Statistical Significance

You need to assess what you mean by a significant difference as well as how willing you are for the test to be wrong in the final analysis. Are your outcomes qualitative or quantitative in nature? If qualitative then what % difference are are willing to call not significant. If quantitative then what marginal difference are you willing to call the "same" statistically. Then, once you identify the difference your looking to discern, your sample size will be based on the size of the type I error you're willing to live with.