bankers Trust, where he was a product manager. And it was during this time period that he briefly collaborated on an abortive newsletter by fax service with Halsey Minor. Listen to our interview episode with Halsey for more on that.

and by 1990 Bezos had about had it with the finance industry and was looking to leave Wall Street altogether when a recruiter convinced him to consider applying at a new style hedge fund called D.E. Shaw. Today, the idea of a quant fund, hedge fund that uses sophisticated computer algorithms to spot minute inefficiencies in the stock market and quickly pounce on the arbitrage opportunities is a fairly commonplace thing. But in the late 1980s, this was a brand new field. Bezos himself described the fund this way, quote, traditionally, quantitative hedge funds had built sophisticated computer-based tools to help traders make trading decisions. What D.E. Shaw and company did was invert that whole model. Computer programmers would program the machines and teach the machines financing.

The machines would actually make all of the trading decisions. End quote. The company's founder was David Shaw, who had gotten a Ph.D. in Computer Science from Stanford University, where he had rubbed shoulders with the likes of Jim Clark. In Shaw, Bezos found a mentor and a kindred spirit, a brainiac and a polymath, a, and a polymath, after his own heart, who used computers to beat Wall Street at its own game. Bezos was hired in December of 1990 as a vice president. Only two short years later, at age 28, he became the firm's youngest ever senior vice president.

David Shaw liked to think of his company as a sort of tech firm, but a tech firm that just so happened to apply itself to finance as its first market of operation. Given the firm's technical nouse, Shaw had an instinct that said his company might be perfectly positioned to seize any business opportunities that the rapidly growing and recently commercialized internet might pop up. And so Jeff Bezos and other executives at D.E. Shaw were given the task of researching and investigating any ideas surrounding the internet that might be marketable. See if any of these ideas that they came up with and pitched to David Shaw sound familiar to you. One was a free advertising-supported email service. Or how about this one? An online financial service portal that would allow people to trade stocks online at a reduced fee.

It turns out that D.E. Shaw would later develop and spin off both of these ideas. One would be called Juno, which if you'll remember was a low-cost advertising-supported ISP that later merged with Net Zero, and the other would be Farsight Financial Services, which was an early competitor to E-Trade, and which would later be sold to Merrill Lynch. But one idea that David Shaw himself was particularly fond of, and that Jeff Bezos specifically was researching, was something that D.E. Shaw executives would later remember was nicknamed, quote, the everything store. The idea was simply to harness computer networks and the internet to be a sort of intermediary between buyers and sellers of every product imaginable, creating efficiencies in the market and taking a small percentage for the trouble.

This was not a unique idea at the time. AOL would spend several years in the 90s trying to leverage its platform to do something along the same lines, and at Microsoft, product visionary Nathan Meyerhold was obsessed with a similar notion of how Microsoft could leverage its dominant platform to extract a tiny, what he referred to as a vigorous, cut from all the transactions in the future. Once those transactions moved online, of course, and presumably onto Microsoft's Windows platform. researching this angle for Shaw, Bezos decided that an Everything store was maybe a bit too grandiose, and so instead he began investigating product categories that might fit as a test case for the Everything Store concept. He researched roughly 20 different possibilities, including computer software, office supplies, and music.

But he settled on books as the best test case because as Brad Stone put it in his Everything Store book, books were, quote, pure commodities. A copy of a book in one store was identical to the same book carried in another, so buyers always knew what they were getting, end quote. This is, of course, different than something like clothing, which has all sorts of vagaries when it comes to things like size, cut, shape, color, etc. One has to imagine that music was strongly considered, but as Stone points out, books had the advantage over CDs at the time because there were only two major book distributors in the country that every bookseller had to work with, Ingram and Baker and Taylor.

This is in comparison to the several major and hundreds of minor record labels in the country at the time. And as Stone also points out, books have what we would now refer to as a strong long tail. There were three million different titles of books in print worldwide, as opposed to only 300,000 different titles of music. No single store, therefore, could shelve all those titles, but an online store could, at least virtually. So here was a differentiator that was meaningful. as Bezos himself would later say, quote,

with that huge diversity of product, he means titles, of course, you could build a store online that simply could not exist in any other way, end quote. And finally, the book business was fragmented, but it was enormous. Even the book-selling behemists of the time, Barnes & Noble and Borders,

only controlled about 25% of the market, and that market represented an estimated world, sales of $82 billion by 1996. It seems that in the course of his research on the Internet as a business opportunity, Jeff Bezos, like Jim Clark, and others, was suddenly bowled over by the sheer growth numbers that the Internet was experiencing at the time. Apparently, Bezos ran across some data by an analyst named John Quarterman, who claimed that the amount of bytes transmitted over the Internet from January 1993 to January,

to January 1994, so 12 months, had increased by a factor of 2057. I'm not saying that this was a percentage increase of 257. I'm saying it was increasing by a factor of 257, so roughly a 200,000 percent increase. As Bezos himself later pointed out, quote, things just don't grow this fast outside of petri dishes, end quote. And so this is one of those cases

where I have to say what happens next is frankly unclear to me. What is clear is that in the spring of 1994, Jeff Bezos left D.E. Shaw and struck off on his own to found what would eventually become Amazon.com. Now, why he was allowed to do this is what I'm not exactly sure about. After all, as we've stated, David Shaw did launch companies based on the other internet brainstorm ideas that his employees came up with, Juno and the online stock trading site.

And from Bezos's point of view, it's obvious why he would want to strike off on his own, because if Amazon launched as just another experimental division of D.E. Shaw, it would only be something that he could possibly run on behalf of David Shaw. Instead, Bezos was allowed to launch the idea on his own and plant his own flag, testing the Everything Store theory, by using books as the list. the first guinea pig product category. Several sources say that David Shaw took a long walk with Jeff Bezos in Central Park and maybe tried to talk him out of it, leaving the company,

but that in the end, Shaw reportedly told Bezos that, though one day they might be competitors, he was free to start Amazon by himself. It seems that at least tacitly, Shaw gave Bezos his blessing to take a run at the everything store. From Bezos's side of the story, he tends to focus on this moment as epitomizing the classic entrepreneur's dilemma, i.e. leaving the safe job to strike off on his own, and in this case, leaving Wall Street without even collecting his yearly bonus. In later speeches, Bezos would recount, quote, I knew when I was 80 that I would never, for example, think about why I walked away from my 1994 Wall Street bonus right in the middle of the year at the worst possible time.

That kind of thing just isn't something you worry about when you're 80 years old. At the same time, I knew that I might sincerely regret not having participated in this thing called the Internet, that I thought was going to be a revolutionizing event. When I thought about it that way, it was an incredibly easy decision to make, end quote. There's a well-worn legend that what Jeff Bezos did next was pack up his car with his young wife, McKenzie, by the way, who he met at D.E. Shaw when she was an administrative assistant, and head west, unsure of where he was going, typing up a business plan on his laptop as they drove, and phoning angel investors on his cell phone. But I think that Bezos is a smart guy. I think it's highly likely that things were slightly more planned out than that. For one thing, he had already flown out to California to recruit software engineering talent.

And for another thing, Bezos likely knew the destination of his cross-country car trip would be Seattle. There are several reasons for this. Seattle had the advantage of being a tech hub. It was home to Microsoft, of course, and thus filthy. with tech talent. But Seattle was also a six-hour drive from a major warehouse that book distributor Ingram had in Rosenberg, Oregon. Obviously, this would be key for a online bookstore to be close to a major distributor.

And finally, Washington State is not nearly as populous as California. No doubt in his research, Bezos had probably realized that a company didn't have to to charge sales tax unless it had a physical presence in the state that a customer ordered from. So Washington state being less populous than California was a major plus, so many fewer customers that would have to pay sales tax. Other locations Bezos reportedly considered for these benefits of tax purposes were Portland, Oregon, Boulder, Colorado, and Lake Tahoe in Nevada. The company that would become Amazon was founded in the summer of 1994 in the garage of the home that Jeff and McKenzie Bezos rented in Bellevue, Washington, at 10704 Northeast 28th Street. Jeff and McKenzie were the founding employees, along with a couple of programming talents that Jeff had recruited earlier.

One was Shell Caffin, who would go on to write much of the initial structure that would become the Amazon site, and who many people think of thus as a co-founder of Amazon in All But Title. The other was a programmer who had been on staff at the University of Washington's computer science and engineering department, Paul Davis. The garage they were working in was tiny, uninsulated, lacked adequate electrical hookups, considering the amount of workstations and servers it would soon house, and the desks were made out of doors purchased from Home Depot. It soon became a sort of tradition for new hires at Amazon to construct their own door desks upon joining the company, thus explaining that reference I made to Jane Slade in her interview

episode. As Paul Davis would remember, quote, it was just Shell, myself, and Jeff in an office, sitting around a table with a whiteboard and discussing how to divide the programming work, end quote. Mackenzie Bezos set up the company's bookkeeping operations, such as they were at the time, using Peach Tree accounting software, and helping with the hiring. When the group needed a change of scenery for meetings and such, they would convene at a cafe inside a nearby Barnes & Noble, something that Jeff Bezos has delighted in telling reporters about over the years. So at this point, Bezos and company know they're going to set up a bookstore online,

but of course none of them have any previous experience selling books. Bezos's solution to this was quite practical. In September of 1994, he took a four-day course on bookselling, sponsored by the American Booksellers Association in Portland, Oregon. He was also studying the competition at the time, such as it was, because, believe it or not, there already were a couple of online booksellers. Probably the first bookseller on the web was the famous Silicon Valley chain of stores called Computer Literacy Bookshops.

On August 25, 1991, computer literacy registered the domain, CLBooks.com, and began using the domain to take special orders from customers. They just didn't publicize it very much, because at that point, commerce on the internet was still forbidden. Another company registered books.com

on October 9, 1992, and that was the Cleveland, Ohio-based Bookstacks Unlimited. By 1994, at least to the new crew in Bellevue, Washington, books.com, with its 400,000 available titles, was the target to beat. As Paul Davis remembered, quote, as crazy as it might sound, it did appear that the first challenge was to do something better than these other guys, end quote.

one of the first big decisions was to come up with a name for the website and the company. As we've seen with Yahoo, the right name and the right domain name, could go a long way to defining a new company on the web. A famous Star Trek fan, of course, Jeff Bezos originally kicked around the idea of make-it-so.com, after Captain Picard's famous directive. Relentless.com was also, considered as a way to suggest that the company would be relentlessly focused on customer service.

But that idea was rejected as sounding a little too menacing. For a long time, a strong contender was cadabra.com, but Shell Caffin especially disliked that name, claiming that it sounded too close to cadaver. Other ideas kicked around were browse.com and bookmall.com. Finally, a major consideration at the time when it came to naming was that most web directories put sites into a category, which was then sorted in alphabetical order. And so, ard.com was strongly considered, A-A-A-R-D, also Awake.com. Finally, Bezos himself settled on Amazon.com.

As Bezos would later say, quote, this is not only the largest river in the world, it's many times larger than the next biggest river. It blows all the other rivers away, end quote. The Earth's biggest river, the Earth's biggest bookstore. The URL was registered on November 1, 1994, and interestingly, we should note that the company always referred to itself as Amazon.com, rarely just Amazon. At the time, this was controversial internally and probably even nonsensical, but in a few short years when dot-coms became a thing, it looked brilliant.

Amazon was consciously trying to be the first.com. Unlike other startups that we've spoken about, who shipped quick and dirty, Amazon took a lot of care testing its product and its setup before going live. Books were ordered from Books.com to test out that competition's operations. Bezos himself tested e-commerce as it existed at the time, for example, by purchasing network routers from the Internet shopping network. The site they were building was written in C and Pearl by Caffin and Davis and was hacked together in unison with a database and customer record system

that the team built on their own, because at the time there were no off-the-shelf packages that existed. In addition, systems for order processing, order tracking, and inventory management were also all hacked together in-house because, again, nothing existed that would yet suit the e-commerce model they were building. A couple of interesting things should be noted about the early Amazon model as it was developed. Early on, the earliest of early on, the Amazon team focused more on doing business over email than the web. Remember, this is late 1994, so most people are going online via things like CompuServe and AOL.

And most of those users probably didn't have access to the web yet until CompuServe and AOL turned on true Internet access as a feature. a few months later. But most users at the time did at least have an email address. So at that point, there were probably 10 times more email users than web users. Bezos wanted to go where the customers were, and as Davis recalled at that point, quote, Jeff thought that email would probably be more important than the web, end quote. So a lot of effort went into programs that would parse customer requests in email for particular titles and email-back automated results

that would hopefully match what the customer was looking for. This early search over email system was later ported over to the website when, by 1995, it was clear that the web was definitely becoming the more viable platform. Secondly, we can note that the initial way that Amazon operated was on a sort of zero inventory basis. Remember, the D.E. Shaw brainstorm for the Everything Store concept was to be a middleman, but not a merchant necessarily, a middleman that facilitated transactions, taking a cut, but not necessarily fulfilling the orders completely. This is interesting to realize, given Amazon's later reputation for warehousing, logistics, and fulfillment mastery. But at the time, the company didn't have the resources for a proper warehouse.

So the way the initial system was set up to function was like this. Amazon's entire database came from CD-ROMs entitled Books in Print, sent out by R.R. Broker of New Jersey. This was basically the book industry's Bible, the source that every bookseller in the country, large and small, used to order books. When you went to your local bookseller and asked for a specific title to be special ordered, books and print was the resource they referenced to see if they could accommodate you. All Amazon did was take this resource, cutting out the middleman,

and make the search function available directly to online users. This is a classic case of the internet cleaning up a tiny inefficiency. Could RR. broker have simply put their catalog online and done what Amazon did, probably, but they apparently weren't interested in doing that at the time, and Amazon was. So Caffin and Davis manually loaded the books and print CDs into their own database. This was a time-consuming process, actually, that Shell Caffin compared to emptying a swimming pool with a drinking straw. The process took an entire day just to load the weekly updates. Amazon supplemented this with inventory data,

from the two major book distributors, Ingram and Baker and Taylor. Some attempts were made to get supplemental categorical data from the Library of Congress, but this was a technological dead end at the time. Kaffin and Davis created their own ad hoc database, building off the open source Berkeley DB package, mixed with the Unix-based M-Map. When a customer successfully found a book that they were looking for, Amazon, as the middleman, simply ordered the book, took delivery of it temporarily, and then turned around and shipped it to the customer. Again, there was nothing to it. It's just, instead of having a physical bookstore that you went to to pick the book up at,

Amazon, as the middleman, shipped the book to you. Not to get all old man waves cane and back in my day on you, but if you're thinking here that there was nothing really revolutionary that Amazon is doing, then you probably weren't around in the early 90s. One of the reasons that big box stores like Barnes & Noble and Borders took off in the 1980s and early 1990s is that they could offer several times the selection that mom and pop stores and existing book chains like Walden Books, which existed in the... small sliver-like stalls in the shopping malls of the 1980s could offer themselves. This was true for all retail at the time. I remember the first time I ever visited New York City

in the late 90s. The first place I went was not the Statue of Liberty, but to the Virgin Megastore in Times Square because I knew they would probably stock the obscure album by the specials that I was looking for for several months. If things like Barnes & Noble, virgin and best buy were a revelation to us in the 90s, it was simply by virtue of having more inventory. So Amazon turned this up to 11 by offering all the inventory in the world, every book in print, well, almost. It might be hard to imagine it now, 20 years on, in this everything on demand when you want it world, but at the time what Amazon was promising, though seemingly obvious and simple to us now, was absolutely and self-evidently

revolutionary to users at the time. Either Jeff Bezos understood this instinctively, that this idea of a long-tail, truly everything's store, was a key differentiator and an innovation, or at least he must have realized it eventually as Amazon conducted its semi-private beta test. and the first orders started trickling in. Because, sure, Amazon offered the bestsellers, and to be competitive, they heavily discounted the entire bestseller list, but what they saw over and over again

was that people jumped on the opportunity to acquire books that they otherwise couldn't have. The first-ever order on the beta test, and thus the first ever Amazon order, was to a former co-worker of Shell Caffin, named John Wainwright, who Caffin had invited to the beta test. He ordered the book Fluid Concepts and Creative Analogies by Douglas Hofstadler on April 3, 1995. If you remember the episode that we did with the co-founder of Excite, Ryan McIntyre,

Douglas Hofsteadler's more famous book is Goader Escher Bach, which McIntyre used to explain the college major of symbolic systems, aka cognitive science. Again, it was long-tail titles like this that really showed the revolution that Amazon was capable of. Plus, Amazon's early customers were probably classic early adopters, so they maybe weren't as interested in the mainstream bestseller lists anyway. In fact, the best-selling title for Amazon's entire first year of existence was a book called How to Set Up and Maintain a Worldwide Website, the Guide for Information Providers by Lincoln D. Stein. It was, in fact, this long-tail revolution that Amazon was able to exploit while it was in the

beta test and working out one of the biggest kinks that it encountered during its beta test. test. It turned out that book distributors required retailers to order a minimum of 10 books at a time. But during the beta, of course, Amazon didn't have anywhere near that sort of sales volume. I'll let Jeff Bezos himself explain how they got around this little kink. Quote, we found a loophole. Their systems were programmed in such a way that you didn't have to receive 10 books. You only had to order 10 books. books. So we found an obscure book about lichens that they had in their system, but was out of stock. We began ordering the one book we wanted and nine copies of the lichen book. They would ship out the

book we needed and send a note that said, sorry, but we're out of the lichen book, end quote. Amazon didn't carry a million book titles, as their early promotional copy claimed, but it could get you a million different titles. So that's where it had its value as this new kind of middleman. And again, it's striking to compare what Amazon promised then as compared to what they are known for today. It usually took a week for Amazon to deliver most books to customers. The books were shipped to Amazon, and then Amazon shipped them to the customer.

More obscure titles could take as much as a month to deliver. Remembers Paul Davis? We came up with a phrase, almost in time delivery. In other words, we don't have the books you want, but we can get them real soon. End quote. Again, just the ability to order anything and everything was the amazing, revolutionary idea that Amazon launched with. As initially projected,

a lot of the early orders came in over email, and half of all the early customers phoned in their credit card numbers. And actually, we hadn't even mentioned that key aspect of e-commerce yet, the actual electronic payment part. As mentioned previously, early e-commerce customers were wary

of using their credit cards online. So, again, Caffin and David, hacked together their own solution to the problem. The order system was housed on two separate computers at Amazon. These computers weren't connected to the internet. When an order was placed on the one computer and the credit card number was recorded, someone on the Amazon team removed the floppy disk for the first computer and literally walked it to the second computer, which would then be connected via modem to the credit card

processor, and the card was charged. This was known within Amazon as sneaker net, as in actual sneakers, shoes, walking the credit card number from one computer to the other. The only way a customer's credit card number could be stolen was by physically being in the Amazon offices and interrupting the sneaker net flow. Said Paul Davis, quote, one of the reasons I designed it that way, was so we could specifically claim that it's secure because it's off the internet, end quote. These systems were, of course, brought online eventually, and sneaker net was retired,

once Netscape introduced and popularized secure socket layers. Amazon's full website launched to the public came on July 16, 1995, And when it went live, it went live with some key innovations that we take for granted today, but which Amazon absolutely pioneered. For one thing, think of the basic user interface of e-commerce. Think of the shopping cart, the virtual shopping basket. Kaffin and Davis invented that metaphor. If a user is shopping your site, they might have several things that they want as they browse around.

You don't want them to have to begin the checkout process for each item that they want to order, so you need a virtual place to store the items they want to buy. You want a virtual shopping basket. From a technical perspective, remembering a given customer from one visit to the site to another visit would prove to be useful, of course. Maybe you could make it so that the customer didn't have to log in every time. Maybe you remembered what that customer had browsed or even possibly ordered previously, or more importantly, almost ordered, so you could store that information and prompt the customer accordingly. Caffin and Davis set up Session Identifier CGIs to do simple things like store shopping cart contents from one page to another,

and though they were not the first to develop this, they were certainly the first to popularize this sort of innovation in an e-commerce context. And then there was the brilliant innovation of product reviews. And this really was an innovation when you think about it. Prior to the internet, can you remember any other retailer offering reviews of the products that they're selling? A supermarket, for example, doesn't say one brand of toothpaste is higher rated by its customers than another. The initial rating system was hacked together by Shell Caffin one weekend and was initially designed to provide editorial content from Amazon itself,

but eventually expanded to allow reviews from anyone and everyone. This caused a bit of controversy for Amazon at the time, as obviously what author wants bad reviews of their book posted prominently right alongside that book on its sales page. But to their credit, Amazon would stick to its guns, believing that honest reviews, and generally just being an honest broker who could be trusted to help the customer make the best purchasing decisions, would again be a key differentiator. In a sense, it was a way of trying to provide the same level of customer service that a local book merchant might be able to provide.

You know, just simply recommending a good book for you to read. And this, of course, grew into the famous Recommendation Engine, which was absolutely another key Amazon innovation, and one of the main things that e-commerce could do things that traditional commerce found difficult. Tying in with the Session IDs system, the recommendation system would combine your own browsing history, your own purchasing history,

as well as the purchasing history of everyone else, on Amazon to help give users that classic, if you liked X, then you will probably like Y prompt. Today, this is such a key component of not just e-commerce, but things like Netflix and Yelp restaurant reviews, that it's easy to forget that Amazon gave us this whole review slash recommendation revolution first, and some might argue they still do it the best to this day.

But that innovation would come down the road away after Amazon worked out more of the kinks in its system over its first year in business. And we'll get to those kinks in our forthcoming Part 2 episode on Amazon, where we will follow the company all the way into the dot-com era and through to the point of the dot-com bust, by which point Amazon was well and truly the king of e-commerce