The Wall Street Journal finally picks up on the freemium trend (it's only been 2-3 years, right?):
Read more at the Wall Street Journal ›
The problem is, anyone reading the article is probably under the impression that what succeeds and what doesn't is an art, not a science. It quotes one entrepreneur as saying:
"He blames his woes on the freemium model, which he likens to a Samurai sword. "Unless you're a master at using it, you can cut your arm off," he said."
People treat freemium like a generally applicable strategy. The truth is that only certain products can pull off freemium.
You can't make freemium work based on features. A crippled version of a full product is a terrible lead generation mechanism. You're not putting your best foot forward, and you're asking people to pay for features they haven't experienced.
Quantity limits are better, but hard to pull off. Potential customers twist themselves into knots to avoid user limits. Rate limits are better (e.g. up to 50 invoices/month) but are tricky to set.
The ideal freemium product bases upgrades on a persistent, cumulative metric. That is, something that always increases with usage, and is difficult to game.
It's no surprise that some of the top freemium companies like Dropbox and Evernote specialize in storage--storage has both these qualities. I even tell my companies, "When in doubt, charge for storage."
Before you simply sprinkle freemium pixie dust on your startup, ask yourself if you can base the model on a persistent, cumulative metric. If not, no amount of freemium expertise can make it work.