Google announced its own wireless network today. That sounds pretty ambitious and expensive. Plus you get what arguably the best, or least the most flexible, data plan offered by anyone, even trumping T-Mobile’s "Uncarrier" approach. But building a mobile network is a massive expense. Is Google betting everything on this new project?

Well, no. Instead of spending billions to create its own network, Google is piggybacking on Sprint and T-Mobile. It’s called an MVNO — mobile virtual network operator. The model has an up and down history in the US. Around 2005 every tech blogger worth their salt thought they were the next big thing. A few years later we were strolling through the MVNO graveyard (RIP Disney Mobile, ESPN MVP, Helio, and many more).

The basic arrangement of an MVNO works like this: a big carrier sells its unused network capacity at wholesale prices to a smaller company that can capture a part of the market that is less profitable or more difficult to reach with its existing plans. "It's a balancing act," T-Mobile’s Doug Chartier said back in 2013. "In a perfect world, the MVNO uses enough network to create a contribution to the organization without being detrimental. That's the holy grail."

Typically, MVNOs have targeted lower income sections of the market, offering the ability to work with cheap monthly fees or pre-paid devices, but with less widespread coverage and without the ability to provide premier devices. Project Fi seems different, at least in the sense that you need to own an expensive flagship phone to participate and that it would work for global travelers in 120 different countries.

But the basic principles still apply. Sprint and T-Mobile get to grow their subscriber base (although it's not clear how they will tally the shared customers), and in this case add early adopters who probably consume an above average amount of data each month. Google incentivizes people to rush out and buy a Nexus (I’m looking at you @backlon) and perhaps gets phone manufacturers to build more cross-network compatible phones to boot.

More importantly — and maybe this is Google’s main goal, Fi could push the entire industry toward a model where data is cheaper and more abundant. Just look at the impacts Google Fiber service had on speeds and prices in cities where it was offered. Project Fi is starting with an "Early Access Program" and is only compatible with one phone, so you could argue that this program isn’t meant to make money. Sundar Pichai himself said, "We don't intend to be a carrier at scale," and it seems like he meant it.

If you’re looking at business motivations, it makes sense to see Fi as an example meant to pressure the actual, bigger players into acting in a way that is good for Google. The side benefit is that simple data plans, seamless Wi-Fi transitions, and low prices aren’t just good for Google, they’re good for consumers.

From Fiber to Loon, Google has been operating across a wide range of new industries on the assumption that more internet access will benefit its business in the long term, especially if it plays a role in delivering that access.

The gamble Sprint and T-Mobile are taking is that they can roll up some new customers, get a lot of free press, and associate their brand with Google and cutting-edge innovation. If Fi is a massive success and forces them to begin offering a competitive plan, well, they will cross that bridge when they get to it.