Scatter search is a population-based metaheuristic that has been shown to yield high-quality outcomes for hard combinatorial optimization problems. It uses strategies for combining solution vectors, making limited use of randomization, that have proved effective in a variety of problem settings. The fundamental concepts and principles were first proposed in the 1960s and 1970s as an extension of mathematical relaxation techniques for combinatorial optimization problems. Its framework is flexible, allowing the development of implementations with varying degrees of sophistication. This chapter provides a grounding in the scatter search methodology that will allow readers to create successful applications of their own. To illustrate this, we present a scatter search implementation for a \(\mathcal {N}\mathcal {P}\) -hard variant of the classic p-hub median problem, for which we describe search elements, mechanisms, and strategies to generate, combine, and improve solutions.