I think that is where parsing them out on an axial graph helps. A brand can be a pure brand - a classical example is the many cartoon/cutesy characters that are

I think that is where parsing them out on an axial graph helps. A brand can be a pure brand - a classical example is the many cartoon/cutesy characters that are used in Japan to represent everything from supermarkets to local government. A brand can also be polluted with axis-3 psychological manipulation[1], implied claims that aren't true, false differentiation, and so forth. You can separate at least in abstract as a thought experiment, how much it is of each. [1] I'm excepting cuteness from manipulation because although it does make people happy, it's not tricksy, you can see what's making you happy, it's the cute logo right there in front of you.