WebVan isn't the right analogy - a real estate bubble is a better one.

In the .com bubble, no one was really sure how things would shake out, so there were lots of companies who wouldn't have succeeded no matter how big they got (the "losing money on every sale but making it up in volume" business plan). When the .com bubble crashed, real estate was seen as a much stronger investment because no matter how low it went, it was real "stuff" that had a sort of intrinsic value - it may go down, but it wouldn't go to 0. A similar thing happened in Tokyo real estate circa early 90s. A Tokyo apartment still has a lot of value - you'd probably think it very expensive - but it's still 80% cheaper than it was at its peak.

The problem with lots of companies now isn't that they won't ever be profitable, or that they don't have realistic business plans, it's just that their valuation is way more than their earning stream will ever support. They are valued at "we'll eventually take over the market" prices, even though they'll only ever reach niche market status.

This is a great explanation. Remember that the value in a stock is that you're owning X% of a business just like if you owned 100% of a coffee shop.

If you buy a coffee shop for 10% @ $100,000 that makes $10k/year you just bought a company for a P/E of 10. The value of a stock in that company is in the growth of that 10k over time that is either paid as a dividend or retained and valued in to the price of the stock. If you don't make money, or make much less than what you could be expected to grow, their value will fall.

I'm not sure Uber is really who the article is talking about. Some companies really have proven their Unicorn status.

Everyone I know uses Uber a lot

Still possible for Uber to be overvalued though. Fedex, a proven business, has an EV of $39BN and FCF of ~$700MM. Uber has a $63BN valuation and loses money. Uber could take over the world. Or it might not. A lot of assumptions baked into that $63BN number. Will be interesting

Things are different than the webvan bubble: many of these companies have substantial revenues, selling into a market of 10 to 20 times as many internet users, much cheaper infrastructure, etc.

Webvan had $178.5 million of revenue in 2000, selling in only a few markets, and it had semi-coherent plans for growing its revenue quite a lot farther by penetrating existing markets more deeply and adding new ones.

Trouble was, Webvan's way of doing business bled cash at a horrific rate. The bigger it got, the more money it lost. Investors got tired of throwing more cash into the company because the cash-bleed problem couldn't be fixed.

I don't think Webvan is the last company ever to suffer from such problems -- or to be unable to save itself if/when investors go on strike.