You can't ever measure it directly. Only indirect effects.
From that we build models.
And one of those models contains a quantity called "energy [physics/chemistry/etc]" which is in fact a highly abstract concept that doesn't correspond to any actual object in the world. Technically it doesn't even have a definite value in any real system.
But we take it seriously because there's a useful, strong conservation law based on strongly compelling underlying principles (e.g. Noether's theorem), and no known violation of said law, or at least, every time there is an apparent violation, it eventually leads us to an explanation. It's a very useful abstraction indeed, and we can build all sorts of useful things knowing it.
In fact, the conservation is key to making "energy [physics/chemistry/etc]" accurately measurable at all. And sometimes we even get that wrong.
The notion of "subtle energy" is another model, much less well developed and rarely if ever quantified. One can argue all day about whether it's a useful model and what it models, if anything. My view is I'd love to see serious attempts to work backwards from observed phenomena (such as mentioned in the GP), as well as other reports (people report tingling, heat, feelings like electrical currents etc - this is data even if it is subjective, and it has structure). And from that build a more coherent model than the "intuitive" explanations often produced.
But they are both describing abstract properties in models, and measurements are indirect not direct in both cases.