As the price war among Android OEMs kicked from high gear into ludicrous gear and anyone with a grasp of how smartphones are built began wondering how OEMs planned on making money, a few brash new entrants to the space had some exciting new ideas. Instead of the tried and true “let’s sell it for more money than it cost us to make it” business model, these bold corporate innovators would give the phone away—sell it at or close to zero profit margin—and make up for it with revenue generated from the user after the sale.

Foremost among these was Xiaomi, that onetime darling of investors and commentariat. Their executives promised they would build up a user base of people using their low-cost, no-margin hardware and then make money on selling various “Internet Services” to that user base, and investors ate it up. Xiaomi was hailed as “The Apple of China” and attracted considerable investment at a $45 billion valuation, making it (for a time) the biggest privately held startup in the world.

As 2015 draws to a close, Xiaomi has sold a whole lot of phones (although not as many as expected), but what revenue has been generated by these “Internet services”?

¯\_(ツ)_/¯

The “Internet Services” placeholder for making money reminds me of the South Park “Underwear gnomes” episode, in which one kid’s underwear was being stolen by gnomes who had a plan that was as follows:

1. Collect Underpants
2. ???
3. Profit

In the case of Xiaomi et al, the Underwear Gnomes business model might look like

1. Sell phones at at no profit
2. “Internet Services”
3. Profit

To date, the much-vaunted “Internet services” from Xiaomi have yet to materialize, and I don’t think they will.

As it turns out, OEMs aren’t good at making “Internet services”—and even if they were, I don’t think they are well-placed to do so anyway.

If there is a profitable opportunity to be had in the mobile ecosystem, what are the odds that an OEM would beat a quicker, dedicated (and well-funded) 3^rd party to the punch?

Even Apple (a phone maker with better software competence than most, and more money than all)  can’t beat the sheer diversity and quality of dedicated companies focusing on their singular products and services—hence the “Apple crap” folder of Apple apps/services that easy to replace but impossible to remove that you will find on roughly 10 out of 10 iPhones.

If even Apple struggles to make apps that you’d want to use for free, imagine how much harder it is to make an app or service so compelling that a user would want to pay Xiaomi money for it.

Most OEMs are slowly learning this lesson—witness Sony throwing in the towel on their own “Music Unlimited” app for their Xperia phones and swapping it out for Spotify. And that’s fine for Sony, which at least theoretically plans on selling phones for more than it cost to make them. But what about Xiaomi? What are they going to do to justify their sky-high valuation?

Amusingly (tragically?) Xiaomi’s underwear gnomes plan isn’t even a new idea. Among phone makers, you can see it as far back as Nokia’s catastrophically bad CEO (no, not the one you’re thinking of) proclaiming in 2007 that the world’s biggest phone maker was not to be a phone maker, but an “Internet services company” that just happened to be making phones.

That, of course, ended with Nokia being a company that makes neither.