When does real-time data become a scaling bottleneck

hey folks, ive been thinking a lot about real-time data pipelines lately, esp with services that need to push updates instantly. weve seen massive user growth, and while our current setup handles it, im starting to wonder at what point does the sheer volume and velocity of real-time data itself become the main scaling challenge? are there specific architectural patterns or tech youve found are crucial for handling this kind of load before it breaks things?