How the Internet Happened: From Netscape to the iPhone

by Brian McCullough

In the age of ecommerce, advertisers could even measure clicks that led directly to a sale. This made it easier for advertisers to calculate their return on investment by orders of magnitude.

This inscrutability combined with the web’s vastness and anonymity presented a tree-falling-in-the-woods sort of problem. Anyone could now publish anything; but if you did, how would anyone ever know about it? And so, necessity dictated that search engines would become the most popular and most important early websites. And because the problem of a business model had been solved by HotWired, search sites, and Yahoo in particular, would become the web’s first great companies.

To this day, a “search engine” is actually a database of website copies. The search engine sends out “spiders,” which are computer programs that go out onto the web and find new web pages. The spiders locate the pages and then copy some or all of the code into the search engine’s own database. When a user searches a search engine, they’re not actually searching the web itself, but are instead querying the database of copied webpages the search engine has compiled.

the algorithms the early search engines used to sort and rank pages were crude and wildly ineffective. The obvious alternative to this state of affairs was to bring a curatorial element to search. And in fact, the dominant player that would emerge in search was not strictly a search engine at all, but a directory, compiled not by bots but by actual humans.

the two began collecting and trading links to the new websites they found. They started compiling these favorite links into a list, each trying to outdo the other by finding the coolest new site of the day.

Yang and Filo settled on the name Yahoo!, which they claimed stood for Yet Another Hierarchical, Officious Oracle.

at least initially, the university was a generous host of Yahoo’s traffic and content, free of charge. When Netscape launched its beta browser late in 1994, it decided to make Yahoo the default link when a user clicked the DIRECTORY button on the top menu of the browser.

Yahoo, on the other hand, was a service; a destination; a directory; a glorified list. There was almost nothing proprietary about it. Furthermore, it was a service that you could never charge for.

Moritz was a general partner at the VC firm Sequoia Capital. Sequoia had funded such Silicon Valley luminaries as Apple, Atari, Cisco and Oracle, but it had not yet dipped its toe into Internet waters. The vision that Moritz used to argue Yahoo’s case was the one put to him by Yang and Filo.

Sequoia eventually bought the pitch. Yahoo already had an audience of millions, and if the web kept growing at its present rate, who knew how many hundreds of millions could be reached in the near future? By that logic, even the wacky company name could be seen as a plus. After all, as Don Valentine, the legendary founder of Sequoia, put it, “A long time ago, we helped finance a company called Apple.”13 Sometimes investments in companies with silly names could turn out handsomely. In April 1995, Sequoia invested $1 million in exchange for one-fourth of the newly incorporated Yahoo. By early 1999, Sequoia’s initial $1 million was worth $8 billion.

For his part, Jerry Yang was confident that Yahoo had one unique advantage: it had been first. It would become an article of faith during the dot-com era that being early to market on the Internet frontier conferred a magical “first-mover advantage” on whomever was so fortunate.

Yahoo did something that was completely radical for the time: advertise on TV and radio. Yahoo was the first Internet company to market itself via mass media. With zippy, hip ads, matching the slick name and the brash image of the site overall, Americans found themselves being asked “Do You Yahoo?” Yahoo quickly became one of the Internet’s most recognizable names, familiar even to the vast uninitiated Americans who were not yet even online.

That was because there was one looming problem: for all the dollar signs in Yahoo’s bank account, there wasn’t actually much on Yahoo’s bottom line. In its first quarter as a public company, Netscape had recorded revenue of $56.1 million.29 By comparison, in its first quarter as a public company, Yahoo could report revenue of only $3.2 million.30 Even that was better than 1995, when Yahoo reported revenue of only $1.4 million for the entire year.31 Again, if Netscape had gone public with questionable revenues, Yahoo had taken things to the next (more speculative?) level.

But Bezos quickly decided that an everything store was a bit too grandiose, and so he instead began investigating product categories that might be suitable as a proof of concept. He weighed roughly twenty different possibilities, including computer software, office supplies and CDs. He settled on books as the best test case because, as Brad Stone put it in his history of Amazon called, not coincidentally, The Everything Store, books were “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.”

Books also had an advantage over something like CDs because, at that time, there were only two major book distributors that every bookseller in the country worked with, Ingram and Baker & Taylor. This compared favorably to the several major and hundreds of minor record labels in the world. 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 on CD.2 No single store could shelve all those titles. But an online store could. As Bezos himself would later say, “With that huge diversity of products [titles] you could build a store online that simply could not exist in any other way.”3

Bezos had already flown out to California to recruit software engineering talent. And according to multiple accounts, he likely knew the destination of his cross-country car trip would be Seattle. His careful research had shown him that Seattle had the advantage of being a tech hub—home to Microsoft of course, and thus filthy with tech talent—and also that it was a six-hour drive from a major warehouse that book distributor Ingram operated in Roseburg, Oregon. Also, Washington State was not nearly as populous as California. No doubt, his research had also led Bezos to realize that a company did not have to charge sales tax unless it had a physical presence in the state a customer ordered from. So, Washington being less populous than California was a major plus.

Initially, Amazon would take a catalog of available books entitled Books In Print, sent out by R. R. Bowker of New Jersey, bring it online, add some search functionality and allow customers to find the books they wanted.

You need a virtual place to store the items customers are considering. You want a virtual shopping cart. Amazon popularized this metaphor. From a technical perspective, remembering a given customer from one visit to the next is a useful thing. Amazon remembered what a customer had ordered previously—or almost ordered, before abandoning their cart—so it could store that information and prompt the returning customer accordingly.

based on our different shopping histories. Amazon was one of the first sites to tailor its storefront individually in this way.

And then there was the brilliant innovation of product reviews. Prior to the Internet, few general retailers offered reviews of the products they were selling. A supermarket doesn’t say one brand of toothpaste is higher-rated by shoppers than another.

Tying in with the cookies and session ID systems, the recommendation engine would parse 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 prompt: if you liked x, then you will probably like

The true turning point for the company came when Amazon was featured on the front page of the Wall Street Journal on May 16, 1996. Under the headline “Wall Street Whiz Finds Niche Selling Books on the Internet,” the Journal described Bezos as “a whiz-kid programmer on Wall Street” who “suddenly fell under the spell of one of the iffiest business propositions of modern times: retailing on the Internet.”

Amazon didn’t exactly trounce its initial competition. Barnes & Noble is still around (though Borders is gone). Bookstores are still around, unlike, say, video rental stores or music stores. Amazon didn’t surpass Barnes & Noble in total revenue as a company until 2004.33 Amazon didn’t even become the biggest book retailer in the world until 2007.34 But it didn’t really matter, because just as had been the plan all along, while the incumbent booksellers raced to copy Amazon, Amazon was already moving toward new horizons. Jeff Bezos didn’t care if Amazon ever definitively “won” in books, frankly, because his real aim was to take increasingly bigger bites of other markets, and then other markets, and other markets, until one day, Amazon had a piece of every market.

Bezos wasn’t just thinking about books, but about retail itself, a business model that went back millennia to that first day merchants gathered in a central location to hawk their goods to a local population. In Bezos’s vision, the products would come to the people. First books, then anything else. In the end, he would make the everything store a reality.

There was no way of knowing. In a perfect marketplace, the market price is the correct price because buyers and sellers (ideally, multiple buyers and sellers) can haggle to arrive at an optimal result. Classified ads did not allow for that haggling. But what if you could create an auction scenario in classified ads? That way you could find the true market price for any item because the buyers and sellers would arrive at the final price organically.

Why eBay? Well, after cashing out from the eShop sale, he had done some web consulting and freelance work and decided to do so under the rubric Echo Bay Technology Group, a name he simply liked. However, the domain EchoBay.com was taken, so he registered what he considered to be the closest approximation: eBay.com.

But AuctionWeb’s immediate success was also due to structural decisions that would enable the service to scale successfully. In short, Omidyar enabled AuctionWeb’s community to organize itself.

Another way to help the system regulate itself was the Feedback Forum. This was a public online message board where users were encouraged to leave written feedback about other buyers or sellers, in addition to a numerical rating: plus one, minus one or neutral. Once a user’s rating on the feedback forum surpassed a negative four, they were banned from the site.

Omidyar had accidentally stumbled upon one of the longer-term factors in AuctionWeb’s eventual success. A focus on community, on empowering the users and allowing them to function autonomously would prove to be absolutely vital.

eBay was online commerce, but not in the way that Amazon was; it was a platform, but not like the operating system or the browser were. eBay was nothing more than a virtual marketplace, and by being virtual, it didn’t actually do anything other than facilitate the interactions between buyers and sellers. It didn’t store goods. It didn’t ship goods. It didn’t even guarantee the exchange of goods between buyers and sellers! The one truly tangible thing that eBay had was the goodwill of those buyers and sellers and the community they were creating—on their own—to make the buying and selling happen. eBay would be one of the first web companies to understand that all the value of its service came from the users and their community. eBay’s only asset, in fact, was its users, and therefore the only important thing for the company to do was to make sure the buyers and sellers were happy so that they would keep coming back.

This is a key evolution. In so many ways, over the last twenty years, the web and the Internet have slowly trained all of us to get comfortable interacting with crowds and, often, crowds of strangers. eBay was one of the first websites to show that a largely anonymous community, carefully constrained by a few guidelines and regulations, but invested in a system of online reputation, could actually work.

There were several factors leading to this explosion in growth. For one thing, eBay noticed the power of Januarys: they came after the holiday season. That meant millions of people with millions of unwanted gifts. eBay to the rescue.

eBay was creating not just the world’s largest virtual marketplace, but also the first marketplace that could rival the real world. Just as the Internet allowed people to connect to the entire world, eBay allowed a person to sell to the entire world from their tiny little corner of it.

The very act of using Hotmail helped spread the word about Hotmail. This kind of practice is now called viral marketing, the technique of promotion by rabid user word of mouth. Today, this is the very foundation of modern marketing strategy; in Hotmail’s era it was very much new and revolutionary.

Microsoft, who, on New Year’s Eve 1997, purchased Hotmail for $400 million in stock. Not bad for two years of work, and an idea that even its founders thought was so obvious that anyone could have done it.

“You’re a search engine—once they’ve done the searching, why do they need you?”34 Yahoo needed to find a way to keep users on its pages. To use a watchword that was ubiquitous at the time, Yahoo needed to get more “sticky.”

investment. And if they offered, say, airline listings, the search portals discovered they could collect lucrative promotional fees as, obviously,

But if you weren’t content with merely doubling your money on a solid, staid stock like Procter & Gamble, then, by 1998, you might start to look enviously at the returns tech stocks were ringing up.

In the twelve months of 1998, Yahoo stock returned 584%, AOL 593% and Amazon 970%.5 These were three of the best-known, most talked-about stocks of the mid-nineties, widely heralded as the vanguard of the new economy that the Internet was supposedly bringing into existence. They were hardly needles in the haystack.

Dot-com companies were young. They were going public sometimes only months after their creation. When they showed any sign of growth, their stock prices took off because it seemed to validate the notion that there was only more growth ahead. And it was that limitless promise that led to the second unique feature of Internet stocks: the profits didn’t seem to matter. Valuations weren’t tied to things like, you know, income. They were tied to potential fortunes to be made, somewhere in the future. New metrics like counting “eyeballs” and “mind share” were used to show companies were growing, even if that growth couldn’t be measured in dollars and cents. Heck, sometimes a dot-com stock would increase in value even after it announced losses!

Surely that meant that things were functioning better, more efficiently, more profitably. The only problem was, none of this seemed to show up in any of the official numbers. Economic output is easy to measure when you can count widgets coming off an assembly line. But when your “economic revolution” is built around thoughts and ideas, and the speedy new ways you’re connecting them all together, how do you quantify the value of those innovations? ATMs might mean fewer bank tellers had jobs; but think of the time saved by millions of consumers! How did one measure that?

Many people, then and now, feel that Greenspan, at the very least, enabled the dot-com speculative stock market bubble. At the time, American investors came to believe very strongly that Greenspan wanted them to be rich, and if anything went wrong, Uncle Alan would put his finger on the scales and make things right.

By the late ’90s, everyone involved in the stock market seemed to be winning. And the coming of the dot-com stocks only seemed to extend this winning streak. Nobody had any vested interest in questioning the madness, least of all the media. As early as 1997, an estimated 30% of national newspaper ad revenues came from the financial services industry.

In 1997, the first of the real “dot-coms” came to market, totaling 19. In 1998, there were 29. But in 1999, there were 249 Internet IPOs. And those were just the Internet companies that debuted on the stock market. There were untold others that got acquired or went nowhere. It was perhaps inevitable that, toward the tail end of the bubble, there were a lot of young Internet companies being founded that had questionable business plans at best. Some of the companies were so flimsy as to be just short of outright fraud. Investors (both venture capitalists and the public at large) no longer had any interest in discerning true value; any company with a .com at the end of its name might be the next billion-dollar winner.

SO MANY OF THE COMPANIES that would embody what we think of when we remember the dot-coms shared some or all of Priceline’s traits: a business plan that promised to “change the world”; a Get Big Fast strategy to reach ubiquity and corner a particular market; a tendency to sell products at a loss in order to gain that market share; a willingness to spend lavishly on branding and advertising to raise awareness; and, above all, a sky-high stock market valuation that was divorced from any sort of profitability or rationality.

This was a not uncommon problem for the ecommerce players. A startup named Furniture.com raised $75 million only to learn a lesson that Ikea had known about for years: you can’t exactly send a couch via UPS. “There were many cases when we would get an order for a $200 end table and then spend $300 to ship it,” a former Furniture.com engineer would admit. “We never could figure it out.”

In order to entice customers, it often waived the delivery fee that was crucial to covering costs. In essence, Webvan tried selling groceries at a loss in order to achieve scale. But that was standard practice at this point, and in no way prevented Webvan from enjoying a typically buoyant IPO. When it went public in the fall of 1999, the company had recorded only $4 million in revenue in its entire existence. Nonetheless, the stock went out at $15 and rose to $34 before ending the day at $25. Webvan had an $8 billion valuation.

Kozmo would send it to them for less than it would cost to buy at the local bodega across the street. And Kozmo still had to pay the army of bike couriers who made the delivery. It was retail without the overhead of real estate, sure, but what about the costs of warehousing, of labor, of the website and logistical back-end systems?

For a couple of years there, it seemed like everyone was begging us to buy cheap stuff, subsidized largely by generous, unseen piles of venture capital money. Ironically enough, far from engendering customer loyalty, consumers tended to treat the dot-coms as a fly-by-night bonanza, taking the deals when they presented themselves, but often not repeating the experience.

DOT-COM COMPANIES FELT they had to spend in order to brand themselves like Yahoo had done. They felt they had to be first to their particular market in order to lock in customer loyalty, just as Amazon had done. They spent because they felt they had to be the first in their category to IPO, like eToys had.