The
Definitive
oral
history
of
online
travel

Dennis Schaal joins host Hannah Sampson — with audio clips from several interviews — to talk about the oral history, what surprised him about the project, and what innovations he thinks will shake up the industry next.

Listen closely for insight into what really went down in the decade that shaped the future in ways we may not yet understand.

By the early 1980s, well more than a decade before the arrival of online travel agencies such as Travelocity, Preview Travel, Expedia, Priceline.com, and Hotel Reservations Network in the mid-to-late 1990s, American Airlines' Sabre unit had created a pre-Internet, direct-to-consumer booking tool for flights, hotels and cars called eAAsySabre. The awkward spelling of the product reflected American Airlines' parentage, which stuck "AA" anywhere it could. Throughout this oral history of the early days of online travel, the founders of companies such as Sabre-developed Travelocity, Hotel Reservations Network (later Hotels.com), Expedia, ITA Software, and Orbitz kept bringing up eAAsySabre as a precursor to, or inspiration for, their subsequent online travel sites and entrepreneurial efforts.

Kathy Misunas, former CEO of the Sabre Group and current consultant and advisor: eAAsySabre was not easy at all but it was easier than anything else that was available at the time. It was a request/reply-type system. We're in the early 1980s, a long time ago. Think back before any type of Internet. There were some private networks and some public networks like CompuServe or Prodigy, and Minitel in France. There were a few. Our product [eAAsySabre] allowed somebody who wanted to work through one of those networks to be able to answer questions. It would come back and say, 'Where do you want to go?' The person would type in, 'San Francisco.' It would say, 'When do you want to go?' They would type in, 'June 3' or whatever. 'How many people are going?' The same questions that an agent would ask.

Consumers could make reservations for flights, hotels and cars but airlines or travel agents had to do the ticketing. Sabre, which provided computerized reservation system [now known as global distribution system] services to travel agents, had to walk a fine line between this consumer-oriented eAAsySabre product and Sabre's travel agency customers. Travel agents resented the idea of consumers and airlines transacting directly and cutting them out of the selling and booking equation. It was a harbinger of the juggling Sabre would have to do for nearly two decades after launching Travelocity in 1996 and before selling it to Expedia Inc. in 2015.

Misunas: Way back then we had two different paths going on. We were totally supportive of the travel agents but we knew at some point, some day, consumers would be doing their own thing. We played with this, and I would get all sorts of questions from agents in seminars because they would hear about it and say, 'What are you doing here?' We'd say, 'How many people in the audience have heard about this? One, OK, so let's not worry about this. Let's move on.'

Then what happened was, on that platform, we decided that it didn't have to be request/reply. We would put up a template and it would be fill in the blank. Technology was starting to change. We could convince CompuServe and Prodigy to allow those types of templates. At the same time, travel agents were starting to really have a different type of automation on their own desktops.

IBM introduced a personal computer in 1981 and the first Macintosh computer showed up about three years later. eAAsySabre was evolving although the idea of an interactive Internet had yet to take hold.

Misunas: We weren't prescient enough back then to know that it was going to be the Internet in 1995. We knew that people were adapting a different way of purchasing and we felt, for travel, we had the tools. That's when we started saying, 'OK, let's put hotels in here. Let's put cars in here.' The things we did already had been available to travel agents. We started testing those. We started putting in other product. Then in the early 1990s, we created a department called Sabre Interactive. That's when we knew the Internet was going to become public in our lifetime.

In 1991, as eAAsySabre was evolving, Bob Diener and David Litman decided to get into the hotel business and within a few years developed a business model that would revolutionize online hotel selling.

It was at a time when travel agencies were mostly interested in selling airline tickets.

Bob Diener, co-founder of Hotel Reservations Network, which became Hotels.com, and currently president of Getaroom.com: Hotel Reservations Network was our original company. We started it in 1991. Well, I was recently in the airline business. I was an airline consolidator. I sold my airline business. After about 6 months of sitting around the beach and being bored, my partner Dave Litman and I, we were talking about what we were going to do next. So we went down to Belize and spent a few days in a hut. Between scuba diving, talking about what we were going to do next and a lot of paper crunching, we came up with the hotel business.

We saw a great opportunity in the hotel business because the state of the industry then was that occupancy on average was about 60 percent. Hotels needed about 70 percent at the time to break even. We're in a recession, and we saw great opportunity to negotiate with hotels to offer, #1: consumers better rates and, #2: there was really no central source other than travel agents to find availabilities and rates on hotels.

Most travel agencies didn't want to book hotels because it wasn't profitable, and most people just picked up the phone at the time to book the hotel. At the time, the airlines were the ones that paid the big commissions, and hotels that actually paid the commissions paid 5 to 10 percent but you had to chase them.

This was still pre-Internet and online travel so Diener and Litman's Hotel Reservations Network started out taking consumer bookings over toll-free numbers but it was an uphill battle collecting commissions from hotels.

Diener: We learned after a very few months that it wasn't a very good business because very few hotels were actually paying us. We ended up spending 90 percent of our time as a collection agency and 10 percent of our time expanding the business. We brainstormed on how we were going to change this. It was our first major pivot. We came up with what we called the merchant model, where basically we approached hotels and said, 'It's much better if we get paid commission in advance.' Why don't we pay hotels a net rate and we'll sell the hotel at a gross rate. The customers will pay up-front and we could collect the commission in between.

Tim Poster, who founded Las Vegas Reservation Systems (Travelscape) and was former owner of the Golden Nugget and COO of Wynn Las Vegas: I started my company in 1990. We were just Las Vegas. I grew up in Las Vegas. Las Vegas back then was very, very different than it is now. It was very much of a who-you-knew sort of thing. We ended up doing a deal together, Diener and I, because he couldn't get into Vegas. We had Vegas kind of locked up. We had the merchant model. We did that. Again, it's a little bit difficult to say that we were the first because there were other people doing the merchant model, but not for just hotel rooms. Tour companies were doing it with airfare and big packages with show tickets and maybe even car rentals and Grand Canyon tours and things like that.

My idea was to just buy the rooms wholesale, get room blocks and go direct to the consumer because the margin was so much greater. I was able to do it because I was a local guy and I had connections with the hotels even though I was pretty young. We definitely were the first ones to do that in Las Vegas. As much respect as I had for Diener, I definitely did not want him coming into Vegas. I did what I could to keep him out.

This was in the early 1990s, maybe '93 or '94, something like that. I met Diener in Miami and he said that they had no interest in selling. I think they more or less told me that the company was worth way more than I probably could afford, although I don't remember any numbers being mentioned. What he was interested in was, of course, getting into Vegas. I said, 'Well, look, there's one way that you can get into Vegas. One way for sure you'll get into Vegas.'

By then I had a lot of juice with the hotels in Vegas because I was giving them so much business and the hotels weren't going to do anything to get sideways with me. I said, 'Look, if you really want to get into Vegas, I have the inventory and I can get more. What we can do is, I can basically give you blocks of rooms and I buy them from the hotels and I mark them up and you buy them from me.' He agreed. He said, 'OK.' That's what we did.

Diener: There were several problems with that [merchant] model. Number one, people we talked with about it said, 'Well, who's going to pay in advance? Who are you?' We realized if you give consumers a good deal, they will actually pay in advance. That was challenge number one. There was no pre-paid model. At the time, consumers went to the hotel and you paid at the hotel. The second issue was getting terms from hotels and having hotels bill us versus just having the consumer go and pay directly by credit card.

Diener personally visited the now-long-gone Dorset Hotel on W. 54th Street in Manhattan and it became the first hotel to submit to Diener and Litman's then-nouveau merchant model.

It is a business model that transformed online travel and the hotel industry. In July 1999, Barry Diller's USA Networks acquired Hotel Reservations Network, and in March 2003 bought Expedia.com, and the two merchant-model practitioners came to wield tremendous power over hotel distribution – much to the resentment of the hotel industry.

Diener: This was our first hotel and I actually went there personally and talked to the general manager and really pleaded with him to work with us. We were actually selling a lot of their rooms on a commission basis. He said, 'Bob, you have no references. You have no assets, but I'll tell you what: If you pay us weekly and promptly, we'll continue to give you terms.' That was our first hotel. Then we went to the next hotel and after months of working with the Dorset, we give the next hotel the Dorset as a reference. That blossomed and then all of a sudden we expanded to five other major cities and we're having thousands of other hotels giving us terms.

Diener said in 1995 he and Litman came to understand the potential of a new technology called the Internet.

Diener: We had a friend of ours knock on the door. His name was Dave Rae. He came in and he showed us the technology. This was before Travelocity and Expedia. Most people had never heard of the Internet. There were really no sites for commerce. The Internet was not interactive at the time. The only way to access the Internet was through a phone connection. It was a very slow connection. You couldn't send something and get an instant response. You had to send something, they would receive it. Then you'd have to respond separately. It wasn't an interactive back and forth. We were very reluctant, mainly due to the speed and the fact that it was not interactive. But Dave Rae said, 'Let's make a deal. I'll build you the site. I'll pay for it and you just pay me a 10 percent commission.'

So we said 'fine, that sounds great.'

At the time you could register any domain for about $16. We went and bought the name Hoteldiscount.com and Hoteldiscounts.com and that was our first website. It was a shocking yellow color, similar to the color that Spirit Airlines is using on their website and planes right now. But it stood out. And it turns out that we launched the site in late 1995, early 1996. And in 1996 it was about five to six percent of our business. We were really impressed with it.

It was still non-interactive. It was very slow so someone would come in and look at the site and request a room, and then we would have to look at it, book it, and respond back to them. It could take hours before there's a completed transaction, but it was a great source of business. We said, 'This is great.' As far as I know, I believe it was the first hotel website. Terry Jones can tell you about this: Sabre had something at the time called eAAsySabre. It was very difficult to use. Prodigy and AOL were really the two main Internet sites that were out there at the time.

We were so impressed with this that we went to Dave Rae and said, 'We want to buy the site from you.' He was busy with his gaming, which was doing well. He was more than happy to sell the site. So we bought the site. It was a minimal amount.

It was 1994 and Rich Barton, who went on a couple of years later to found Expedia out of Microsoft, had transitioned from stints as a product manager for DOS 5 and DOS 6 and working on the Windows 95 team within Microsoft to its hot and sexy consumer division.

At Barton's first product review meeting in the big boardroom he was to present to Microsoft co-founder and CEO Bill Gates; executive vice president of worldwide sales and support Steven Ballmer, and group vice president of applications and content Nathan Myhrvold.

But Barton, 27 at the time, had a surprise in store for the trio because he was going to tell Gates that his pet project in travel – to develop an Encarta-style and CD-ROM-based travel guidebook – was a dumb idea.

Rich Barton, Expedia founder and its first CEO; currently executive chairman of Zillow and a venture partner at Benchmark: I didn't say it that way, but I said, 'I just think this is small.' The whole book-publishing business is maybe a $100 million a year [opportunity] and if we got 100 percent of the business that didn't matter much to Microsoft. But in the course of that project and building the business plan, I did discover eAAsySabre because I was a geek using online services like CompuServe, Prodigy, and the like. This was before the graphical Web.

I discovered this thing called eAAsySabre, which was a command-line program with basically the old DOS interface. It was intended to be used by travel agents who weren't in their offices. If they were home and had to field a call from their clients and they had to deal with making changes to a trip. It was their online access to their Sabre reservations where they could look, shop or modify, etc. Sabre decided to make this user interface, this little program, available to the general public. They made it available to people like me who were using CompuServe and Prodigy.

I looked at that and I thought, 'Oh wow, this is cool. This is really cool.' Consumers don't want to have to pick up the phone to talk to a travel agent just to understand what the schedules and availabilities are. They don't want to have to pick up the phone to get pricing. They want to actually do it themselves.

I was one of those guys who was a frustrated business traveler myself, and I remember being on the phone with the corporate travel agent at Microsoft and wanting to jump through the phone, turn the computer towards me and do it myself. I could hear the click, click of the keys that the corporate travel agent was using so I knew she was using something on a computer.

Bill Gates and his colleagues in the Microsoft boardroom saw eAAsySabre and bought into Barton's vision. But what also enticed them is they saw a business opportunity in trying to replace a lot of those big-box mainframe computers that were crunching data at companies such as Sabre and major airlines and replacing them with PCs running Microsoft Windows NT software.

Barton: I demoed this for Bill Gates and Steve Ballmer and Nathan Myhrvold. I can't remember who else was in the room, and said to them, 'Look, I think we could do a Windows version of this and really blow the doors off of a travel product.' It didn't just help people with travel guide information, but actually helped people make and take and plan and purchase trips.

They loved it. They were familiar with Sabre, and they knew the travel industry was one of the biggest industrial consumers of mainframes for IBM and DEC [Digital Equipment Corp.]. It was our mission at Microsoft at the time to rebuild everything that a mainframe could do on a modern PC by Windows NT and other modern software and hardware products, so I posited that we could actually rebuild the reservations systems on Windows NT and build a better mousetrap. They got really excited about that for obvious reasons. We ended up doing that, rebuilding the whole faring system.

Gates indeed was all in and along the way he approached then-chairman and CEO of American Airlines Robert Crandall about the idea of Microsoft and Sabre developing an online travel site together.

Terrell Jones, former chief information officer, Sabre; founder and first CEO of Travelocity; former chairman of Kayak, and currently executive chairman, WayBlazer: There's another part of this story that Rich [Barton] may have told you that, during this period of time, Bill Gates visited major accounts with his sales person. He would come at least once a year. We were a big Microsoft shop, and he was interested in travel. He proposed to Bob Crandall that we do a travel site together, and that Microsoft would do the front end and Sabre would do the back end because they were good at UI and we were good at reservations. Bill and Crandall talked about the idea of a joint venture online service. eAAsySabre was already operating. Kathy Misunas and I spent a lot of time with Ballmer trying to negotiate that deal. Steve Zollo, who was running eAAsySabre, says he showed it and our Internet idea, not yet named Travelocity, to Rich. Microsoft eventually decided to go their own way as did Sabre, and Travelocity was born and later Expedia. A deal was not possible. We both wanted to drive the car.

Barton: That's the first I've heard of Terry's story. Or at least that I can remember.

My pitch to Bill [Gates] and Steve [Ballmer] was in the context of an Encarta CD-ROM for travel and I was telling them we shouldn't do it. My eAAsySabre demo was a small part of the meeting. I don't recall if Gates had already seen eAAsySabre, but I'd guess he had.

After we got the green light, Greg Slyngstad and I spent quality time with Terry and American Air and Sabre folks trying to do a deal for connectivity or co-development. I don't recall Bill and Steve being involved at all, but it's possible they chatted with Crandall. They only cared because AA bought so much mainframe hardware-slash-software, and Microsoft wanted to get PCs and Windows NT in there.

It was a losing battle. Terry didn't want to do a deal with Greg and me. He was pretty cocky and comfortable that Sabre/AA were in the driver's seat and didn't want to invite Microsoft into the party.

I wouldn't have either.

AA people were very dismissive of me, too. So, we spent a ton of time with Galileo, Amadeus, and Worldspan, as well. Ultimately, we partnered with Worldspan for a global distribution system connectivity, and away we went. GDSs were all functionally equivalent from our perspective. One of the early R&D projects for us that was so cool was Best Fare Search, which rewrote airline pricing on a modern tech platform. This was a game-changer for Expedia as it was something the GDSs didn't want to have happen and therefore didn't do.

After getting approval from Gates around 1994, Barton and his team started building a booking service but it wasn't going to be a website at first.

Barton: I thought of those guys [Gates and Ballmer] as my venture capitalists. I pitched them on starting it on the outside of Microsoft because I figured it was going to be a travel business, not a software business, and that's how we got going. That was in 1994, before Mosaic and Netscape. Netscape and Mosaic happened pretty shortly thereafter. We originally started building it for what was called MOS [Microsoft Online Services], which was a competitor to AOL, intended to be, and it was a proprietary online service. When Netscape and Mosaic happened, I and my team knew right away that these proprietary online services were not going to survive, so we pivoted in the middle of the development project, even before we released it, to the graphical Web.

[The team included] the names that you'd know who have stayed around such as David Beitel, who ended up being the CTO of Expedia. He was one of the first engineers we hired. He is now the CTO of Zillow. Robert Hohman was the development manager and engineering manager. He's now the co-founder and CEO of Glassdoor. Byron Bishop [became a senior vice president at Expedia] and Greg Slyngstad was the first GM of the project. Greg was there. Simon Breakwell was one of the first guys that we hired from the industry, and then Erik Blachford, who later became CEO of Expedia after I left.

The idea was to build the largest seller of travel in the world. That's kind of how I pitched it. They wanted me to do that one step at a time, of course, and we went out and we got a deal with Worldspan so we were able to access air, car, hotel in version one. It wasn't pretty, but in version one we shipped it on the Web in 1996, and we were in business. We weren't making much money, but it had caught on.

I mean nothing in Web 1.0 was all that pretty. A lot of clutter, a lot of design hangover from apps and multimedia CD-ROMs and desktop apps. It was like a lot of content stuff that turned out not to be that important. It was really slow. I don't know what the baud speeds were at that time.

We were transactional right from the start. We actually licensed a bunch of travel content as well, so we did have a bunch of travel guide information that we put up as well.

Expedia was not the first to make it to the public Web. Travelocity beat us by probably a year. Travelocity was out in front, and when we went public, as I recall, Travelocity was probably twice the size in gross bookings as Expedia was. I maybe misremembering a little bit, but that's probably about correct. Preview Travel was number three. Preview Travel, they were the AOL partner. The apps, if you will, the vertical destination sites to all these travel brands, they all had to have a Big Brother portal. Travelocity's was Yahoo, Expedia's was MSN, and Preview Travel's was AOL.

Terry Jones founded Travelocity within American Airlines' Sabre division in 1996 but its roots go back to eAAsySabre a decade earlier.

Jones: I was at Sabre, and you need to go back even farther than that, probably 10 years earlier, or maybe eight years earlier. Sabre, at the behest of Max Hopper, who was our CIO, the visionary guy at Sabre, really, thought that this online thing is going to be big. We should put Sabre on AOL and other networks like it. You may remember there was a product called eAAsySabre. It was fairly primitive, but it was on AOL and CompuServe and Prodigy and several other networks.

One of the issues with it was the user interface, of course, had to reflect the design of the host, so AOL and CompuServe thought that they knew all about travel. So they kind of forced us into their design, but it was successful and it had hundreds of thousands of users. Now you could make a reservation, but that reservation was then automatically sent to a travel agent and you picked up your ticket from them. For an an airline ticket and for hotels and cars. TWA had a product, as well, called something like Travel Shopper.

I've been deposed on this many times because there are all kind of patent disputes about that because everybody says they invented online travel and, frankly, they didn't. So that had been around for about eight years, and what happened was eventually the travel agents woke up to the fact that long-term this thing was going to be a competitor. They went to Sabre sales and said, 'You guys need to kill this thing.'

Instead of that, they gave it to me because I was running Sabre computer operations. They said, 'Why don't we put it over there?' I'd run it [eAAsySabre] during part of my career before, I had been a travel agent, and they said, 'Let's just get it out of Sabre.' Let me back up. I actually asked, when I was running it maybe four or five years before, 'Why isn't this thing on the Internet?' Because we found that we had customers coming to us from the Internet via CompuServe. At the time, we were pretty aggressive about getting it on more and more networks, and so they looked at it. At that time, the Internet was still under government regulation and you couldn't have commerce on the Internet.

Talks with Microsoft hadn't gone anywhere because each wanted control of the online travel site.

Jones: No question that Rich Barton drove him Gates to do their own thing. Microsoft had long discussions with us. I believe they were in good faith. You know, I mean I wasn't happy about it. I wanted to do our own things but we had our marching orders from our boss to go do it. Max Hopper, I think, was still there, who was the CIO, and Kathy Misunas. Maybe Hopper left, the boss was Crandall, and Kathy was running Sabre. So, he said, 'These guys are really smart about this stuff.' No question, Microsoft with their coverage, particularly at that time, they owned every PC, so it was powerful. It could've been a power combination, but it didn't work.

I go to this meeting in 1995 and they say, 'We're going to talk about putting this thing on the Internet and we're going to partner with another company.' We negotiated with this company, Worldview Systems. This was a content company. They had lots of destination content, and, at the time, we didn't know whether booking or content would be big. We did a joint venture with this company, we called it Travelocity. I think one of us owned the name and one of us owned the URL, so it was sort of a Mexican standoff. There were four buttons on the home page, and two went to them and two went to us.

Al Lenza, former vice president, distribution and e-commerce for Northwest Airlines and founding board member of Orbitz: Early '95 is when the noise started. Right around this time, the Expedia guys came over, early '95, to talk about their vision to have an online travel agent. It was not just Rich Barton. It was another guy, Greg Slyngstad, I believe was his name. The airlines would save on commission because they would be interested in only five percent commission. Then they were just trying to position themselves as, 'We're good for you. We're good for the industry. We're going to be good for consumers, and we're going to give Travelocity competition, which is owned by a GDS, and we're independent.

Actually, Expedia gave us the idea to reduce commissions to five percent for all online travel sites and to do so early when they were small. We did that in the middle of '95. Traditional commissions were starting to come down. I think January '95 was the Delta cap, the $50 cap, the first hit. Our initiative to reduce online sites didn't really get much attention.

You had the kiosks starting. You had commissions being cut. Online OTAs. All these things were happening. The e-ticket was the enabler for e-commerce. There's just no doubt, without e-ticket, e-commerce would not have flourished, so that was the enabler, and it was comical. I don't know if you remember, but there were a million skeptics in our airline and in the industry, saying 'Nobody will ever go to an airport without a paper ticket.'

I don't know how many times I told the story about if you lock your briefcase in your trunk and you're stuck and your paper ticket is in there, you don't ever have to worry about that again. We spent more time selling internally the value of this and the potential than we did externally.

The external adoption was amazing. There were all these writers saying 'Always have a paper ticket because you can go to another airline faster, and you're trapped if you don't have a paper ticket.' It was just the stories that just kept being spread for people to fear having this. It was amazing. We implemented paper-ticket fees. That took 100 percent in no time.

We could do anything on our website and not have to do it anywhere else. That was how the word got out about CyberSavers. All of a sudden, the word kind of spread that if you shopped online that you get a better deal. Whether it was fact or not, in some cases and not, customers believed it. That's what brought the huge increase in volume in the early years. I remember getting complaints that this was unfair to the elderly and to the people who were technically challenged, that it was just not fair. In socialist Minnesota, it was a big thing about that it wasn't fair, and it wasn't equal.

It reminds me of the way people talk about Uber now versus taxis, that it's not fair. They're playing by different rules.

Jay Walker, Priceline.com founder and currently chairman of Walker Digital and chairman and CEO of Upside: The idea emerged because we were looking to solve a problem. I ran a private an R&D lab called Walker Digital, and Walker Digital specifically worked on solving problems that we felt would be completely changed in what we saw was an upcoming digital age. We said, 'Look, in a world where digital tools and online networks existed, problems that in the past had been really impossible to deal with would be reinvented. Solutions would be reinvented because the world was about to change.'

The problem we were working on was how can a seller of perishable inventory discount their product to clear the market without disrupting the retail price for their own product? In other words, how do you have a sale and at the same time not hurt your regular sales, your regular customers. It was good for any perishable products. The question is if you had inventory that was perishing, networks would allow you to advertise that inventory in a unique and discreet ways. We knew that was coming but the problem was discounting it would only have disrupted your old retail prices.

So that and hence the name of the company is "Priceline" because there is a line which is the current market price for your product, the price line and you can't go below that line without destroying your own retail prices. Nobody is paying at retail if you're having a sale. That was the problem we were trying to solve.

One of the industries that we worked on to solve the problem in was the travel space because we knew that online travel marketing was very much how travel agents sold the business. And we knew that if you could create a solution to the problem in travel, there was an extraordinary amount of perishable inventory. At that time when we started, the airlines were flying about two million empty seats a day. There were many more airlines and the industry load factor was probably around 61 or 62 percent, unlike the 87 percent is today. The airlines were flying a lot of empty seats around, which meant there was a lot of perishable inventory but at the same time we knew there was a fair number of people who would fly if they could get the price they wanted. Those people were browsing, they were sitting on the couch at home or they were using alternate forms of transportation like the bus or whatever. So, effectively, there was a capacity to expand the airline marketplace if you can figure out how to sell to those people, how to capture that demand.

So we began working on the idea of a demand collection system. How would you collect demand for people who wanted to fly and were willing to fly, but at the same time were not getting this right price? That's where we had the insight to say, hey, if they told us the price they were willing to pay and secured it with a credit card, we would know that the unit of demand was indeed a real customer, not some fake customer, right? The only question was how do we collect demand that airlines would want to fill and that solution was to create a new product which was a product that said, look, you're going to have to agree as a leisure customer to fly anytime of the day on any major airline with up to one stop or connection on a non-refundable, non-changeable ticket. If you're agreeing to all those things, then guess what, the airline knew that you really were a leisure customer.

On that basis we said to the airlines, look, we have a new product, this super-leisure product. We have a new way to collect demand for the product and we're going to use this new-fangled thing called the internet and we're going to let the customer price the product so that will allow you the airline to see the demand curve all the way below the price line.

Walker: The reaction from the airlines was very negative. So, the airlines basically said, 'We don't like you; we don't like travel agents; we don't like anybody who wants to sell our products. We don't believe that there is such a thing as super-leisure customers. We believe that people prefer our brand over every other brand. Our customers want our brand so we don't believe there's anybody really out there willing to fly any airline at anytime of the day. We don't agree that your product has validity, we don't agree that your market has validity, and we don't like you.

'Well,' we said, 'That's all very well and good but we're going to have a lot of customers who are willing to give us demand. We're going to come back here with a big, thick book of demand that's going to perish every day. And, at some point, you're going to say that you'd have rather had those customers than simply tell us how this doesn't exist.' That's a typical innovative problem: data shows up in an established corporation and the established corporations typically say, 'Well this isn't any good. If it was, our people on third floor would have thought of this already.'

I went down to go see Bob Crandall at American Airlines [before the launch of Priceline.com] and Bob, as you probably know, was a legend in the airline industry and was a tough, smart, bare knuckles airline guy, and I went in because his people had said, 'You know, there's this guy you need to meet. He has invented a way to use the Internet that could actually help us, Bob.'

Bob hated the Internet. He said, 'Alright if I got to see the guy, I'll see the guy.' I went down to the boardroom at American Airlines in Dallas, Texas. The boardroom at American Airlines has got, like, its own ZIP code, it's so damn big. It's like a giant room. Here I come in with two of my guys, and Bob Crandall comes in and goes, 'I hear you've got something that I might like. I doubt it, but show it to me, OK.' I said, 'Fine.'

I lay out the business and I lay out the concept and I said to Bob, 'Look, here's how the product's going to work. You got to have to agree to fly any airline except you'll be able to tell us one airline you don't want to fly. You've got to fly any airline but you get to pick one that you specifically don't want to fly.' Because I knew that most people had one airline they just didn't want to go on.

Crandall interrupted me and he said, 'Jay I hate that. I hate the fact that they can eliminate one airline.' I go, 'Bob, why do you hate that? You're American. They're not going to put an X next to American, they are going to put an X to somebody on the airline they don't like, some little small major at the time.' I said, 'Not American.'

He goes, 'You don't know the airline industry. You hate the airline you fly the most.'

He said, 'Jay, here is the story: One out of every seven trips in the [same] airline is a service failure. It's not our fault: weather, mechanical, problems with the airport gates, one out of every seven trips is a failure. Which means if you fly me regularly, guess what? You have experienced more failures, whereas when you try my competitor only once in a while, you get on his flight, it runs perfectly and you say, 'Why when I get on the other guy the flight is perfect and when I get on with you, Bob, your airline always breaks?'

I never forgot it: you hate the airline you fly the most. So I changed the product. I said, 'Fine I will commit you will not be able to eliminate an airline. You will fly any airline, just fly any major airline, it doesn't matter which one, you can pick one, OK. He looked at me and he said, 'Jay, let me just be perfectly clear. We would be better off if you were hit by a truck.'

Then I said, 'You're speaking metaphorically, right? You are not taking it personally are you?' He says, 'You know, it's nothing personal Jay. The Internet is the worst thing that could ever happen to the airline industry. It's going to allow for total transparency on our prices, it's going to allow the stupidest customer to control the pricing to the consumer when we can't even control it to the travel agents let alone to the consumer. It's going to upend this industry in a way that we will not be better off. So therefore, OK, we don't want you to succeed.'

And I said, 'Bob, you can hit me with a truck. The Internet isn't going away. It's coming.' He goes, 'We will see, Jay. Usually it's just a handful of people that if they don't make it happen, it doesn't happen and you're one of those people.' I never forgot that – you hate the airline you fly the most – and American will be better off if I were hit by that truck.

The meeting ended and he said, 'We'll think about it, but don't hold your breath.' I suspect that the lack of cooperation in the airline industry the day before we launched might have had something to do with good old Bob. It's a real airline story, and I don't hold any ill will. I love the guys at American. They eventually did a lot of business with us. They're great people. I never took any of it personally. You know, the airline industry is a tough business. They hadn't made profits until the last year or so in its whole history on a cumulative basis, so anybody who think that they can run an airline go try. It's really hard.

Walker: I don't want to name names for a variety of reasons but they all pulled out the day before we launched. On the exact same day, they all decided that, 'maybe we don't want this to exist.'

That was a fun story.

The day before we launched. No, they didn't test it. They might have spoken with one another but that would be certainly against the law, so they probably didn't do that. That would be wrong. But we did get some very strange phone calls the day before we launched the business. 'You know, we told you we were going to do this but thanks so much.' The day before we launched, they all pulled out except for one airline [that] didn't pull out.

TWA didn't pull out. I think somebody forgot to make a call. That's exactly what most likely happened. We launched the business with our advertising and promotion. And, we told consumers if you're a leisure traveler and you're willing to be super flexible, we've got a new way for you to save money and the truth is the airlines are eventually going to lock in. It doesn't cost you anything to fly. You go on this new-fangled Internet, tell us the price you're willing to pay, and we'll come back to you in just a few minutes and give you a yes or no answer as to whether or not there's any airline willing to give you a seat at that price.

Hugo Burge, former CEO of Cheapflights Media, co-founder Howzat Partners and currently CEO of the Momondo Group: In the UK, the landscape was very dominated by traditional travel agents, and also charter flights and charter packages. Also, with a company, which I don't think you've ever heard of called Teletext, which was a television data service that painfully rolled through the pages every few seconds. It was very hard to navigate. It was a monopoly at the time.

This was throughout the '90s. This was something that was prevalent. If somebody was looking for cheap flights, a lot of people would filter through Teletext or even through the deals pages of the travel adverts in the back of the newspapers. It was also a world in which things were beginning to be shaken up. Low-cost airlines were beginning to shake up the landscape and starting to seize the initiative from the charter airline. Ryanair and easyJet: Those were the two pioneers and later British Airways launched it's own one with Go. What was interesting was they started off bypassing the travel agents. At that point, it was pretty radical to bypass the travel agents because all consumers would book through the discounters or High Street stores or telephone numbers in the back of the newspapers. Initially it was radical enough to book on the telephone.

Then, of course, the Web started. The traditional companies like Thomas Cook, Thomson, and Airtours, and some of the bigger names of High Street like Trailfinder and Flight Centre didn't really seize the initiative on the Web. It was interesting, the Web initiative was seized primarily by a handful of airlines, the low-cost airlines. I think Flight Bookers, which with a consolidator, was one of the first to seize the initiative and launched the eBookers brand. It eventually got big and then was purchased by Orbitz, as we know, later.

But the early players and the early entrants into the online sector were really a handful of British companies, but Expedia was one of the first. I think in October '96 – it actually coincided with the launch of Cheapflights – so it's interesting that we know the story of Expedia born within Microsoft and funded to the tune of several hundred million dollars to get going. They were launched in the U.S. in October '96, but in the UK they followed fairly shortly after, I recall. But Cheapflights was at the right place at the right time. It was founded by a travel journalist John Hatt, who had been travel editor of Harpers and Queen for 10 years. And his very simple idea was to make finding cheap flights simpler. His idea was to make cheap flights advertising more logical and simpler by organizing deals by destination.

Cheapflights started as a listing service for flight deals from consolidators, from people who had websites, but predominantly people who had telephone numbers so it was kind of a Teletext killer in many ways. Because Teletext was a non-user-friendly source of cheap flights deals, and Cheapflights knew [it could succeed] in terms of being user-friendly and offering a great range of deals in an easy format.

John had a goal of transparency and aggregation from a very early stage. He felt that by not selling and by aggregating all the different suppliers and sellers of products that he could offer a more interesting and better product. And that was at the heart of his vision, simply buying cheap flights.

I think he loved the idea that [Cheapflights] did what it said on the tin. In fact, another anecdote which I find quite amusing – as long as it comes across right – is that John had always used to say: 'Cheapflights is an amazing name. It's like the naked lady of the travel industry. It's highly attractive, people know what it does, and people love it. People love Cheapflights.' He was a great advocate of that and it was really his personality and drive that really got the initial website going, and the vision.

Geert-Jan Bruinsma founded Booking.com's predecessor company, Bookings.nl in November 1996 a few months after graduating Universiteit Twente in the Netherlands and discovering the Internet. Bruinsma sold Bookings.nl to Bookings portal in 2003, roughly a year after IAC's Expedia unit had done its due diligence and decided – in what might be viewed as an historic misstep – not to acquire Bookings.nl. The Priceline Group acquired it in 2005, combined it with Active Hotels, and rebranded it as Booking.com.

Geert-Jan Bruinsma, who founded Bookings.nl and currently works on business development at Booking.com: I got the idea in August '96. It was after I graduated from university. In my final year of university, I got in contact with the Internet and I found it really interesting. And I always wanted to start my own business after university. At the end, I decided it had to be something with the Internet. I had no idea what. Then, during a dinner with friends, the idea of hotel reservations came up.

In the Netherlands there was a national reservation center. And you could call them and they had contracts with many hotels and so you could actually call them to make a reservation. It was a semi-private business. It was part of a public organization, it was the Netherlands Tourism Board but they put this into a private section of it. I think most common was just call the hotel directly. Get a phone number from a phone book or get the phone number from a travel guide if you want to go to Paris. Everyone knew that people in Paris don't speak English so you had to speak a couple words of French to make a reservation. It was always a hard time, especially the cross-border thing. There was a huge language barrier herein Europe.

The morning after I made a connection to the Internet with this slow, slow modem, I searched for online hotel reservations. I didn't find anything for the Netherlands but I remember I found Hilton.com in the U.S., which at that time, you could actually already make a reservation online on Hilton.com. I must admit, I did some copy-paste from them in the early stages of the project.

In Europe, people who had Internet access were university students and they don't book hotels. There were some tech companies where people had access, but it was completely different from the U.S. where in the U.S. there was AOL, which was at that time quite big already. I think in the first two years of Booking, I think 60 to 70 percent of all reservations came from the U.S. people traveling to Amsterdam or to Europe.

Well, it was just me so I needed to learn how to build a website. I wrote a business plan because I needed some money, very little, but I wrote a business plan and I sent it to everyone I knew had an email address. There were 51 of them. Also, because they had an email address so at least they knew something about the Internet. Most of them probably never visited the website at that time. It was just friends. Yeah, the total investment was 50,000 Euros. There were 18 shareholders. It was a small amount of money. I think there were even some who didn't even look at the business plan, they said, 'OK, let's do it.'

Bruinsma: I made a nice brochure, printed it on paper and put it in an envelope with a registration form and sent it around [to hotels]. I never visited one, just sent it around. I had a list of 2,000 hotels in the Netherlands to start with. At the time I had 10 hotels signed up when the website went live. I started with five percent commission. Hotels considered it a new market. You always have the first hotels who say, 'I'll give it a try.' That was more or less the contract, Give it a try. No contracts or a very basic one, no price deals, the hotels had to set their own rates on the website.

I had very little knowledge about the whole industry. I was a night porter in a hotel as a student. It was a good time. It gave me some inspiration and at least I knew how I the reservation process went because we got people in at night who didn't book and so they came in for a reservation. I had no clue about commission rates; that's why I started with five percent. I had no clue about room rates, about how the whole industry worked. To me, it sounded very logical that hotels themselves, they should know the best room rate they can charge at anytime. From the beginning, it was the hotel that decides about the rate on the website. I know that in the beginning there was no credit card guarantee, you just made a reservation. But, I think after the first – after two or three – reservations, hotels started to ask about credit card guarantees so we added that one as well. Up until a couple of years ago we were never involved in payments.

Bruinsma: I wanted to put an advertisement in the Telegraaf, which was the biggest newspaper at that time. I got a phone call that they refused my advertisement because they talked about it and they decided not to accept any advertisements with an Internet address and there had to be a phone number.

This is '97. I think the day after when I had to come up with new ideas I started looking at search engines and what we call now Search Engine Optimization. I was pretty successful in that. I think 60 percent came from Alta Vista and we had some local Dutch ones, some national ones. They're all gone, but Alta Vista and Yahoo was not a search engine, it was still an index. You could go through it and you could click through travel the Netherlands. But AltaVista was the main supplier of visitors.

Bruinsma: The second step was affiliates. My ultimate goal was to be number one if people were looking for hotels in Amsterdam because that was most of the business. I remember there were two websites about Amsterdam, about hotels in Amsterdam, which were number one and two, and I was number three. I couldn't get rid of them. I also realized that they had no online reservation system; it was an email form. I could see there was a lot of manual stuff behind it. The other one was just advertising, asking 20, 25 Euros a month to have an advertisement on the website. I sent an email to both owners and I proposed to them, 'Let's say, if I put my hotels behind your website and we split commission 50/50. How does it sound?' I think within an hour they both replied like, 'Wow, this is a great idea.' It was Channels.nl, which is still live, and the Amsterdam Hotel Guide. It doesn't exist anymore. It took about two days to program it and I remember after two days they put it live. I know two of those guys from the Amsterdam Hotel Guide. For years they haven't done anything. It was so successful from the beginning that they didn't have to do anything. They were just looking at statistics all day long, how many bookings they got.

That was a big step because it was scalable. Every day we were adding new affiliates. This was in '97, beginning of '98, I guess. Search Engine Optimization was number one. We had landing pages before the term landing page existed. I didn't know it was called affiliates. We had a system in place and then only a year or so later we heard it was about affiliates. But that was the second big growth path. This was all before Google. After two years, business was going so well that I could hire someone. At the end of '98 there were the two of us.

AOL's purchase of Time Warner in January of 2000 marked the beginning of the end for the rapid growth of the first generation of Internet companies. It left over-hyped companies outside of travel such as Pets.com, Boo.com, and Kozmo.com with nothing. It also left travel brands that had done very well in their initial public offerings, including Priceline.com, on the edge of disappearing if they couldn't quickly justify their valuations with revenues and other marks of success. It wasn't the best time to launch an online travel company, either, but Orbitz carried on.

It was also the era of making mergers of the previous years fit in a sensible way, as well as creating new executive teams that didn't always match with the founders that had fought the early battles.

Boyd: The company missed its earnings in the third quarter of 2000 by, I think, a penny. It was just a very, very small miss. With a high multiple growth stock, even a small miss is essentially greeted with a big sell-off. The stock price traded down. With the Internet bubble essentially blowing up around us – nobody wanted to own Internet stocks – we got sold off. We had a sister company call Priceline WebHouse Club [for buying gasoline and groceries], which had to be closed down at the same time.

We had a car mart. New cars and long-distance telephone. At one time, we had 18 different verticals that were either up and running or under construction. The original vision of the company was Name Your Own Price for essentially everything. We narrowed it down to travel once the market conditions basically made it clear that there wasn't going to be any capital for that kind of expansion.

Boyd: I was chief operating officer. We implemented that plan at the time that I came in as COO. Bob Mylod came in as CFO. Mitch Truwit came in as [executive VP of operations]. Peter Millones became general counsel, and that was the team. We put together a plan to try to address all the major elements of the company's problems. Financial problems, problems with the marketplace, problems with regulators, and sort of tap away at it one at a time.

No turnaround is easy. We had to lay off a lot of employees, which was very difficult, and cut our costs. We shut down every vertical at that time except for travel and mortgage. We basically had to make substantial changes to the products to address the regulatory and consumer issues that were coming up at the time, and get our customer service in order.

So it was a long-term process. We were able to essentially get the financial aspects of it in place fairly rapidly. I think we made a GAAP [generally accepted accounting principles] profit in the second quarter of 2001. We thought it was very important to show a profit in this business because of the uniqueness of our business model, the Name Your Own Price business model. We wanted to sweep away any remaining concerns that it was somehow it was a gimmick.

[After 9/11happened] We focused on that [Name Your Own Price for hotels] product. The business did very well and grew very rapidly. But, during that time we came to realize that the universal customers that were willing to accept the trade-offs for Priceline's Name Your Own Price product were a small subset of the entire market. And companies like Travelocity and Expedia were growing to be much larger because they had much broader appeal. And, we made a decision to get into the retail side of the business initially by selling retail airline tickets. We came out with, what I think was one our best advertisements with William Shatner and Leonard Nimoy, where Shatner is being fired by Priceline, he says, 'He's going to replace me.'

Lenza: We tried to get the big European carriers to take a third of the ownership,and we'd be global. We had meetings with BA and Lufthansa and Air France primarily, several meetings in Europe. In the end, they couldn't swallow partnering with the U.S. carriers holding a majority. We also had many, many meetings with the big hotel companies, and in the end, they couldn't swallow being a minority owner of an airline-owned OTA. The Europeans ended up building Opodo, which was a big disaster. They could have been part of something bigger with us. It was a wasted opportunity.

Hafner: I drafted, with the help of our attorney, what we called the Charter Associate Agreement that we then sent out to, I think, 37 airlines, and got 23 of them to sign. It was basically a commercial agreement by which we accepted the same commercial terms and conditions that the founding airlines got – there were four founding airlines – in exchange for guaranteed access to all of their publicly available (including Web-only) fares. I think the initial wave was 23, but when we finally got the press release out it was like 37.

We said, 'OK, the scope of this agreement is for the U.S. market. Why don't we do a similar joint-venture approach for Europe, and a similar one for APAC?' So, all this stuff was going on at the same time, and we thought we'll have the U.S. entity own a stock in the European entity, and own a stock in the APAC entity.

I was flying around nonstop. That [signing joint-venture agreements with Europe and APAC carriers] was going nowhere because what can a U.S. company do to help the European company off the ground? And then the same type of conversation in APAC. We went down there and asked for 25 percent equity share in this joint venture, and they told us basically to go stuff it. And they said, 'Hey, if you're taking your market – this is the airlines talking – and all you're bringing is your own fares to this. You don't really have any inventory that provides for the density in APAC. You have no local market knowledge. Why would we give you 25 percent, not even putting in any cash?' So it's a tough sell. Ultimately, they should have taken the deal because the JVs that they created in both markets totally sucked.

Lenza: Katz was an interesting guy. He was difficult, but he deserves full credit — he and Hafner and some of the others – what they did. Katz had an insistence that you had either the [airlines'] CFO, the chief marketing guy, or the chief distribution guy, two of those three had to be on the board. That was one of his conditions. We had two board members each, and he insisted that they be the most senior either marketing, finance or distribution, so that's how I ended up [on the board]. He would have these big limos to pick us up to take us to the board meetings. He would insist we go, but he would really take good care of us. Somebody at one board meeting said, 'We're losing all kinds of money. Can you find us a smaller limo next time?' I think the next time, it was a little car that waited for us.

Katz: There was one funny story of how, just to keep their attention and make sure it wasn't painful for them, yeah, I sent limos down to pick up all the board members, and one board member – I'll withhold his name for the moment – he'd claim that I'm spending all this money on limos for board members. He took a cab back to the airport. The next time he came to the board meeting, he calls me up and says, 'Hey, do you mind sending a limo for me this time?'

I insisted that there be two board members from each of the airlines, and that each one of them had to report directly to the CEO, and it had to be the CFO or the CMO. We had very high-level stakeholders as board members, and I think that was essential. That was an essential ingredient to what ultimately became a successful enterprise. One measure is 16 years later, whenever the last sale of Orbitz was, we'd sold for roughly the same price as it did when we sold it in 2004 ... Ignoring a lot of shareholder splits, that would happen. They took out maximum dollars once the company sold. That's a measure of success.

Ellen Lee: Honestly, my take about Jeff – because Jeff is a complex guy – but I honestly believe that nobody could have made those airlines do what they did to get that business off the ground besides him. They needed somebody who knew them, who was strong, who had a point of view, and I honestly believe that he, while not sort of a founder of the idea and all this other stuff, I think the reason Orbitz was as successful as it was in the early days is because he could herd the cats. Better than anybody else. I give him a lot of credit for that.

Of course, you have to give Alex [Zoghlin] a lot of credit about the technology. They could have just built something like everybody else, but he really was innovative about the product he could build, and everybody replicated it. Everybody has a matrix now, but that stuff was brand new when Orbitz first did it. I think Alex obviously deserves a lot of credit for that too; he's a visionary.

Lenza: Now I think the e-ticket thing was revolutionary for global commerce, buying online for concert tickets and StubHub. Our industry can say, 'Shit, we led the way.' We were the ones who implemented in a big industry. Granted, hotels is a different question, but in terms of documents that were inter-lineable and global among hundreds of airlines and how it's been emulated in sports and concerts and plays, Broadway shows, it's really quite remarkable. It's one of the things that I'm kind of proud to have been part of that from the early going because it resonated way beyond and really made e-commerce possible. Without that, it would have been slow going.

Really, a lot happened in that window. If you take the window from 1995 through 2002 or so, in those six, seven years, the whole world changed. That's really kind of the takeaway in a big way. And, 9/11 did a dent on a lot of stuff, but it could not stop the unstoppable. Enough momentum had been built that it didn't slow the growth at all of this whole area.

By then, Expedia was established and growing, and Travelocity was growing. Then, of course, there was Priceline, on the other hand, and we were going to create a big OTA owned by the airlines that would do what Priceline and Expedia both did. Eventually, Texas Pacific Group and David Bonderman, who was an investor in America West Airlines and Continental and was close to many of the senior leaders, he said, 'No, no, no, I want to fund and build a competitor to Priceline.' We split the initiative. We created, on the one hand, Orbitz with the four airlines – United, Northwest, Continental and Delta – and then Hotwire [TPG with American, Northwest, Continental, America West, United and US Airways]. Because Delta was so invested in Priceline, Delta was left out of Hotwire. The airlines fully funded Orbitz, the creation of it. TPG and those investors funded Hotwire. The airlines gave their content, and TPG and investors put in all the money.

Orbitz came very close to folding before it succeeded. We had a board meeting a week or two before 9/11. We approved a significant funding and if that meeting had happened a week or two later there probably would have been no Orbitz because there would have been no more funding. It had been draining a lot of money up to the launch, and in August of 2001, they needed another funding round.

Had that meeting not happened when it did. But fortunately, Jeff and his management team did a great job. Steve Hafner was one of those that did a great job on the hotel side. Of course, the airlines after 9/11 hitting and the big guys heading into bankruptcy in the next three or four years, even with all the government help, all of a sudden you were selling assets, and Orbitz and Hotwire were caught in that.

I think there were some things where Orbitz was constrained by ownership, and one is that, the M&A process was really constrained. That ultimately was the launching ground of the massive success of Priceline and Expedia too, which was burnished by acquisition. So that was severely limited by what the governance structure. The good and the bad of the governance thing was a single-minded advantage in distribution costs, and creating great products for consumers, well for a good 10 years. A good 10 years that model dominated air distribution models in the world. It changed everything and everybody who wanted to be in the air business. That's a pretty big imprint. Was it the market cap success that Priceline became? Not at all. It had a different agenda for its shareholders.

Wertheimer:Anything based on a GDS [search] in the early days was going to have very few answers. The state of the art was that Sabre Bargain Finder would look at 10 outbounds and 10 returns from the 100 possible, price it out, and take the four best. We looked at literally 10 to the 10 [10 billion fares]. The main thing that I understood later was under Carter when they deregulated airline travel in the '70s, the systems were built for regulated industry. They were built when computation was fairly expensive on mainframes and when it was only used by a small set of people, by professional travel agents, and when searches were trivial because everyone charged the same price by law. Then to go from that to a point where you had more different prices than any other industry because they're perishable products and because the actual cost of production is zero. So it's all margin, and because of the crazy price elasticity, you had this very interesting yield management and revenue management variable-pricing scenario, but there weren't tools to deal with it.

We showed that you could do it. We took what was a pretty opaque market basically, and made it so that you could actually find the best answers. I think we were the first guys to do comprehensive shopping.

Katz: We forget that in those early days, when you did a search for a flight, which is what most people were experimenting with, you would see, sometimes three flights. Travelocity, when it started – and I was at Sabre when Travelocity was started– and we thought it was kind of an expensive experiment at the executives committee at that time. But, it was a browser view of Sabre, and it was an incarnation of eAAsySabre. Our CTO Alex Zoghlin created an architecture that was very much duplicative of what Google was creating at the same time – all Linux based, distributed microprocessors, really highly scalable, cheap infrastructure that could enable this massive search.

Then of course, we were the first to license – that really enabled – ITA, which was kind of an upstart product that kind of looks good, but nobody credible believed in it or even ITA at the time. But we sort of worked with them to help make it better and integrated their product into our data centers. So we had this idea of a low-cost infrastructure, massively scalable, and let's show everything. And when we turned it on in June of 2001, it worked. It started attracting a lot of attention. I can remember when we were doing a press tour right after launch, in the first week or two of June, I was in Times Square, and there was literally a line of camera crews all around Times Square to interview me. I'd go up to one camera, briefly a few feet away and there'd be the next camera. In between camera shots, I'd call the op center, and say, 'Is it still working?' People didn't want to turn it on. I remember we were basically ready. We had done months of testing and the ops team said, 'We're not ready, we're not ready.' And I remember saying some version of, 'Light it up, see what happens.' We had to start sometime so we did.

It was eye opening to travel agents, it was eye opening to consumers, it was eye opening to airlines. I don't think they knew. I'm pretty sure they didn't know because I used to get calls from a CFO or two at an airline, saying, 'Where'd you get this fare?' We opened up something on that scale that had never been seen before.

Gerstner: In the fall of 2000 wouldn't you know it, the dot-com bubble burst, but our business was still going gangbusters because we were kind of the only game in town. Priceline didn't have the time focus on this, Expedia didn't have the time to focus on it, Travelocity didn't really have the time to focus on it so we were able in a short period of time to basically sign all of these folks up. You know, Lycos, Yahoo, offline guys like Costco. We became the outsourced technology and fulfillment provider for all cruise and vacation packages for most of these players, if not all of the players.

They were all using us and obviously the nature of that partnership – that's obviously how I got to know Jeff Boyd and Bob Mylod [at Priceline.com] very well because I had to go negotiate a deal with them. Then Lloyd and Rich at Expedia because I went and negotiated a deal with them and Terry Jones and the guys at Travelocity; and Site 59, Michelle Peluso, because we negotiated the deal with them; Steve Hafner [Orbitz], which is a story unto itself that I'll get to. But I became friends with all these guys because we were their partner.

In November 2000, Expedia and then USA Networks independently start talking to NLG about acquiring it.

Gerstner: Now we have both Expedia and USA Networks interested in buying us. This was November of 2000. We meet with Diller and we say we don't know if this is going to happen because Expedia's also interested in buying us, and he said, 'Well tell me about this Expedia company,' which we did. We said we think it's got the best team in the business, we think they're going to emerge as the major players in the online travel space. He said, 'great introduce me to them I want to buy them too.' That was just bizarre, right? You didn't remember, we had only closed on the acquisition of NLG six months earlier in March of 2000. I had graduated in May of 2000 and here I am four or fives months later, after the dot-com crash, the rest of the world is melting down and I'm on this little rocket ship that both Barry Diller and Expedia want to buy.

Then we spent the next four months negotiating the deal and it turned out that Barry Diller did exactly what he said he wanted to do and by April or May of 2001, we had signed a term sheet and he was going to buy both companies. There was a three-way negotiation. We were still negotiating to sell our company to Expedia, Expedia was negotiating to sell themselves to USA Networks. It was just a complex set of conversations. In hindsight, but for the terrorist attacks, it wouldn't have mattered, right?

We signed a deal. I remember Diller gave an ultimatum, must have been at the end of April, that both deals had to be prepared to be announced I think in early June or late May. He gave us a date and he said, 'I'm going to announce both these deals on this date, they both have to be signed, sealed, and delivered, and if they're not both deals are off.' We all worked hard, we all scrambled, we all got the deals done, and whatever that date was we met in Hollywood and we announced the deals. And USA announced it alone, they were buying NLG and they were buying Expedia. Both deals had been signed and that was that.

I remember being in the back seat of a limo with Rich because the celebration party for that was at Diller's place, his house in Beverly Hills, and he was throwing a big party in the backyard. Here we were, it was just such a bizarre transition for this kid [Rich Barton] who had worked at Microsoft, and this kid [Brad Gerstner] who just graduated from business school and now we're on our way to a party in some Hollywood mogul's backyard and he'd just bought the companies. I will never forget that backyard event and Aaron [Gowell] was there. I'm trying to remember all the other players who were there. Bob [Diener] and Dave [Litman] from Hotel Reservations Network were there. Obviously Dara [Khosrowshahi] was there. Another guy who was involved definitely was Jon Miller because Jon Miller was the CEO of USA Networks and went on to become the CEO of AOL, and he played a very significant role in all of this.

On July 16, 2001, USA Networks announced it intended to acquire a controlling stake in Expedia and to purchase outright National Leisure Group, then the largest online cruise and vacations packager. But then came the terror attacks of September 11, 2001.

Khoswrowshahi: [USA Networks' deal to acquire a controlling interest in Expedia from Microsoft] had been finalized. And we had signed for the deal and that deal – as many deals of the time – had what's called a Material Adverse Change clause in the deal. September 11 was, if nothing else, quite material to the travel business. Unbelievable tragedy, and all the talking was about how we really didn't know where this was going in any way, shape or form. And we had some time to make a decision as to whether or not we wanted to go forward with the acquisition or not. And, there was a debate happening.

Rich Barton called up Barry Diller and Rich said, 'Listen, we want to be in business with you but people at the company are nervous enough right now as it relates to September 11, and if you don't want to close the deal that's fine, just tell me now. I want to know now whether you want to close it or not. I completely respect your walking away from the deal but time is my enemy. During times of difficulty and crisis, leaders need to be clear. They need to provide clarity to their employees. I just want to provide clarity to my employees so please tell me sooner rather than later whether you want to do this deal or not.'

We all got together as a small group of people around a table: Barry, myself, Victor Kaufman, the vice chairman of the company, and a few other folks. And we talked it over and I remember someone at the meeting – Barry and I still debate who it was – said, 'If there's no travel, there's no life. It's going to come back. Expedia is going to be an important part of it. So either the whole world is going to blow up or we're going to be OK and who knows how long it will take, but we'll be OK.' I was at that meeting and Barry said, 'Let's close' and I forgot whether it was myself or Barry who called Rich, we said, 'You know what? We want to go forward with our partnership.' And in hindsight it was a great decision and I've got to commend Rich for looking out for his employees and asking us to make that decision sooner rather than later.

[The deal to acquire National Leisure Group] wasn't as big a deal as Expedia. And, to some extent, we felt like we needed to hedge our exposure. If we're going to go forward with one player, it was Expedia. In hindsight it was definitely the right bet to make.

Expedia had its own path. I think Hotels.com led us to look at Expedia. After we bought Hotels.com we said, 'Hey, we like this travel thing. We like this sector. This is going to go online fast. Let's look for who else is playing in the sector.' It led us to Expedia, it lead us to Hotwire. It really got us in the travel field and certainly got us interested in the lodging sector as a particularly interesting area to organize. I think Expedia on its own, really the credit goes to that team on their own. Their recognition of product and technology got them there and also their recognition that they shouldn't just be an air brand. They should be a full-service brand, including hotels and dynamic packaging, were the real keys in their overtaking Travelocity.

One company that tied its fate to dynamic packaging was New York-based Site59, which specialized in spontaneity and last-minute, weekend trips when it was founded in the Fall of 1999. Site59 co-founder and CEO Michelle Peluso went on to become Travelocity CEO in late 2003.

Michelle Peluso, co-founder and former CEO Site59, former CEO Travelocity and currently venture partner at Technology Crossover Ventures: At Site59, we recognized the value of dynamic packaging early, thanks in large part to co-founder and COO Tracey Weber's leadership for both suppliers who wanted price opacity and for consumers who wanted real availability and great deals. We were the first to bring a lot of this to market, along with compelling content, for last minute trips.

We were one of a number of startups calling on airlines in the early days, and I found it often frustrating to try to get them to meet with us. Then, Travel Weekly wrote its cover story on Site59, and perhaps more importantly, included a photo and a line about how many beautiful women in fishnet stockings there were at our company. Needless to say, the airlines started calling us right after that, offering meetings in our office in NYC, and our business took off!

Travelocity acquired Site59 for its packaging business for $43 million in March 2002. Peluso concedes, however, that Travelocity got bogged down with acquisitions in Europe, such as Lastminute.com and a focus on profitability. Its technology didn't keep pace with those of its rivals.

Becoming CEO of Travelocity was a huge privilege and a great deal of fun. We focused our energy and effort around bringing the Roaming Gnome to market, the Travelocity Service Guarantee, broadening our international footprint, our best of class merchant hotel and dynamic packaging programs, and greater use of content and editorial through Experience Finder. Much of this enabled us to reaccelerate growth.

Two companies, Hotels.com and Expedia, were both powerhouses in the hotel industry with the merchant model. That didn't mean there wasn't a culture clash when a parent company tried to make them work together.

Khosrowshahi: We got into business with Hotels.com and they always complained about these Microsoft guys, these Microsoft/Expedia guys, who were their toughest competition. And we actually tried to buy a company called Travelscape out of Las Vegas early on, another competitor. Actually, Expedia ended up buying it. We tried to buy Travelscape, that didn't happen, we stepped away. Then actually Expedia bought Travelscape. At the time, Expedia was a public company; it was partially public and we set up a meeting with the team there and it was a bunch of young guys. There was Rich Barton and Erik Blatchford and Barney Harford there, who were out to revolutionize the travel distribution world. They were a group that couldn't fail. And so we were very excited about the Hotels.com business. We decided to buy the Microsoft ownership in Expedia and make a bet on the Expedia team and we did it. Then at that point we were really in the travel business so we had controlling stakes in two different players [Hotels.com and Expedia], both of whom hated each other and competed with each other. And one thing lead to another and we brought it all together under the aegis of what was that time, IAC Travel, which was spun off and became the Expedia that we know now.

Rosner: If you take people who have been competing and have been diametrically opposed for years, and then you suddenly say overnight, 'Hey, now you guys work on the same side of the table.' Culturally there's a clash from the competitive side of things.

So how do you different people do different things. And I'll never forget one year at a conference Barney Harford and I were on way opposite sides of a table. He was at Expedia [corporate development and strategy planning and later CEO of Orbitz Worldwide]. Before we came together he was on one side of the table, I was on the other, and I remember Michelle Peluso [former CEO of Site59 and Travelocity] was sitting on one side between us. And, we were basically just completely opposed, then the following year at the same conference a year later, Barney and I were sitting next to each other because we're family.

Rosner became president of Hotels.com in December 2003 as IAC acquired the shares of both Hotels.com and Expedia that it didn't already own. The task was to integrate their sales teams.

Rosner: I think it's human nature when you're bringing a sales team together. How do you do that? It was a great case study, frankly, for the learnings there, and how to integrate companies who have been really, really just dire competitors for so long. It's great, great learning.

Life would be kind of, I guess, cyclical. I would feel like one year I was a prom queen, the next year I'd be, not even invited to the prom. Depending upon what's happening in the market there's different views. Which is if demand is low for whatever reason then you wind up in much friendlier relationships with the suppliers. If demand is high then things become acrimonious. At the core issue was channel conflict, and not getting on the same side of the table to actually figure that out.

We did a lot to try to basically get on the same side of the table to resolve channel conflict. Things like being sold out on Expedia versus being sold out at the hotel, it's not like someone sat down and said, 'We're going to deliberately do something like this.' Most likely – I can speak to it from the Hotels.com side, not the Expedia side – but it was most likely if you have an allocation and the allocation is gone, then the hotel is sold out in that marketplace.

Bruinsma: Yup, that [around 2001] was more or less when Expedia started and grew. And when Expedia came to Europe they had the merchant model. In Europe booking a hotel was like what we do today. The agency model: You make a reservation and that's it and you pay at the hotel. I always compare them. The merchant model is like what we use when buying an airline ticket; you buy it at the time you book, you can't cancel anymore, that's it. I know that for Expedia it was a hard time to introduce the merchant model into Europe because no one was used to it. Also, Europeans making a hotel reservation didn't understand that they had to pay at the time of booking.

If you booked a complete package like a tour operator, then you were paying at the time you were booking but a hotel, just a hotel reservation, was not about paying.

I think this: The way we did the agency model was so much easier. No contracts, just sign up. At the end of the month or the beginning of the next month we were sending an overview with all the reservations and they had just to check those reservations – which were cancelled-- and we didn't charge them for cancellations. When we had the idea that the hotel was misusing this thing then we just selected some visitors, some guests and we asked them how their stay was and if they said, 'We had a great stay,' and the hotel said it was cancelled then we knew something was wrong. I think we only did it once or twice.

Bruinsma: In 2002 we were very close to a deal with Expedia and I think after six months of negotiation they cancelled it. There was a group of investors in the Netherlands who wanted to go on and so they came with an offer I couldn't refuse. For me, there was a reason to sell my shares and to leave the company. I had a one-and-a-half-year old son at that time and I would love to spend more time with him. I took my chance.

It was the director's decision [at IAC/Expedia]. I remember that we did all preparations and there were lawyers from Expedia in the office for weeks. They must have spent quite a lot of money on it. At the end, it was a board decision and the board decided not to do it. As far as I knew, the next morning after they told us it would be decided in the board meeting and the next morning we got an email like, 'Oh, sorry guys. The board decided not do it.'

Khosrowshahi: I think the Expedia team had taken a look at Booking.com and Active [Hotels, acquired by Priceline in 2005 and 2004, respectively] and again had passed. And, I think it was because we were attached to the merchant model and we were attached to high margins at the time. And I think in hindsight that blinded us. I think companies get very comfortable sometimes that there way is the only way and hats off to Glenn [Fogel, Priceline's mergers and acquisitions lead] and team for recognizing Booking and the potential there. And there's another deal that's been a home run on the Internet: the Priceline purchase of Booking.com. It's been absolutely extraordinary.

Bruinsma sold Bookings.nl to Bookingsportal in 2003. Arthur Kosten, who later became CMO of the Priceline Group's Booking.com for about a decade until 2012, recalls meeting Priceline's Glenn Fogel at one of Kosten's predecessor companies, Bookings.nl, before the Priceline Group acquired it in 2005.

Arthur Kosten: I think, around that time [2003 or early 2004] we [Bookings.nl] were talking to the guys at Active Hotels like, 'Hey, you guys are really good in the UK and in affiliates. We're much better on the continent and we're primarily good in SEO and SEM. What if we merged these businesses together?' … So essentially Priceline came by – Glenn Fogel, you obviously know him, a great guy – who did all the right acquisitions when everybody did the wrong acquisitions and he has a nose for finding these companies

Definitely for us, it was a bit scary. Suddenly to have these American guys that want to buy your company, and you didn't know them, and obviously you think they're way smarter than you are. Imagine we were a bunch of tech guys doing these projects, and acquired a company, and actually had quite a success growing it but relatively inexperienced young guys. Suddenly these guys with banker backgrounds and very experienced M & A guys come along and we were a little bit scared.

But you know Glenn, he's a very charming and friendly guy and we had a great conversation. I showed him, essentially, all the flywheels in the business, and the level of insight we had in the business, and the buttons we could press in turn to actually grow the business. For us, we couldn't envision that it would become so big as it is now but we were fairly confident that we could grow it a long way.

Priceline acquired UK-based Active Hotels in September 2004, purchased Amersterdam-headquartered Bookings B.V. in July 2005 and merged them, forming Bookings.com.

Kosten: Glenn [Fogel] came over and he was instantly excited with a couple of the ways we could show him how we managed and grew the business. So it was a weird deal because essentially every week we were beating the plan so much that we were afraid to tell Glenn because we thought that he wouldn't believe us and think we were manipulating.
I'm now an investment man too and you see all these hockey sticks and I've never seen a hockey stick play out but this was a hockey stick playing out during the process. Where essentially with it at some moment Glenn is just going to think we're cheating him because the numbers got better and better than expected every single week. So that was actually a fun process.

I think it helped that we had a couple of companies that we deeply admired like Google and Amazon that both emphasized to actually not look at competitors but look at your customers. Typically, we had a lot of new recruits coming in and say, 'Oh Expedia does this, we need to do it too.' We always said, 'Why? How does that validate? Maybe they're testing it, maybe it doesn't work for them, maybe they'll pull it back in half a year. Maybe it works for them in their model in their market but not for us.' We had a feeling like, hey, Europe is different because Priceline [Name Your Own Price] tried it and then it failed. Expedia tried it, and it wasn't as big a success as it was in the U.S. Obviously, there's something different to Europe. Essentially, I think, originally we thought Europe is different and we'll try to win in Europe. Yeah maybe the feeling was that these well-funded companies that have a lot of success in the States seemed not to be able to replicate that success so easily in Europe. That gave us the feeling that maybe we shouldn't look too much at them because then maybe we'll make the same mistakes as them.

Boyd: At that point in time, we then had a large and growing retail airline ticket business and a large and growing retail hotel business. So that allowed us to think of the business internationally as a broader business than necessarily Name Your Own Price. The historical context is that that when I joined Priceline, the company's strategic plan was to have a Name Your Own Price business in all the major parts of the world. We had a joint venture in London, in Hong Kong, and in Australia. We had businesses that were intended to build the Name Your Own Price brand around the world and we closed those down because they were very small and losing a lot of money. We couldn't afford to keep them running.

Glenn Fogel, who was our corporate development guy, and still is today, shut down our business in London for us. But to his great credit, he continued to stay very close to the market and ultimately identified Active Hotels in Cambridge as an agency model hotel business. Hotel only – it was growing very well, and had terrific people working there. We made that acquisition first. A year later, Glenn identified Booking.com. We bought Booking.com, and ultimately combined Active and Booking and that was one of the few mergers that the integration really did provide some top-line synergies.

What the guys at Booking.com did the best really was that they understood how to basically address the entire market and not just the big chain hotels, the Accors and Marriotts and Hiltons in London, but ultimately leads to find demand for and service hotels in all countries, small towns, small hotels.

We believed at that time that the agency model was preferable for consumers obviously because they don't have to pay up front. And while collecting commissions can be a challenge, our businesses had the processes and understood how to do that. And so we actually felt it was an advantage. It was very hard for the others to really move away from the merchant model because of the very substantial cash flow benefits that accrued to the merchant that collects the money and hangs on to it until the guest check out.

I think that was an advantage for us. I think the businesses were initially misunderstood as affiliate businesses that were relying on a bunch of marginal click-farming websites for their demand. Obviously, that was the not truth. These businesses were growing very substantial, direct business, and brand strength over time.

It was a great acquisition. We were really lucky that we have tremendous entrepreneurs at both businesses. In particular, the guys at Booking.com, after the integration of the two really with Kees Koolen, Arthur Kosten, Pietes Doorne, and Adrian Currie came from Active Hotels; they really did a great job scaling the business. And I think one of the insights that we came up with, and it turned out to be the right one, was to really make sure that we gave them an extraordinary amount of control and independence to the guys running all of our businesses. And that's still a hallmark at The Priceline Group.

Burge, CEO of the Momondo Group: You've probably heard this from the Booking and Priceline guys. When Priceline first launched in the UK, I think they hired the former marketing manager of Burger King, and spent millions upon millions of pounds on advertising with a product that just didn't work in the UK market. It's quite interesting that the first attempt at Europe by Priceline was – I think I've heard even them talk about it as – an embarrassing failure. But they picked themselves up. And then instead of growing organically, they changed models and they bought a completely different model at a very good price, and the rest is history. They've made the most successful travel acquisition in the history of the Internet by buying Active Hotels and then Booking. So it's interesting how things go.

Not all of the interviews we conducted with early players fit into the big story, so we are publishing these interviews with key figures at Ctrip, Hostelworld, Travelstart, and Decolar. These conversations offer deeper insight into what happened at online travel brands outside of the United States both during and after the early years of online travel booking.

Ctrip's Different Path to China's Consumers

Ctrip started out as a TripAdvisor-like review site in 1999 and soon developed a small hotel business. Among the marketing tactics, Ctrip staff began handing out advertisements for the hotels at the airport in Shanghai and rail stations. This didn't happen in isolation: Ctrip's founders were well-aware of what was going on in Silicon Valley at the time and took encouragement from the growth and development of Travelocity, Expedia, and Priceline.

Ctrip obviously had to adapt its own development to conditions in China at the time, and emerged as the strongest player after the financial and SARS crises.

"[Co-founder] James [ Liang] wrote an email to all the employees telling them that after SARS, Ctrip will become stronger," says Ctrip co-president and COO Jane Jie Sun. "And, sure enough, SARS plus the financial crisis swept out the companies that were not very efficient or innovative. But Ctrip stood out. Ctrip and eLong were the two companies after this turmoil that survived."

Skift: I was very interested in the founding of Ctrip and what was going on in China at that time. Were there other online travel companies already doing business when Ctrip was founded and what was the situation?

Jane Jie Sun: Ctrip was founded in 1999. When we founded our company at that time in the USA we had already seen the takeoff of Silicon Valley. James Liang, our current chairman and CEO [and co-founder], evaluated the different options … At that time, payments were really an issue because credit cards were not widely used in China. When they tried to establish Ctrip, at the beginning, it was just a review community similar to TripAdvisor. It was a community with no monetization at the beginning.

Then the capital markets tightened up. All of a sudden, we had to turn it into a money-generating business right away. Again, I think James evaluated different options. So the first line of business he went into was the hotel business because hotels are relatively fragmented. Then we established a model, which is we make the reservation and the customer will pay the hotel directly. By doing that, you avoided the issue of credit card payment. At that time, Ctrip was very small. We concentrated in bringing volume to very specific hotels. That way nearby hotels would realize, ‘Wow, Ctrip is valuable. We want to put our inventory on their website, too.’ So gradually, Ctrip established partnerships with a majority of the hotels. Ctrip also established a strong sales force on the ground to recruit memberships. Ctrip was the first to give out Ctrip cards at the train station, bus station and airport, all these traffic hubs. That was one method to acquire traffic.

Skift: The cards were just like an advertisement?

Sun: Yeah, for example, when you landed in Shanghai airport our sales team would give you a Ctrip card. And, by using this card, you would be able to enjoy the hotels that we negotiated on behalf of our customers.

At that time, the Internet was just at an early stage. It wasn’t very easy to find a hotel if you did not use the Ctrip website. The sales and marketing side was also quite labor-intensive at that time. That’s what was going on. We had two teams. One is the sales and marketing team, which was recruiting the customers by giving the cards to the customers on the ground near the airport, the train station and bus station. We also had another hotel team working with the hotel to enlarge our inventory pool. Both sides needed to grow very quickly in order to sustain strong growth. That was the first line of business. It was very successful.

At the beginning, when we established our business, we were very innovative. The majority of the travel business was done offline by establishing roadside travel agencies with very, very low efficiency. But what we did was we created a centralized call center. It was a paperless operation. All the information from the data was stored in our proprietary software. In a traditional travel agency, in order to train a sales rep, it takes about one year to make sure the sales rep is equipped to host customers. And if this sales rep leaves, then he or she will take away all the information, including the customer’s data. Our founder, James, used to work at Oracle. He is a software engineer so his net mindset was all the information needed to be stored in our system.

Our staff normally, when you train them, would take three months to be put in a position to host and to take care of the customers. Even if they leave all the information remains in our system so the next representative can also help them. That’s what was going on during the early days. The majority of the calls at that time came to the call center because in 1999 and 2000, the Internet connection was not very easy. Broadband was not widespread. So PC volume was still quite low at this time.

Skift: Did any other companies, such as airlines or hotels, have websites where they took online bookings at the time?

Sun: There were a lot of online travel agencies but after the financial crisis, only two survived. One was Ctrip. One was eLong, but eLong was much smaller. What Ctrip did is after we successfully launched the hotel business, we then acquired a very small air-ticketing center and we started our air-ticketing business because we already had a big pool of the customers who were using hotels. And, naturally you could promote the air-tickets business. About one to two years later, the air-ticketing business took off. Again, it was very successful.

At that time, credit cards started to take off. We issued the first joint credit card with China Merchants Bank. By doing that, I think we were able to take the payments upfront from the customers and make a reservation for the customers. Again, that was a huge undertaking, but it was again, very successful. After the air-ticketing business, we already had two killer businesses under our one belt. One was hotels. One was airline tickets. The third line of business was vacation packages. Vacation packages were a combination of hotel, plus air and plus local activities. Since we already had two major products under our belt, the vacation packages had a good foundation. That again also took off very well.

Skift: When James was first getting the idea to do the hotels, what was the reaction from investors? Was it easy for him to raise money to get Ctrip started or was there a lot of skepticism?

Sun: We did our IPO at the end of 2003. So for four years Ctrip was a private company. I believe the first couple of investments were from IDG Capital Partners and the Carlyle Group etc. I think the investors had a good vision and they felt the team was very solid. And everyone was burning money. When Ctrip burned money, the efficiency and scalability achieved far exceeded everybody else. That is why the company stood out among hundreds of startups.

It’s very interesting. In 2003, I believe, SARS swept over the whole country. So a lot of companies shut down. James wrote an email to all the employees telling them that after SARS, Ctrip will become stronger. And, sure enough, SARS plus the financial crisis swept out the companies that were not very efficient or innovative. But Ctrip stood out. Ctrip and eLong were the two companies after this turmoil that survived.

Skift: There was still a lot of competition, I imagine, from the offline market?

Sun: The offline market was the majority. Even today, the offline market still is the majority. I think if you have a new technology that is more efficient you will serve the customer much better.

Skift: You said that after the financial crisis, a lot of the competition didn’t make it. Yet, it became such an extremely competitive environment with all the price discounting. When did that take shape?

Sun: The China market attracts lots of capital so lots of foreign investors really do not know the market very well. So they pour in lots of money into this market. But really it’s very difficult for the smaller players to know [the market]. Travel actually is very intricate. You also need to make teeny little steps to be flawless in order to execute well. Lots of smaller companies tell a very good story to the investors and gets lots of money. Then they try to beef up the GMV [Gross Merchandise Volume] and then they set a GMV as their target. If you do that, it’s very easy to inflate growth because it is very easy to inflate GMV If the team feels GMV is the target, then it could lead to loss-making sales.

Skift: Yeah, I’m sure there were many doubters about Ctrip, especially in the beginning.

Sun: Fortunately, I think in travel we already had a couple of players in the USA. The USA was very advanced in online travel. We had Expedia, Priceline, Orbitz, and Travelocity. The U.S. was always the leading player. Investors saw what happened in the USA and our model was very easy to explain it to them. What was difficult was that China at that time was quite backwards. When I first joined Ctrip, investors always asked me, ‘Do you think we should invest in Ctrip?’ I told them, ‘Every investor, every fund has its own criteria. But I can share why I came to China to work for Ctrip because for me I’m investing my youth and time, and my life in this company.

I told them, ‘You really need to ask three questions. First of all, do you want to invest in China?’ Because 10 years ago, some investors still had a bias against China. It is a country very few people knew about it. The reason I came back was I was born in Shanghai. I went to Peking University law school. I went to study, work and live in Silicon Valley for almost 15 to 20 years. The GDP growth rate at that time in China was about 8 to 10 percent. In the USA it was about 3 percent.

For me, as a young person, if the country is developing faster that means they need more talent and the career path will be better. China obviously was going to surpass many countries because the government is very efficient and people here are very hardworking. It was going to surpass a lot of neighboring countries like Japan, Korea, Southeast Asia, all these. So that’s a great opportunity. The first question is whether you want to invest in China.

The second question is, ‘If you want to invest in China, which industry do you want to invest in?’ I feel travel is a great industry. A lot of other industries, such as alcohol, tobacco, and the gaming industry are very profitable. They’re high-margin. For travel, you bring happiness to people’s lives. So you are very rewarded. Also when so many Chinese people go out into the world we always ask our customers to become the diplomats for China, and to bring the best of China to the world. Let people know China through you and ask these customers also to bring the best of the world to China so people in China will know what’s going on in the rest of the world. From that perspective, I think it’s increasing the friendships between Chinese people and the rest of the world.

That is very meaningful to me and to the rest of our team. Every day when we are issuing a ticket or booking a hotel, and we don’t see it just as people working. It’s a mission that we are actually doing. We’re training the whole country. We’re bringing China closer to the rest of the world. I think travel is a great industry. And as a percentage of GDP, the China travel industry still is much lower than in Europe and the United States. Again, the travel industry will be among the fastest growing industries there.

The third question is, ‘If you want to invest in China and in the Chinese travel industry, which team has the best chance to win?’ Throughout the past 15 years, every year there were newcomers that came in to compete with us. In the beginning, I think, it was China Mobile, then banks and lots of players. In the end, it’s always whoever has the strongest customer service, whoever has the strongest technology, branding and product will win.

Internally, we have five very important items. One is service. That’s important. The second is the product. You need to have all kinds of product rather than one product. The third one is technology. The fourth one is price. And if you have all these four items, then your branding is very strong. If you can do these five items very well, then you will be leading the competition. That’s what happened.

Feargal Mooney, who currently is CEO of Hostelworld Group, the largest standalone hostel-booking platform on the globe, joined the company as COO in 2002 when he says "there were 12 people working out of a bedroom or two or three bedrooms in a house" in Dublin.

The company, founded by software developer Ray Nolan, brought hostels online by offering them free property management system software and only charged them 10 percent commission. Hostelworld developed a business model that niftily resolved commission payments and collection by making consumers prepay 10 percent and forking over the rest when they got to the hostel.

Nolan tells Skift about the founding of the company in Dublin. "I’m a coder, had written the in-hostel software used by hostels in Ireland, and saw that connectivity with the Web was the only way forward. The business model was entirely my creation. And it was the 10 percent deposit-based business model that was most important: We could grow without having to build administration systems to repatriate money to hostels, and of course they would never worry that we be slow payers. It was in place from the first booking and never changed."

Nolan adds: "The first booking was made August 3rd, 1999. Coding began in February of that year. I know because I sold 10 percent of the company for 30,000 Irish pounds to fund my wedding on April 16th that year. You don’t forget stuff like that."

Along the way, in late 2002, Hostelworld offered by fax to acquire a rival, San Francisco-based Hostels.com. After the offer was accepted -- again, by fax -- the team went out and celebrated an extended Christmas and New Year's holiday, and when they returned to the office in the second week of January 2003, they realized they neglected one little detail -- they didn't have the money to pay for the acquisition.

Skift: Tell me about the creation of Hostelworld. What was going on?

Feargal Mooney: The business was started back in 1999 and unlike a lot of businesses it didn't necessarily start off doing what it would eventually end up doing. It was started by a techie guy called Ray Nolan. He was a computer programmer and used to write all sorts of software packages.

And one of the hostels in Dublin came to him and said, ‘Will you write a software package to help us run our hostel more efficiently because we're keeping all of our reservations in notebooks and so on? There are lots of property management systems, or PMSs, out there for hotels but they don't work for our hostel because in a hotel you're reserving a room where in the hostel you've got beds within rooms. And you've got beds in male dormitories and female dormitories, all those sorts of things that hotel systems don't cater for. So can you write us a package to help us manage this?’

He wrote the package and then he realized if one hostel has this problem then all the other hostels probably have a similar issue. He then went out to all the other hostels in Dublin and sold them the same system. In some case he went in and said, ‘I've written a property management system specifically for hostels and I'll sell you a copy of this software for ... ‘ I can't remember what the price was but it was big fee and ‘Here you go.’

Skift: Did this company have a name?

Mooney: The name of the company was called Raven Computing. And one of the hostels that bought it was a hostel called Avalon House in Dublin. The guy who owns that and ran that was a guy called Tom Kennedy. Tom bought a copy of the property management system software and started using it. After a couple months using it he said, ‘Look, this is great. We used to have notebooks and people would bring them to cancel reservations and if they booked for three nights we might scribble it out on one page but forget to cross out the other pages and so on. But now we have it all logged by the system. It's great but I still have a couple of people full-time employed who are either answering the phone or responding to emails from customers who are saying, do you have availability from the 23rd of June to the 27th of June and if so what's the price? I know if I respond to that email 10 other hostels in Dublin probably got the same inquiry from the same customer and we're all putting effort into responding to these email inquiries and only one of us is going to end up getting a booking from it. So is there a way that we can integrate the email inquiries with the property management system?’

That was where the idea came for what started off as HostelDublin.com. When you start a business in Ireland in a city as small as Dublin you realize fairly quickly that Hostel Dublin just isn't big enough. The market in Dublin isn't big enough. A couple weeks later it became HostelIreland.com and then again in a month realized that Ireland isn't a big enough market either and it became Hostelworld.

Skift: Are you talking about around 2000 or what year are you talking about now?

Mooney: The second half of 1999. Effectively what they did was say was, ‘OK, let's start this business. Let's change it around.’ At that time there were online travel businesses there. The big guys then were Expedia, Hotels.com, or Hotel Reservations Network, as it was called at the time. Travelocity and then Orbitz and Priceline were kind of small players. And, effectively there was no one that was doing online reservations for hostels because the economics of it just didn't make sense. The economics didn't make sense for the OTAs because the price point of a hostel bed was too low. Remember at that stage most of the OTAs that were there had a lot of their business still in call center reservations. Even though people might come and see the hotel on the websites they'd ring up to make a reservation and the payment processes were fairly manual. There was a lot administration follow up, which is fine if you were booking a $250 to $300 hotel room and you're taking a 30 percent commission off it but for the price of a hostel bed it doesn't make sense. And it wouldn't have made sense for the hostel operators either because again the price point is lower their and their margins are lower. They couldn't afford to pay the OTA commission rate.

Skift: Who were the customers at that time?

Mooney: [The customers] would have been students. Before Hostelworld developed what hostels would've done to get business -- their marketing and so on -- would have been going down to the ferry port or the railway station or the airport and handing out flyers, making sure they were included in the Lonely Planet guidebook or Let’s Go or whatever. Maybe going to and or handing out flyer or posters out to universities and colleges because it would've been very much a young and student-based customer audience at that point. Effectively what the guys did was say, ‘OK, let's change the model here. Let's start getting hostels online,’ and they developed what eventually was an OTA dedicated to and focused on hostels.

There were two issues that needed to be resolved. One, how do you get hostels onboard and signed up to your site? With any of these businesses it's a marketplace. It's a two-sided market. You've got the traveling customers that want to book somewhere to stay, and you've also got to get the accommodations to sign up. Without products to sell you can't get customers and without customers it's hard to get product to sell. They needed to figure out how they were going to get hostels to sign up and get them beds to sell on the site.

They also needed to figure out a revenue model that number one was going to be attractive to both sides, but number two, a model that was going to be highly efficient and automated. Because, if you're selling a hostel bed for, at that point in time, it would've been, $10 to $15 you can't have somebody answering the phone to take that reservation because the economics of it just don't work out. What they did was they said, ‘Let's start giving out the property management system for free.’

The property management system, when they finally put a name on it, was called Backpack. Instead of charging hostels for a copy of the software they said, ‘Let's give out Backpack for free and the condition is that Backpack is integrated where it's the Hostelworld back-end booking system. If you were running your hostel on Backpack it means that we get visibility and access to your beds to sell. And you don't have to pay us anything unless we sell a bed or process a reservation for you. If we process a reservation then we take a commission.’

The commission was set at 10 percent, which at that stage was significantly lower than most of the OTAs [online travel agencies]. At that point, [the OTAs] would've been charging 25, 30 and sometimes higher, up to 35 to 40 percent commissions. It was a very attractive commission rate to try and encourage hostels to come onboard. Remember again with hotels you've got lot of chain hotels. If you were Expedia, Hotels.com, or Travelocity, if you got a contract with Hilton Hotels and Marriott and InterContinental and other chains you probably had 30,000 or 40,000 hotels on you site. Whereas with hostels you are dealing with individual hostels signing them up one by one. There just weren't any chains of hostels. Things have changed very significantly in the last few years but at that time most of them were single owner-operators. They weren't online. A lot of the original hostels that we signed up meant we had to buy them computers. We had to install the Internet connection lines for some of them to get them going and get them up and running.

Skift: What was the reaction from the hostel owners once you started to get them online?

Mooney: We were saying to hostels, ‘We're not going to charge any listing fees. You don't pay us anything unless we deliver a booking. If we don’t deliver bookings you don't have to pay us.’ That helped up get the hostels onboard and on the site. It was very much a case of going around kind of door-to-door, doing the hard sell, and explaining to them the whole Internet thing and online reservations and so on. Like any startup, we had a small team. We had a couple people out in Australia. Australia has a big backpacking culture so they could buy a second-hand car and drive around Australia signing up every hostel there is in Australia.

The key to success was the model that we developed. The models that were in operation by the mainstream OTAs at that stage -- what I would refer to as the Expedia model -- was where you pay the full price of the stay to Expedia and then they settle with the hotel. Alternatively, what's now the Booking.com model is travelers don't pay anything to Booking.com; they pay the full price to the hotel when you arrive and then Booking.com invoices the hotel for commission. Those were the two models. They were fully pre-paid or fully post-paid and both of those involved a huge amount of administration. All this stuff with a 10 percent commission on a low-price bed, you're not going to be able to make any money with that.

The model we developed was one that was new and fairly innovative and the one that we still operate today. The customers come to the website, find the hostel they want to stay in and they enter the credit card details. We process the 10 percent deposit non-refundable and the customer pays the balance of the 90 percent to the hostels when they check in. What that meant was that 10 percent deposit was effectively our commission and we kept that and the hostel got their 90 percent when the customer walked in the door. We didn't have to do any of the administration of invoices for commissions or collections. It meant that the whole booking process was 100 percent automated.

The search engines, even Google when it came along, weren't that smart and there wasn't anybody in the whole world who was doing online reservations for hostels. So, what we did was we registered about 700 different domain names. We had a hostel domain for every city, every country, every continent in the world. For example, if you had a hostel in San Francisco you would be featured on HostelSanFrancisco.com, HostelUSA.com, HostelAmerica.com, HostelNorthAmerica.com, and on Hostelworld.com. So we had for every city, every country, every continent a specific domain name that was localized to that. That was really because if you went on to any of the search engines and searched for any hostel-related term all 10 of the first 10 listings that would come up would be one of our websites.

I met the guys in late 2001. At that stage they had the website up and running. They were processing bookings. My background was on the finance side of things. I had worked as a financial analyst with a number of organizations including technology companies. I'd also run a guesthouse at an earlier point in my career. I was introduced to the two guys. There was a techie guy and there was a hostel operator. They needed someone with a finance background to come in and sort the business side of things. So I met them and I started working with them in the beginning of 2002. Still in the early stage, I remember when I joined it was like a startup. There were 12 people working out of a bedroom or two or three bedrooms in a house. It was very much at the stage of can we afford to pay the bills at the end of the week? Nobody was getting paid a fixed wage. It was a case of, well, do we have any money to pay wages at the end of the week or not. If we don't let's go to the pub for a few beers and so on. That was the stage we were at. I think shortly after I joined we had a big celebration. It was the first day that we did 400 reservations in one day and that was a massively big record for us and we all went out and celebrated that.

It was also at a time when the idea of raising significant capital to drive this wasn't really an option. Some of that was related to the fact that there's quite a difference in U.S. -- and particularly Silicon Valley-funded models -- compared to the funded models they're operating in Europe. We don't have the same sort of VC infrastructure that might be in place in Silicon Valley. You have the whole 9/11 thing and people saying they'll never want to travel again. The appetite just wasn't there. Effectively what that meant was this was a business that was funded by friends and family and we knew that we had to get to break-even and profitability very quickly. We did that. I started with the guys in early 2002 and we got to profitability by July of 2002.

My first title was chief operations officer, which was a very fancy title in the start of an organization working with 12 people out of a bedroom. I was chief operations manager. The techie guy was CEO and the hostel guy was director of sales and marketing. Effectively we built it up, the three of us together. What was also interesting was the other two guys had looked at this and said, ‘We think it'll be great. This has got huge potential. I think it's a great opportunity. We kind of got into it by accident. We want to build this up. We want to sell it in a couple of years and make a fortune and retire and live happily ever after like a lot of dot-com people at the time.’

What it meant was that their focus was always on once we got to profitability, how can we start extracting cash out of the business? This was a business that the total sum of investment was less than a million euros. It was giving significant returns to the founders. Both of them left the business. The hostel guy, Tom, left in 2004 or 2005, then Ray Nolan left in 2008. In 2002, we got to profitability. At that point things kind of took off over the course of the next couple of years. We were seeing triple-digit growth. It was flying.

There was one other site at the time founded by a guy in San Francisco who owned a hostel and who had set up a directory site, kind of listing site of hostels, called Hostels.com. Around 2002 -- it might have been 2001 -- we took a look and said, ‘Why don't you put our booking engine in?’ You've got all these hostels. He had this idea of having every hostel in the world listed on the site with directions and phone numbers and everything. We said, ‘We've got a booking engine for hostels, why don't we do a partnership? You use our booking engine and you can make some money from this.’

With a domain name with Hostels.com he was obviously driving a significant amount of traffic. We then approached him and said, ‘We'd be interested in buying your business, your domain name or website.’ We made him an offer to buy the business. It was Christmas Eve of 2002. We sent him a fax offer to buy Hostels.com. He faxed back and said ‘offer accepted.’ In the terms we said that we would close the transaction by the 31st of January. He accepted the offer and we said great. It was Christmas Eve. We went off and we celebrated Christmas and ate our turkey and ham or whatever. In Ireland, Christmas is a fairly extended holiday so it goes on for a couple of weeks. We arrived back to the office probably the 6th or 7th of January and said, ‘We've got to get this deal sorted.’ We contacted our lawyers and said, ‘Look, we need to get legal agreements, well we want the money.’

We didn't think that was going to be a problem because we thought we'd go to our local bank and borrow the money from them because now we were a profitable business. When I say we were profitable, in the full year of 2002 we probably turned a profit of a couple hundred thousand dollars or whatever. We went to our bank and put together a presentation saying, ‘Here's a great opportunity for us to buy Hostels.com and here's all the reason we think we can do a better job to drive revenues and so on’ and the bankers sat and listened to our presentation. At the end of it they said, ‘That's all very interesting but there are two problems with it. One is hostels’ -- because they just didn't get the hostel thing as a legitimate business. So, ‘one is hostels and the other is dot-com’ because that was after the whole dot-com bubble had burst. Surely this wasn't going down well. Then they said, ‘Next time you come looking for a multi-million dollar loan from a bank it might help if you put suits and ties on.’

So that didn't work. They didn't give us the money. At that stage our lawyers were saying, ‘Look you've got a legally binding contract here. If you don't find the money by the 31st of January then the guy who owns this will sue you and then your business will be ruined.’ We effectively had less than three weeks to find the money. I put together a loan-note instrument and went round Dublin city to everybody and anybody that I knew or I'd ever heard of that was anywhere reasonably wealthy and offered them an opportunity to subscribe to some of the loan notes and put a fairly effective coupon, -- I think it was a 12 percent coupon --on the loan notes and through that raised about one third of the money. With this, we then went to another bank and said, ‘Here, we've raised this ourselves,’ and we ended up getting the balance and completing the deal. That was a fairly significant milestone for us. It was a little bit scary in the outset in terms of whether we were going to get that or not but we did. Hostels.com returned the full purchase price in profit for us in the first 18 months.

We were probably 30 to 35 people by 2004. In terms of the destination markets, Europe would've been the biggest piece because there are more hostels in Europe than anywhere else. Europe would've been probably 75 percent of the destination market. Although in terms of sources of customers it would have been pretty global and the U.S. has always been our biggest nationality of customers.

Argentina-based online travel agency Decolar, which counts Brazil as its largest market, was founded in 1999 at a time when just about all airline or hotel bookings were made in a brick-and-mortar office, on the phone, or at the hotel and airport.

When Decolar started dabbling with selling airline tickets online, there was a lot going on behind the scenes and even in subsequent years it was common for customers to visit Decolar offices to verify that the company really existed or to make changes in payment methods.

"For example, after the customers finalized the process to buy online -- a ticket, for example -- we had to check with the GDS [global distribution system] in Sabre or Amadeus to see if everything was OK with the booking," says Bob Rossato, who was a corporate service manager for Decolar and sister brand Despegar for six years beginning in 2001.

"Then we had to call to each customer to request some information about his or her credit card," Rossato recalls. "Every customer had to send to us the documentation filled out in a form, a copy of the credit card, a copy of the ID, front and back, and send it to us by fax."

We spoke with Rossato, who is a co-founder and board member of Brazil-based ViajaNet, as well as ViajaNet co-founder Alex Todres, who was an advertising and marketing manager at Decolar/Despegar a decade ago, about the development of online travel in Latin America.

Skift: How did online travel get started in Brazil?

Bob Rossato: Yes, I started to build the OTA [online travel agency] at the end of the 1990s with the guys from Despegar.com here in Latin America. I worked for Despegar from '98 to '99 and we started to build the OTA industry here in Brazil, in Latin America especially. The OTA industry, to us, was the beginning of the Internet years here. It is so funny looking back and very interesting because we had some barriers to be overcome. For example, I remember now that in that period personal computers and Internet communications were so difficult in terms of customers having personal computers in their houses. There were no computers during that period -- it was so expensive here in Brazil. Another barrier, for example, is that this was the period when the customers started to carry credit cards.

The second barrier that we had here was in terms of market acquisition. Because in that period, Google didn’t exist. In terms of the marketing online, we used Yahoo, AltaVista, and these kinds of search engines. We used offline marketing like print marketing, including newspapers, magazines, and TV advertising because online marketing was so small in that period.

One thing, in looking back, is very funny in terms of the operation. Because in that period, the website, was only something for lead-generation. Everything was done totally offline by our operations teams. For example, after the customers finalized the process to buy online -- a ticket, for example -- we had to check with the GDS [global distribution system] in Sabre or Amadeus to see if everything was OK with the booking. Then we had to call to each customer to request some information about his or her credit card. Every customer had to send to us the documentation filled out in a form, a copy of the credit card, a copy of the ID, front and back, and send it to us by fax.

Skift: How were people booking hotels and booking airline tickets before the Internet?

Rossato: Just by phone here in Brazil. Just by phone. After Decolar started, the bookings in the beginning were 95 percent by phone and just 5 percent online. Because it was a totally manual process in terms of the ticketing. For example, there were printed tickets. Not e-ticket like nowadays and another challenge for us was that you had to deliver the ticket by courier or by traditional mail. It's a crazy thing nowadays thinking about what was happening in the beginning years here.

Rossato: They didn’t believe in us. They didn’t understand our project. They opened access in the end. We just started the process. Some airlines that existed in that period, like Varig or other airlines that had their operations here in Brazil back then, don’t exist anymore. In the last 17 to 18 years, the whole market would change. These old airlines don’t exist anymore.

Skift: Right, so how did consumers react? How quickly did it take to build up a customer base?

Rossato: It definitely took awhile to acquire customers in that period. Most of the customers made a lot of calls, I remember. They made a lot of the calls to check if the company existed. They were so uncomfortable to shop online. Because it was uncomfortable for them to put their credit card on the website, on the Internet. We had to do things to really make it comfortable for our customers like saying that this is a big company, we had big investors inside the company, that we are a true company so people felt comfortable to use their credit cards on the website.

Skift: When did that start to change? When did consumers start to trust you and you got a bigger audience?

Rossato: Internet adoption -- when the people start to buy anything on the Internet -- and e-commerce here in Brazil got better after 2005 to 2006.

Skift: In the late 1990s, did you start to have online competitors in Brazil?

Rossato: We had smaller competitors. They closed their doors when the bubble crash happened. It was the very early stage in terms of the online Internet in the end of the 90s. In the beginning of 2000, Rumbo, an online Spanish company, started and then they closed their doors around 2003 to 2004 when the brand was changed into Submarino Viagens. Submarino is part of the B2W holding company in Brazil. It's a retail company.

Skift: What about elsewhere in Latin America? Were there other big online companies starting to appear?

Alex Todres: [Booking.com and Expedia entered Brazil] maybe four years ago. [In the beginning of Decolar] something that was interesting is that people, after they bought a ticket from your site, they went to your office to be sure that the tickets existed. I remember even after 2010, we had some place in the office to receive the customers who buy online because the people want to go there or change the way that they pay or things like that. It was an online operation but with a very strong offline structure.

Stephan Ekbergh, who founded Travelstart in Sweden in 1999, got bored on vacation in South Africa in 2004. So he did what just about anyone would do when looking for fun on vacation -- he phoned local global distribution system contacts and invited them to lunch.

Ekbergh learned about the state of the travel market there and found out that offline players were consolidating and there were a handful of online players, such as LastMinute.com and Kalahari.com, but none were really making a mark and successfully focusing on online travel.

"I thought to myself, ‘This will be like stealing candy from a kid here,’ Ekbergh says. "I think you need to have a real focus if you want to succeed in online, and I thought these guys were never going to have that focus. So I decided tthere and then on that vacation that we were going to set up shop in South Africa."

He moved to South Africa in 2004 and started up Travelstart's operations in Capetown in 2006. Ekbergh soon discovered that banks and airlines had little understanding of online payments or online travel agencies, respectively, and it was a struggle to find reliable Internet service providers.

Skift: I was hoping you might be able to shed some light about what was going on before Travelstart got under way and how you guys started. Can you fill in some of the blanks?

Stephan Ekbergh: I started Travelstart in Sweden in '99. We went here on vacation in 2004. I'd been in South Africa about 15 times or so since the '80s so I knew it pretty well. We went on a mission to see if we could emigrate there. Actually, I wasn't thinking about launching Travelstart in South Africa when we went there on vacation because I thought this isn't going to go anywhere in a million years.

After about a week or so of vacation, I was really bored. So I started to call some of the GDSs [global distribution systems] and inviting them for lunch because, you know, there was nothing going on. And very quickly I realized that the whole industry was sleeping in South Africa; there wasn't really an online space to talk about. I knew Lastminute.com very well because we had been working with them for a couple of years in Scandinavia. I knew them very well, and I knew who their country managers were in South Africa. I also knew that they had spent about $2 million trying to break into the mainstream here, but it was also a joint venture with one of the conglomerates. Travel.com.au also tried to break into the market through joint venture with one of the other conglomerates in the country, and they were literally doing nothing.

Skift: When you say conglomerates, do you mean travel agencies?

Ekbergh: One of the findings in my amateurish research down in South Africa in 2004 was that the GDSs said everything is consolidated here; there's nothing more to do with the industry in this country. Basically both Tourvest and Bidvest, along with some other groups, had already bought a bunch of agencies so I'd say about four groupings had sort of cornered the market in South Africa. Then of course there was Flight Centre, which was a standalone.

I thought to myself, ‘This will be like stealing candy from a kid here.’ I think you need to have a real focus if you want to succeed in online, and I thought these guys were never going to have that focus. So I decided there and then on that vacation that we were going to set up shop in South Africa. That was the chain of events, basically. We moved here in 2004, but only got going in 2006 because we had so much to do on the home front in Scandinavia. We were very busy with that so we didn't manage to set anything up until 2006.

In South Africa, there were a few companies [including traditional travel agencies] that offered off-the-shelf booking engines, but I didn't really count that as any type of competition.

Skift: The consumers weren't gravitating towards it?

Ekbergh: No, apart from Kalahari.com, which in those days was the Amazon of South Africa, was around but also very weak. The service was weak, and the products were crappy. I just thought to myself that at some point something's going to happen there.

And I could do this because I controlled my company. I did it because I thought it was fun. I also thought I've always had this love for Africa in general. So I thought at one point, everybody's going to be buy their trips online here as well, so let's see if we can have the patience enough until it happens.

Skift: Where were you born?

Ekbergh: I was born in Sweden.

Skift: But you just had a love for Africa?

Ekbergh: I came to Africa the first time in the '80s. I've been working in travel most of my life so when the Internet started to happen in the mid-'90s, I of course knew that was going to be the future. So that's why we started. Lo and behold, what I found out very early was that I could see different dynamics in Africa than what I saw in Europe, and I thought, ‘Man, here you can bring a lot of change to the consumers and to the suppliers, both with distributions and payment.’ Even up until very recently, most of the European airlines servicing out of South Africa, you could book online but you had to go to the airport to pay. That was only recently changed, you know! It's so retarded in many of the countries here that there is a lot to do. You have to do a lot of heavy lifting.

Skift: For hotels, the same thing obviously.

Ekbergh: I'll tell you what the problem with hotels is. Most of the hotels -- if you take away the big chains (which there aren't many) -- they haven't got any PMS [property management system]. They don't have any proper technology to service that. There is a big, white, empty sheet in many cases. Of course the Hiltons, and all these guys have got it all covered. Booking and Expedia, they have them. I haven't done any studies, but I would probably guess that at least 80 percent of the accommodations in Africa are not covered by these guys. They [hotels] have very crappy, if any, technology.

Skift: So Travelstart started with ...

Ekbergh: We started with flights. We were the first ones to do that right; we had all the low cost carriers. You could pay anyway you want - cash, credit card or EFT [electronic funds transfer] or whatever. And then we bundled up with really focused and great service.

There were tons of challenges at first, of course. Talking to banks about accepting payments online was ridiculous. They literally had no idea what we were talking about. Getting ISPs [Internet Service Providers] that offered Internet that had uptime of over 70 percent was literally impossible. Internet was extremely expensive back then. It's still expensive, but back then it was extremely expensive. Then getting airlines to understand what we were doing was virtually impossible. We had to scrape the airlines [their websites] because they didn't have any APIs and they definitely didn't want to work with an OTA if there was such a thing. We scraped them, and they tried to sue us, they tried to threaten us, and all kinds of things; that was of course a challenge.

Skift: What was your resolution to that?

Ekbergh: Sales figures are always a good resolution. We showed them that we could sell, and then they were quiet.

Skift: What were the challenges in getting consumer adoption?

Ekbergh: That wasn't the problem at all. Actually, it took off literally immediately. The problem in the beginning was how to get to them because offline didn't work well at all. We had to create an ecosystem of affiliates and train them how to make money. That took us a few months and then we got about 10 or 20 really good anchor affiliates that could start to drive a lot of traffic to us. They made a lot of money; some of them became really rich, actually. We did that really well.

Skift News Editor Dennis Schaal interviewed more than two dozen people who were instrumental to the early days of online travel booking. In the process he recorded more than 24 hours of interviews upon which this series was based. In addition to the interviews, Skift collected archival materials from participants and institutions involved. Those items appear here with their permission.

To produce this feature, Schaal was aided by members of Skift's editorial, production, and development team. That team included (in alphabetical order):