maximum principle
Here is a fact that feels obvious once stated yet is enormously powerful: in a region with no internal heat source, the hottest spot is always on the boundary, never strictly inside. Heat does not spontaneously concentrate; a steady temperature field cannot have a peak in the interior that exceeds everything around it. The maximum principle is the rigorous statement of this intuition for the Laplace and heat equations, and it is one of the sharpest tools in all of PDE theory.
For Laplace's equation (harmonic functions), the principle says a non-constant solution on a bounded region attains its maximum and its minimum only on the boundary — the interior values are squeezed between the boundary extremes. The reason is the mean-value property: a harmonic function's value at any point equals the average over a surrounding sphere, and an average can never exceed the largest value being averaged, so a strict interior peak is impossible. For the heat equation there is a parabolic version: the maximum of the temperature over a space-time region occurs either at the initial time or on the spatial boundary — never at an interior point at a later time. There is a weak form (the max is attained on the boundary) and a strong form (if the max is attained inside, the solution is constant).
The maximum principle is the engine behind several deep results obtained almost for free. Uniqueness of the Dirichlet problem follows immediately: if two solutions share the same boundary data, their difference is harmonic with zero boundary values, so by the principle it is zero everywhere — the solution is unique. Continuous dependence (stability) and a priori bounds follow the same way, which is much of what makes elliptic and parabolic problems well-posed. It also gives qualitative physical sense for free: a body with no internal source never spontaneously develops a hotter interior than its surface, exactly as thermodynamics insists.
To prove the Dirichlet problem has a unique solution, suppose u and v both solve nabla^2 (.) = 0 with the same boundary data. Their difference w = u - v is harmonic with w = 0 on the whole boundary, so by the maximum and minimum principles 0 <= w <= 0 inside; hence w = 0 and u = v.
A two-line uniqueness proof that would be hard to get any other way — the maximum principle does the heavy lifting.
The maximum principle holds for elliptic (Laplace) and parabolic (heat) equations but not for the wave equation: a hyperbolic solution can oscillate and exceed its boundary and initial values in the interior, so do not expect this tool to constrain waves.