Basically review the papers that are related to the respective theory or perspective and criticize and/or support their points. You may want to check "Review" or "Meta-Review" papers for guidance.
iportant is that you find a theoretical gap. It means you look through a lot of papers and reviews as Ahmed suggested. Then you find something what was not teorized or theorized wrongly. This is the hardest work in all this story.
There are of course recommendations concerning structure of the paper, its elements and the journal you would like to send it to. The best recommendation is to download 3 top conceptual (theoretic) papers from the journal and follow their structure.
In such cases it is important the gaps in theory. To illustrate can consumer buying behaviour models (developed for off-line marketing) be applied to online marketing? Meta analysis should be carried out to formulate the same.