Significance testing on share of spend

Hi all,

can someone point me in the right direction for how i would work out stat sig of share of spend?

eg - 5 companies in market, ask ppl how much you spend with each for say phone bill. This will then give total spend for each of the companies, and total market spend, therefore can easily work out share of spend, but how to do sig testing on this? im told its not as easy as a z/t test of proportions.
Before you even begin to solve a stat. significance problem, you must ask yourself what is the alternative and null-hypothesis for the population parameter. It doesn't make sense to do a significance test when you don't know what is the null-hypothesis. So, in order to solve this problem, you have to know what the null-hypothesis is.