You need to figure out what it costs you to support each user, make a ballpark of how many free users will upgrade (it's less than 2% for most sites that share their numbers, including my own W3Counter.com), and do the math to see if freemium will work for you. You might find that you won't be profitable if your cost to support the free user base is more than the paying users can cover at prices they're willing to pay. You can't just make the price arbitrarily high, sometimes freemium really doesn't work for a business.
W3Counter barely gets by as a freemium service. The paying users cover the servers it takes to run the site, but not enough more to pay even a single person's salary. I'm always waiting until service degrades to the point someone starts complaining before I add more hardware, otherwise enough new users wouldn't have started subscriptions to pay for the expansion.
The next service I built, W3ROI, does not offer a free plan. It's harder to build traffic to the site, and there are less people talking about it on blogs and social networks, but each new user it adds pays for themselves. There is never a risk of not being able to afford the hardware to provide a good quality of service.