The main differences with Star Citizen are that it's

• Funded in advance
• Funded by people who have no say in how the product/company should work
• Massively overfunded

This means, CIG has no pressure to ship soon or even at all (if the project fails, they have no liability). They also have nobody telling them what to with the money. They have already made their profit.

I am not knocking CIG for this situation, but if you put it like this, it's easy to see why for each CIG out there, there are tens of thousands of games on crowdfunding sites that either

• Failed to raise funds
• Failed to get a decent company/legal structure running with the money they raised
• Failed to actually ever deliver anything in an usable state
• Are just pure scams

So as a general business model rather than just an insane stroke of luck, I don't think this is a good option.

A business model that only earns money after release (like the classic publisher-funded development model) is bad for the obvious cash-grabby and buggy reasons, but at least it consistently delivers games. Contrary to the "earn money before you start development" model that is enabled by crowdfunding, which in general does not deliver games.

In my (not very educated) opinion, early access is probably the best middle ground. You start off with little initial funding required, but by the time you turn to the crowd, you already have a working prototype and company structure. That makes it much more likely for the game to eventually be released in a full version. This option obviously comes with its own downsides as well, but many of my favourite games have been small studios or even individuals who use early acces to fund development.