There is always a gap between theory and reality.
A theoretically correct idea is not necessarily a right answer in reality. You may observe the same issue in the data analysis.
When you drill down or breakdown a set of data by some axis (age, area, product etc.), you may come up with lots of options of axis. Even if it is theoretically possible, I would suggest that you should make sure that the axis you select will lead to some effective action(s).
For example, if you try to categorize the sales results by the weather (sunny, rainy or cloudy), you may able to technically do it. But it would not help you determine the actions you take for sales improvement because you cannot control the weather anyway.
If you use some axis by which you drill down some data, you may consider whether the axis is actionable or not. This is extremely important from the practical viewpoints.
If you select “age” axis for your analysis, then you may reach some actionable ideas on how to improve the sales from a certain category of age( eg. 20’s).
In business, “actionable” is always a priority.