I've worked for a few video game 'startups' that tanked or are in the process of tanking. Here are some pitfalls to avoid:
* Make sure you focus on the company aspect of things. Why are you making games? If the answer isn't 'to make money' you're doing it wrong - make them as a hobby instead. Starting a video game company because you have great game ideas, or because making games is something you love to do is a bad, bad idea. The reason for this is that at the end of the day, if you don't focus on the financial aspect of things, your company.will.fail. Most game companies fail because they're living month to month off of publisher's paychecks without trying to establish revenue streams of their own (other than royalties).
* Speaking of royalties. If you're getting funding through a video game publisher, make sure you understand how the payout will happen, how your advance is being payed off, etc. You probably won't have much negotiating power here, and this isn't really a route I'd recommend unless you _really_ need the funding.
* Start small, don't try to make a game engine right out of the gate. Put something simple together and iterate on it. If it turns out you're making a lot of the same type of game over and over, then think about abstracting an engine out of it.
* Reconcile the awesome game that's in your head, with the awesome game you can make given the time and resources available. Don't overcommit, you'll just end up implicitly cutting features instead of explicitly cutting them, and the game will suffer.
* Don't forget about the importance of marketing and timing. Great games fail _all the time_ because they weren't marketed well, or because they released and then got buried by the latest AAA or hyped up title.
* Make sure you focus on the company aspect of things. Why are you making games? If the answer isn't 'to make money' you're doing it wrong - make them as a hobby instead. Starting a video game company because you have great game ideas, or because making games is something you love to do is a bad, bad idea. The reason for this is that at the end of the day, if you don't focus on the financial aspect of things, your company.will.fail. Most game companies fail because they're living month to month off of publisher's paychecks without trying to establish revenue streams of their own (other than royalties).
* Speaking of royalties. If you're getting funding through a video game publisher, make sure you understand how the payout will happen, how your advance is being payed off, etc. You probably won't have much negotiating power here, and this isn't really a route I'd recommend unless you _really_ need the funding.
* Start small, don't try to make a game engine right out of the gate. Put something simple together and iterate on it. If it turns out you're making a lot of the same type of game over and over, then think about abstracting an engine out of it.
* Reconcile the awesome game that's in your head, with the awesome game you can make given the time and resources available. Don't overcommit, you'll just end up implicitly cutting features instead of explicitly cutting them, and the game will suffer.
* Don't forget about the importance of marketing and timing. Great games fail _all the time_ because they weren't marketed well, or because they released and then got buried by the latest AAA or hyped up title.