← all stories

After the Web · Era 3 · The early Web

1996–2018stock crash in 2000; in 2004–2005 it bought Active Hotels…

Priceline.com

An American site where the buyer named their own price for a plane ticket or a hotel, and the seller decided whether to take it. It soared on the stock market in 1999, lost almost everything in 2000, and then grew on the purchase of a European hotel booking service that became Booking.com.

Founders Jay S. Walker (founder, author of the patent) · Richard S. Braddock (CEO from 1998) · Jeffery H. Boyd (CEO from 2002-11-25)
Domains priceline.com
travelmarketplacepricingipodotcom-crash

In brief

Priceline was an American site where the buyer named their own price for a plane ticket or a hotel room, and the seller decided whether to agree. In March 1999 the shares of a company that had opened less than a year earlier and was selling tickets at a negative gross margin soared from $16 to $69 on the very first day, according to Motley Fool estimate. In the fourth quarter of 2000 they fell as low as $1.125. Today Priceline is worth telling about because of its second life: after the crash the company bought two European hotel booking services, one of which was Booking.com, and eventually became Booking Holdings.

How it started (the founders)

The idea belonged to Jay Walker, founder of the research company Walker Digital. The patent on "buyer-driven commerce" was filed on September 4, 1996, in the name of Walker Asset Management and issued on August 11, 1998, as US 5,794,207. Walker Digital transferred the rights to the patent and the related intellectual property to the company, receiving back a perpetual non-exclusive license; after the IPO, Walker Digital owned about 5.3%. The company itself was created on July 18, 1997, and the service opened on April 6, 1998, selling airline tickets.

The mechanics were simple: the buyer names a price and commits to buy if the seller agrees; Priceline earns the difference between the named price and the seller's fare. Airlines get extra revenue on seats that would otherwise have flown empty, and their published fares stay untouched. Over the course of 1998, tickets were joined by cars (a test, in July), hotels (October), and mortgages (January 1999).

Walker himself did not run the company for long: in July 1998, three months after launch, he was replaced as chairman and CEO by Richard Braddock, an outsider (the prospectus biography gives August 1998; in 1973–1993 he worked at Citicorp and Citibank, including as president and chief operating officer). Later, from November 25, 2002, Jeffery Boyd, who had previously shared the duties with Braddock, became sole CEO. Walker left the post of vice chairman and the board at the end of December 2000 to focus on Walker Digital.

Year-by-year timeline

Lesser-known but significant facts

  1. The huge 1999 loss was an accounting one. The net loss came to $1.055 billion, of which $998.8 million was "warrant costs," that is, warrants issued to suppliers, not operating losses. Operationally, the company was already earning gross profit of $57.8 million in 1999.
  2. $189 million in income that was never there. In the fourth quarter of 1999 Priceline booked about $189 million in income for a warrant on a stake in WebHouse Club; in the third quarter of 2000, when the partner shut down, the whole warrant was written off. A single accounting entry flattered 1999 and made the 2000 loss heavier.
  3. Walker sold a large block of shares almost at the bottom. Hutchison and Cheung Kong bought 11.3 million shares from him for about $24 million, roughly $2.1 apiece (my calculation). For comparison: at the IPO a share cost $16, and at the peak of the second quarter of 1999, $162.375.
  4. The 2004 and 2005 deals were two purchases, not one. Active Hotels (about $161 million, 2004) and Bookings B.V. ($133 million, 2005) together gave, according to Priceline, about 18,000 properties with exclusive rates. In October 2006 the two companies were combined under the Booking.com brand (the paired dossier), which is why retellings easily merge the two deals into one.

Legend vs. the record

The first growth lever

The first lever was not "name your own price" but the sale of unsold seats. From launch to the end of 1998 the company collected offers for about 1.9 million tickets ($400 million in demand) and sold about 134,900 tickets for $30.4 million in revenue. Conversion was about 7% by tickets (134,900 out of 1.9 million, my calculation) estimate: the flow of bids showed how much buyers were really willing to pay. Taken together, the tickets sold for less than they cost: gross profit for 1998 was negative.

The second, far more significant lever turned out to be the turn of 2002–2005: hotels, and then European services with an agency commission instead of a spread. It was the international, mostly European, business that grew to 55–58% of bookings over 2006–2008.

The paired story

Parallels today (projects from the catalog)

What a builder can take from this in 2026

  1. A patent and a brand are no substitute for margin. In 1998 Priceline, taken as a whole, sold tickets below cost; as long as the spread is negative, scale does not cure the loss.
  2. Do not license your model to a partner without checking the partner's economics. The whole WebHouse Club story is a $189 million warrant booked as income and then written off.
  3. When the core is broken, look at where things are already growing. After 2000 what grew was not "name your own price" but hotels and Europe.
  4. Buy what complements your model, not what copies it. The international, mostly European, agency business delivered more than two thirds of operating profit by 2007 on 55% of bookings.
  5. An accounting loss and an operating loss are different stories. In 1999 the $998.8 million in warrants looked like a catastrophe, but gross profit was positive.

Discrepancies and what we could not verify

Sources (primary first)

Internet History Keeper

Every historical dossier here is free and open, and it will stay that way. If you want the series to keep going (new dossiers, checks against primary sources, the English edition), support it with a subscription for $9.90 a month. It is support, not access, and you can cancel at any time.

Become a keeper — $9.90 a month →

The same thing, about today

The same breakdowns, but of projects launching right now: what the product is, where the first users came from, how they charge. The card is free, the full dossier is $5 (the dossier itself is written in Russian).

Browse the dossier catalog →