Idiano - a good question. It's the industry that dictates. My take on it is 'in the beginning', and to keep the process and metric as easy as possible for people to understand, compare and calculate overall IF was the 'king' value. That tradition is still strong. The other comparators i.e. h-index (traditionally more related to individuals) have come more visible and into play more recently and are useful as a more accurate measure of how overall IF is broken down - and what it's true value is overall. However, even that is open to abuse. For instance, a journal may have a low IF but a higher citation rate - especially if they practice the ethically dubious practice of 'forcing' authors to cite their own journal works. In this case, they will promote the highest rating they can - which may not be IF. That becomes confusing. If there were 'industry standards' to regulate this that would be helpful - but I doubt that will ever come.