Three evaluations, three different phone lines.

Each study runs the Voice Agent rubric against a different kind of inbound line and returns a scored failure map. Every figure is illustrative — these show the depth and shape of a real evaluation, scored on its own service rubric, not named client results.