Really really good idea to attack something like this as early as possible...
In the spirit of failing early, this is a great way to figure out where people really what what you think they do. If you can't get a few thousand email addresses, you quite likely have a bad idea or haven't figured out how to talk about it yet.
The phone numbers are a particularly interesting idea, but how did you intend/did you get in touch with all the people after launch? Did you and a few others manually sit around and call 20,000 people?
Every person who uses your service is important, if you can call them personally, great. if you have to email so you can have some hours to sleep then do that. i didn't call but suggested that would be a fun idea to try.
after launch vip emails were sent and then emails were sent to different buckets of people.
in case anyone was curious i started doing this and more 9 months before the techcrunch40 competition. i would say 4 months is a good headway to get this going. i think
a) some sites dont need outreach or trust to be built since their initial users could theoretically spread the product
b)you want to build the relationships and things earlier than you need them.
Solve a problem and find the people to whom that solution is most valuable. Start there, cater to them, make them passionate, and voila: there's your base!
In the spirit of failing early, this is a great way to figure out where people really what what you think they do. If you can't get a few thousand email addresses, you quite likely have a bad idea or haven't figured out how to talk about it yet.