A service that processes tweets from the firehose in realtime. Assuming you decide to process everything day 1, thats a case where the overhead from running your app could greatly outweigh the revenue it generates.
I mean, I guess? I would question that business and processing model heavily, though. Most sustainable business models show growing from some kind of smaller MVP to a larger full-featured application.
Significant overhead costs to go from nothing to steady-state instantaneously would be enough to scare me.
Significant overhead costs to go from nothing to steady-state instantaneously would be enough to scare me.