What is the difference between Backtracking and Exhaustive Search techniques?

Backtracking is an algorithmic paradigm aimed at improving the time complexity of the exhaustive search technique if possible. Backtracking does not generate all possible solutions first and checks later. It tries to generate a solution and as soon as even one constraint fails, the solution is rejected and the next solution is tried.

Instead of exhaustive search the backtracking algorithm tries to construct a solution incrementally, one small piece at a time. It’s a systematic way of trying out different sequences of decisions until we find one that works.