Don't just apply to journals and books. They can be used to gather information on presentations, data sets, software and other research outputs too
Allow measurement of early reaction to papers because social media, for example, can provide feedback on research in less time than citations in journal articles
Can demonstrate broader impact because they allow you to show how people from outside of academia have engaged with your work
You can follow the trail of who has mentioned or used your research in order to discover new papers, peers or collaborators
Cons
Alternative metrics look at how many times research is used or mentioned but not at the context. As a result, a simple count cannot be used to demonstrate the value of research alone
A piece could be blogged about many times due to negative feedback
Some people feel that articles get mentioned on social media because they relate to popular topics, not because they are examples of good research
Alternative metrics can be abused by individuals who want to artificially increase their altmetric scores
Some people question the significance of the processes altmetrics measure, arguing that Twitter, for example, is too brief a place for “serious” academic conversations and that tweets are not a useful measure of the value of a paper. For example, see Colquhoun, D. Why you should ignore altmetrics and other bibliometric nightmares. If so many articles are behind paywalls – how can people outside academia re-tweeting articles read the full article in order to verify its content and quality?