One of the biggest challenges is keeping engineering focused on "maintenance" and bug fixes for the current product. I once worked for a secure email company who would get an alpha release ready and then immediately start developing the next release. Needless to say it was hard to get customers to use the product (well it did work *most* of the time).
When it comes to prioritization just remember this one truth:
"The best customers are the ones you already have".
From a pure dollar and cents perspective, current customer revenue is cheaper than new customer acquisition.