It's called selling a product for less than it costs. A variable cost is one that is dependent on the the # of units sold. So, obviously, the more one sells, the more one loses money. Think of it this way: you could make a million dollars right now, really fast, by selling gasoline (in the US) for $1 a gallon. You'd sell a million gallons. And lose $2 million in the process. The more you sell, the more you lose because your costs (which are dependent on your total volume) increase with your revenues. This is why you can't confuse revenue with income.
Amazon had that model for many years of operation. But it went public before it ran out of money. Call it the "cost of acquiring the market". The problem was that Amazon had a much bigger market to work with.
The more product Amazon sold, the more money they'd make (or more accurately, the less money they'd lose).
On the other hand, Ecomom never had a gross profit. The more product they sold, the more money they lost. This is the situation you describe in the gasoline scenario. I suspect the gasoline model was not their strategy by choice, but by accident. The controller's assertion is that it's the runaway discounts that did them in. 50% discounts intended to be for one-time-use-only were used on almost all orders.
Whether it was a long-term strategy or a fundamental mistake, whatever they did is not the Amazon model.
Rather famously, Amazon in its early days included the cost of fulfillment (shipping, warehousing, etc.) in Sales and Marketing expense, not COGS. As a result, gross margins were thought to be somewhat inflated.
Apparently they also counted their equity investments in the stock of companies including Webvan and Sotheby's as cash and marketable securities.
At the time, analysts were worried about what all this implied for operating cash flow and gross margin.
What they were selling was $40 gift certificates for $20.
Then they sold a $30 good for a $40 gift certificate.
They recorded this as a $40 sale with a $30 cost and $20 marketing. The implication that you can eventually cut back on the marketing, which is obviously false.
IMO, the $40 should never appear in anything that could be construed as revenues. Sure, you can work it such that you'll end up with the proper ($10) loss, but they never received an actual $40, their only (well, majority) revenue appeared in the form of $20 payments from the group selling the gift certificates.
And Amazon is one of the very few success stories of its era. Remember that there were dozens of very well capitalized venture funded retailers working one part of the market or another. Remember pets.com?
This is a big part of the reason why the current startup environment "feels" the same as the dot com crash. Valuations and funding have gotten completely divorced from the business fundamentals. YC startups are being handed blank checks, and no one even remembers "ramen profitable".
At this point, IMHO, it's undeniably unsustainable. Whether it constitutes a "bubble" or not is something we won't know until it pops.
Amazon had that model for many years of operation. But it went public before it ran out of money. Call it the "cost of acquiring the market". The problem was that Amazon had a much bigger market to work with.