Defining the size of random samples based upon expected mean differences

Question: How large do my (sub) samples have to be if I want a mean difference between two samples to be significant?

Essential factors for this calculation are:

1. The expected standard deviation

i.e. the distribution of the mean values measured in the samples. If the standard deviation is unknown, the rule of thumb is used:
(scale maximum - scale minimum)/3.

2. The expected mean differences

3. The desired significance level

i.e. the statistical probability that two means actually differ from one another.
Market research normally targets a significance level of 95%.