You are right, sometimes it seems that it is something that is done to fulfill a requirement rather than something else. In favor, I would say that it provides a vision of how the author evaluates the included studies

One of the aims of the GRADE Working Group was to reduce unnecessary confusion arising from multiple systems for grading evidence and recommendations, there are some criteria that should be met when stating that the GRADE approach was used:

GRADE approach graded the quality of your finally drawn evidence from meta-analysis, it will help the readers about the strength of evidence. It helps to determine whether drawn evidence are conclusive (sufficient for clinical practice) or suggestive (need some more studies). I recommend to use GRADE of any meta-analysis. Even in the presence of heterogeneity, evidence should be graded because heterogeneity is also one of the component of grading.