I only found out about google fi recently. Signed up last week. If I had the ability to move to it today, I would try it. That being said I have a couple points of frustration already:
1) While the pricing is cheap-ish, it's not as cheap as it should be. In fact it's *more* expensive than my current plan with Ting (a Sprint MVNO). My wife and I are on a group plan and pay $48per month (on fi this will be 60$). The only reason I'm looking is because the coverage in our house is terrible and the sprint femtocell is an actual joke in hardware form. 20$ base fee for talk and text? That's extremely steep. No family plans/discounts? That's ridiculous. A current MVNO should *not* be able to offer services at a lower price than a new service that will have significantly lower operating costs. Fi will have a significant chunk of its calls/texts/data going over wifi (free for google other than scaling servers). Google's pricing of pay per use model is an excellent choice! However Ting has already embraced this and can offer at lower prices. If you're the new service on the block shouldn't you have a reason to switch?
There are a couple of points to be made here. First, since Google is paying for the actual price of the data, they aren't making money off of unused data. Even on Ting, from what I recall, you still buy "buckets" (even if the bucket can be quite small) of data and don't get credit back for any of the data you don't use. On top of this, Google is working with two different carriers and also has data roaming, which -- particularly with roaming -- costs more for the additional coverage.
It likely could be cheaper, if Google really wanted to, but the other side is they aren't wanting lots of customers as, since this is a "beta" or, as you put it, "proof of concept". Having a higher price per MB helps control the number of customers that will seek to switch to Project Fi.
2) Invite only? Still? I can understand this when the nexus 6 was the only supported phone. Clearly it's a proof of concept at that point. But this week google holds a conference and announces 2 more phones support google fi. Customers hype about the phones and google fi. Then customers realize they can't actually use those shiny new phones on google fi. This is a huge marketing failure on google's part as the customer gets hyped about the new phones then is told by google in the next breath "sorry only the cool kids are allowed on our network - you're an outcast". Huge. Marketing. Failure.
Since you seem to accept it was valid for invite only originally, I won't address that part of it. Instead, I'll say that the invite system now seems to be more to control the geographical location of where users sign up. For example, they may be using it to control the total number of customers they have in particular areas. I suspect this is particularly true in "fringe" areas where Sprint and/or T-Mobile coverage is weak.
3) Zero visibility into when invites will be completed (bear with me). They have a customized message when you hit the google fi page to give you an estimate of how long it will be. This was a great idea that failed in execution. Most people stare at "We're still determining your status. Check back here in a week for an update." message for weeks. If people are lucky enough to get a week range, they might watch it dwindle from 6-8 weeks, down to 1-2 weeks, and then skyrocketing back up to 12 weeks or more. What's the point of a status bar that doesn't communicate status? Get rid of it and don't misinform (frustrate) your customers.
I do believe much of the reason initially, for lack of clear ideas when invites would be extended, was to ensure Project Fi was up and running properly and not to overload their systems. I agree they should be through that phase, though they still may be trying to insure they don't grow Project Fi too quickly for their current size.
However, I think the invite "hold up" is now largely based on my answer to #2 -- how well T-Mobile and Sprint work in the area where the person is wanting service. Both T-Mobile and Sprint are doing a lot of work on improving their network and expanding LTE service into areas that may only be 2G currently. I suspect that, rather than just flat out decline people, they instead (for people in weak service areas) waiting for T-Mobile or Sprint to "fix" those areas. Since T-Mobile and Sprint likely aren't giving frequent updates, they can't give exact promises of when they might invite people in those areas.
My theory is based partially on how most seem to have gotten their invites by now, and many who have requested an invite in the last week have gotten the invite the same or next day.
Am I going to switch? The only reason I'm considering it is the poor reception in my house. If sprint had a femtocell that was worth it's weight in salt I wouldn't even be considering it because i have a better deal already. Despite the disadvantages listed above I'm still planning on trying it out if google ever lets me be one of the cool kids because I am willing to pay a little bit more for reliable wifi calling/texting. Then i won't have to mess with a femtocell at all, i can just use my wireless I'm already paying for.
Lastly, holding a conference and releasing sweet new phones for google fi and then telling everyone "dont worry we'll have all the invites out and completed by *mid-summer* is an epic marketing failure. By the time you release new phones to hype google fi it should no longer be invite - period. It should be open to the public.
This is additionally frustrating because google could have prevented some of this by letting invites expire. We all know there are people who have invites and aren't using them. Some people even *want* to give them away because they know others are waiting. If google were to expire the invite after 2 weeks, those invites would be in a constant cycle of being given to users that will use the service, provide cash flow, provide feedback and in general get this service up faster.
And your specific situation makes me think I have close to the right explanation. My understanding is that pretty much everyone that requested an invite got one. Perhaps in your case something has gone wrong -- I'd suggest you might try reaching Project Fi and let them know -- you may have one of those "growing pain" issues that they are trying to solve, part of the reason for the invite system.
OTOH, I suspect it might have something to do with being in a weak Sprint area and that T-Mobile may not be much better. Particularly if Sprint (or T-Mobile) had stated they are going to improve service in your area "soon", Fi may be waiting to issue an invitation until they know service in your area has been improved.