At present, the network meta-analysis has been rapidly developed and widely used, and it has the characteristic of quantifying and comparing the relative advantages of two or more different interventions for one health outcome. However, the comparison of multiple interventions has increased the complexity of drawing conclusions of network meta-analysis, and the ignorance of the certainty of evidence has also led to the misleading conclusions. Recently, the GRADE (grading of recommendations assessment, development and evaluation) working group proposed two approaches on how to make conclusions from a network meta-analysis of interventions, namely, the partially contextualised framework and the minimally contextualised framework. When using partially contextualised framework, authors should establish ranges of magnitudes of effect that represent a trivial to no effect, small but important effect, moderate effect, and large effect. The guiding principles of this framework are that interventions should be grouped in categories based on the magnitude of the effect and its benefit or harm; and that when classifying, consider the point estimates, the rankings, and the certainty of the evidence comprehensively to draw conclusions. This article took an example to describe and explain the principles and four steps of partially contextualised framework, in order to provide guidance for the application of this GRADE approach in the interpretation of results and conclusions drawing from a network meta-analysis.