My understanding is that the determinant of a matrix can be thought of as the volume of a parallelepiped formed by the column vectors and copies of them linked to themselves.
From this I can conclude making the vectors longer, or more orthogonal should increase the determinant. So if were to divide each column separately by its length (to get rid of the effect from longer vectors) and take the determinant of the resulting matrix, it should give me a sort of overall orthogonality score — the closer to $1$ the more orthogonal all the column vectors are to one another.
Is there a name for this technique/number? I could see it being useful in statistics where each column is a feature and rows are observations, but when I try to Google it I just get results about orthogonal matrices, the end case where the determinant is $1$, rather than discussion of this sort of score.