Hard to say at this point given that this hasn't launched yet. I'll need to see what it looks like when it does its pilot in Israel.

Google's long-term play has a decent chance of involving self-driving cars and monetizing via advertising and data with some fee from the ride. Nothing says "open up your wallets" to large brick and mortar advertisers like "we literally drove the customer to your doorstep."

Getting users hooked and getting this tested via commute data is a great approach before branching out to something else. It also lets users browse the web, and Google likely has clear visibility into the search behavior of users searching while in their rides to determine the incremental ad revenue this enabled.

Google doesn't need to offer this for free to provide a great experience at a lower price than any current competitors like Uber or Lyft. They have substantial revenue from other areas to fuel this and so can price their competition out of the market. Other companies do the same thing to squeeze into a dominant position in a market they want to enter.