So the question I'm currently asking myself whether plugging in a proper pathfinding system into the game would be worth it. It would certainly simplify several aspects of the game - from the editor (where the designer had to lay down waypoints which I really didn't like), to the actual behaviour of the guards (I wouldn't have to write anything to try to mask the fact that they're all following straight lines). I just have to now decide whether I want to implement and debug a propper A* pathfinding algo - which I have only ever read about, never implemented.
All this occured to me when I was writing the level editor and realised how horrible the logic for the guards could be if I continued with this method. The algo would be a fairly basic version because of the simple level format but its not something I was planning to have to deal with for this game. I know I'm going to have to learn to do it sometime (I really want to learn it), but I have to decide whether its worth adding the task to this game.