From what I've (briefly) read, typically the profiles are within about 5 points of each other. There seems to be an assumption that, like building muscles, they can be trained (at least early on) to more or less even out. If there are large chasms between profiles, but nothing dipping below the median, the relative deficit seems to be ignored since the worst-performing area is still "good enough".
Even when one area does dip below the median, it may not be diagnosable since it's only one criterion.
That's how it appears to me anyway. If I've got it wrong, hopefully someone can bring out the proverbial red pen. 