Webvan

What did it do?

Webvan was an online grocery company based in Foster City; customers placed orders on the internet and chose a 30-minute window for delivery. The company built its own warehouses and fleet of lorries, served 10 regions of the US at its peak and aimed to reach 26 cities by 2001.

Why did it shut down?

Under pressure from investors to secure first-mover advantage, the company expanded into many cities without proving its business model in its first market. It placed a $1 billion order with Bechtel for warehouses, and infrastructure built from scratch created very high fixed costs. In 2000 it ran up $525.4 million in expenses against $178.5 million in sales. Targeting price-sensitive customers kept margins thin, and none of the senior executives had any supermarket experience. With losses exceeding $800 million, the company ceased operations and filed for bankruptcy in July 2001.

The story

On the day it went public in November 1999, Webvan was worth more than $4.8 billion. Yet the company's total revenue up to that point was just $395,000, and its accumulated losses exceeded $50 million. That gulf foreshadowed one of the most expensive lessons of the dotcom era.

Webvan was founded in 1996 by Louis Borders, co-founder of the Borders bookshop chain. The idea was simple: customers would place their grocery orders online and choose a 30-minute window for delivery. Benchmark Capital, Sequoia Capital and Borders each invested $3.5 million in 1997; then Softbank put in $160.3 million and Goldman Sachs $50 million, and Sequoia added another $50 million. Venture funding totalled more than $396 million, and the IPO brought in $375 million. George Shaheen, the former head of Andersen Consulting, took the helm.

Investors wanted rapid growth to secure first-mover advantage. Webvan started taking orders in the San Francisco Bay Area in June 1999, placed a $1 billion order with Bechtel for its warehouses, built its own fleet of lorries and in 2000 bought its loss-making rival HomeGrocer in a $1.2 billion share deal. At its peak the company served 10 US regions, and in 2000 it had $525.4 million in expenses against $178.5 million in sales. Nobody in senior management had experience of running supermarkets, and the price-sensitive customers it targeted did not leave enough margin to carry such expensive infrastructure.

Shaheen resigned in April 2001. In July 2001 Webvan ceased operations and filed for bankruptcy, leaving behind losses of more than $800 million and 2,000 employees out of work. Non-perishable food was donated to local food banks, and some executives later moved to Amazon.

Webvan's real mistake lay not in the idea but in the sequencing. The model was rolled out nationwide through warehouses built from scratch before it had become profitable in its first city. Peapod, which used existing supermarkets' infrastructure, survived the dotcom crash, while Webvan passed into history. When rapid grocery delivery took off again in the 2020s, Webvan's story became the warning most often cited to those start-ups.

Lessons for founders

• Prove profitability per order in your first city before opening in a new one; scaling fast multiplies losses.
• Using existing supermarket and warehouse infrastructure takes far less capital than building automated warehouses from scratch.
• Make sure at least one person in the leadership team brings hands-on operational knowledge of the sector; general management experience is no substitute for grocery logistics.

Sources

1. Webvan (Wikipedia)
2. Webvan goes under / Online grocer shuts down -- $830 million lost, 2,000 workers fired (San Francisco Chronicle, 2001-07-09)
3. Webvan Stock Price Closes 65% Above Initial Offering (The New York Times, 1999-11-06)
4. https://www.wikidata.org/wiki/Q17087287
5. Wikipedia (English)

This section is compiled from public sources (Wikidata, news archives and company announcements). Figures are given as they appear in those sources; unknown fields are left blank. If you spot a mistake in an entry, you can send us a correction.

Is something in this entry wrong? Request a correction → Method