"To say that because a researcher is publishing in a certain journal, he or she is more influential or deserves more credit is not necessarily true. There are many other variables to consider." (Jim Testa, a researcher for ThomsonReuters Scientific in a 2008 interview)
1. Journal impact metrics do not assess the quality of individual articles, their importance, or usefulness! A single factor is not sufficient for evaluating an author's work in total.
2. Journal Impact Factor, Eigenfactor, and CiteScore cannot be used to compare journals from different subject fields.
3. Citation-based journal impact metrics are higher in disciplines in which rapid citation is the standard. Fields with a more durable literature, such as mathematics, have a much smaller fraction of short-term citations and hence lower journal impact factors.
4. Journal impact metrics can be manipulated (The Journal that could not stop citing itself ; Coercive Citation in Academic Publishing)