The answer is by using the full potential of the three mandatory parameters that each session has in Google Analytics. I emphasize all sessions here, because even sessions that are not part of a campaign have a utm_campaign parameter : the value is then assigned to ( not set ).
Let’s review these three parameters:
1 – utm_medium : it is not recommended to customize this parameter given its crucial importance in the definition of channel grouping (unless you customize your channel grouping .)
2 – utm_source : customization of this parameter is possible but not recommended given the importance of the name of the site / network (Facebook, Google, Adwords) in the analysis of a campaign. Not to mention that this parameter is already formatted in a particular way in the default script of other source utms in your Google Analytics system. Ideally, as with all other sources of your analytics , it is best to limit yourself to the site your user was on before going to yours. An example: you want to compare the performance of your sponsored posts on Facebook with that of organic posts. If you keep an utm of type: utm_source=facebook , utm_medium=cpc and utm_campaign=mysupercampaign , you can then easily do this analysis by selecting filter the source in your report. If, on the other hand, the value of the utm_source is personalized ( facebook_campagneXcreatifY ), your analysis may be more complicated because the number of variations of your source parameter for your traffic identified by a utm may be significant.
3 – utm_campaign : it is strongly recommended to personalize it, it exists and was created for this purpose. Many combinations for the same campaign don't look pretty in your report by campaign because the results for one campaign are split across multiple rows. But on the other hand, thanks to a rigorous nomenclature, you can easily filter your results both in Google Analytics and in a visualization platform like Google Data Studio, Datorama or Microsoft Power BI, without any sampling .