This study investigates the performance of base-isolated buildings designed according to the recommendations provided in the NZSEE/MBIE base isolation design guidelines. In total 16 case study buildings were designed for a site in Wellington, New Zealand, including four fixed-base buildings (for comparison) and 12 isolated buildings with various inelastic-spectrum-scaling factors, kμ (equivalent to force-reduction factors) and importance levels. The performance of each building design was subsequently assessed using the FEMA P-58 framework. Three-dimensional numerical models were developed in OpenSees to perform the non-linear time history analysis with 180 pairs of ground motions across nine intensity levels. Results suggest that the average annual rate of collapse and the expected annual loss of the isolated buildings are both around four times lower than fixed-base buildings. This study also investigated the impacts of superstructure design ductility (controlled via the inelastic-spectrum-scaling factor, kμ) and the design importance level. Results showed that a high kμ is likely to worsen the performance of the base-isolated building. For a kμ = 2, the peak storey drift demands were increased by 50% ~ 100%, whereas the peak floor acceleration demands were only slightly reduced. As a result, the expected annual loss increased. Observations showed an increase in kμ reduced the median value of the superstructure collapse fragility and could change the failure mechanism from isolator failure to superstructure failure. To improve performance, one could allow for more isolator displacement capacity at kμ = 1 or impose suitable superstructure deformation limits if a higher kμ is permitted. Lastly, the results showed that designing a base-isolated building with a higher importance level increased the peak floor acceleration demands by 50% to 60% and had a mixed impact on the peak storey drift demands. However, it did reduce the annual rate of collapse rate by around a factor of two.