I think that the mining contract notion is brilliant, but either it's not worth while, or this calculator is wrong.
You sell a 200Mhash/s contract for 3 months, at $270.
If I put those numbers into
http://bitcoinx.com/profit/index.php with no power costs (as that's what you'd be paying), at 15BTC to USD, I'd be in the hole $107.05.
It could be that mining simply isn't profitable or viable these days, but you'd really have to run around 350Mhash/s for it to be viable, according to that calculator. At your gigahash rates, I see losses of $535 after three months.
I doubt those numbers are precise, and I don't understand the mathematics behind hashing nor the forces involved. Does seem like it's not the best deal to partake in, however. Feel free to debate my calculated numbers, but I think the community could use some solid evidence of what profit numbers they'd be likely to receive running your service.
--sega01