Everything you say about Groupon today applied to Amazon in the late 90s. They were accused of being a pyramid scheme. They were accused of bad accounting because they counted free shipping as a one time expense. They were never going to succeed. Fast forward 15 years and Borders is going out of business, Barnes and Noble is in retreat and Amazon moved beyond books to retailing everything - including the cloud.
My nature is to be a cynic, and I roll my eyes when people make up new metrics. I look for cash rather than an opinion on earnings. Even Generally Accepted Accounting Principles can be gamed, let alone non-GAAP metrics. But maybe this time they're right.
This isn't to say Groupon will succeed, but there is some precedent. Maybe they are more Amazon than eToys.