Start-ups have a high need for resources yet face significant risks when forming partnerships with incumbents to access those resources. We propose that a partnership strategy based on relational pluralism, forming multiplex and multifaceted ties with partners, can mitigate these risks. Such ties offer the start-up increased legitimacy and a relational safeguard against resource misappropriation by more powerful partners. However, we propose that there is a limit to the effectiveness of relational pluralism. Its effect is weakened when the start-up becomes entirely dependent on a small set of partners, or when an additional tie yields resources that are redundant. We argue that the start-up only benefits when the gains from relational safeguarding and legitimacy outweigh the costs of dependence and redundancy. We empirically observe the co-evolution of start-ups’ interlocking directorate and strategic alliance networks in the Australian mining industry over a 10-year period. Our results show that start-ups that engage in relational pluralism perform better than both start-ups that form no alliances and start-ups that form stand-alone alliances. Having a very small portfolio of partners or one that skews heavily toward local partners, however, indeed limits the effectiveness of relational pluralism. Intriguingly, we also find that the temporal sequencing of relational pluralism matters. One of our central findings is that the best performing start-ups first form board interlocks with promising partners and add a strategic alliance later. This offers a rare glance at the temporal sequencing in which peripheral start-ups can gain exceptional performance through partnership formation.