Spiders employ a diverse range of predator traits including potent venoms, complex silk hunting strategies and mechanical strength coupled with larger body sizes to capture prey. This trait diversity, along with the quantifiable nature of venom potency, makes spiders an excellent group to study evolutionary trade-offs. Yet, comparative approaches have been historically confounded by the use of atypical prey models to measure venom potency. Here, we account for such confounding issues by incorporating the phylogenetic similarity between a spider's diet and the species used to measure its venom potency. Using a phylogenetic comparative analysis of 75 spider species to test how diet, silk use in prey capture and body size drive venom yield and potency (LD50), we show that spider venoms are generally more potent against prey models more closely related to their natural prey, reflecting prey-specific patterns. Despite predictions, we find no trade-offs between body size, silk use in prey capture and venom potency. We also find that venom yield scales sublinearly with size, reflecting the 0.75 allometric scaling predicted by metabolic theory, suggesting venom is metabolically expensive in spiders. Our approach demonstrates how contemporary comparative approaches can be applied to historic venom potency measures to test fundamental evolutionary patterns in predator traits.