Distributed search has become a very important research area in light of the proliferation of peer-to-peer networks and grids, which heavily rely on the effectiveness of forwarding algorithms to propagate queries among peers. Most present search techniques apply homogeneous forwarding logic in each peer. None of these forwarding algorithms is perfect. This article introduces heterogeneous search, a promising approach for improving search performance through the use of multiple forwarding algorithms among peers in unstructured P2P networks. Simulation results demonstrate that this technique shows significant advantages over others.