For decades, scholars have put forward the idea that change in industrial networks depends on the relationships and networks themselves. However, models are still lacking that conceptualize the heterogeneity of relationships and networks and that show how this heterogeneity actually affects change. This paper puts for-ward a model suggesting that network structure-in terms of an open or closed system-and relational embeddedness together affect the ways knowledge is gained, given knowledge flow and problem solving as two sources of knowledge. Moreover, the paper proposes that this effect influences the tacitness and novelty of the knowledge gained. Additionally, network structure and gained knowledge are postulated to have an impact on two changes, the establishment of relationships and the development of technology, which take place in industrial networks. The paper advances six propositions and concludes with implications for research and practice.