In communication network design, connections with few edges reduce the probability of delays and increase confidence. The 2‐path network design problem (2‐PNDP) raised in this context and aimed to build up a minimum‐cost network to satisfy a set of pairs of demands using a path with at most two edges. In this work, we propose two strategies that incorporate integer linear programming (ILP) into a state‐of‐the‐art heuristic for solving the 2‐PNDP. Our approaches collect information from specific components of the original heuristic and use it to reduce the solution space explored by the ILP solver, accelerating its resolution and improving the solution quality. Computational experiments conducted on a set of instances from the literature have shown the benefit of these combinations, achieving better solutions in almost all cases.