Let $a,b,c,d \in \mathbb{R}^3$ be the vertices of a tetrahedron (I’m unsure whether or not the order of the vertices is important for what follows). The volume of the tetrahedron is $$ \begin{align} \operatorname{vol}(a,b,c,d) &= \frac{1}{6}\big((b-a) \times (c - a)\big) \cdot (d - a) \\ &= \frac{1}{6}\det \left|\begin{matrix} (b-a) & (c-a) & (d-a) \end{matrix}\right| \end{align} $$ The gradient can be represented nicely by $$ \nabla\operatorname{vol}(a,b,c,d) = \frac{1}{6}\left(\begin{matrix} (d - b) \times (c - b) \\ (c - a) \times (d - a) \\ (d - a) \times (b - a) \\ (b - a) \times (c - a) \\ \end{matrix}\right). $$ I have verified this formula by hand, i.e. I have manually calculated $\nabla\operatorname{vol}$ componentwise and compared the result to the componentwise expansion of the given formula — but I’m not satisfied. I prefer derivation to “left-hand side equals right-hand side” verification.
Can anyone give an insightful explanation of the gradient formula?