Neointimal hyperplasia (NH) is a main source of failures in arteriovenous fistulas and vascular grafts. Several studies have demonstrated the promise of perivascular wraps to reduce NH via promotion of adventitial neovascularization and providing mechanical support. Limited clinical success thus far may be due to inappropriate material selection (e.g., nondegradable, too stiff) and geometric design (e.g., pore size and spacing, diameter). The influence of pore size and spacing on implant neovascularization is investigated here for a new biodegradable, thermoresponsive shape memory polymer (SMP) perivascular wrap. Following an initial pilot, 21 mice were each implanted with six scaffolds: four candidate SMP macroporous designs (a–d), a nonporous SMP control (e), and microporous GORETEX (f). Mice were sacrificed after 4 (N = 5), 14 (N = 8), and 28 (N = 8) days. There was a statistically significant increase in neovascularization score between all macroporous groups compared to nonporous SMP (p < .023) and microporous GORETEX (p < .007) controls at Day 28. Wider‐spaced, smaller‐sized pore designs (223 μm‐spaced, 640 μm‐diameter Design c) induced the most robust angiogenic response, with greater microvessel number (p < .0114) and area (p < .0055) than nonporous SMPs and GORETEX at Day 28. This design also produced significantly greater microvessel density than nonporous SMPs (p = 0.0028) and a smaller‐spaced, larger‐sized pore (155 μm‐spaced, 1,180 μm‐sized Design b) design (p = .0013). Strong neovascularization is expected to reduce NH, motivating further investigation of this SMP wrap with controlled pore spacing and size in more advanced arteriovenous models.