In the 2016 Mw 7.8 Kaikōura earthquake, widespread liquefaction occurred in the reclamations at CentrePort, Wellington (New Zealand), while less intense shaking during two earthquakes in 2013 produced only some liquefaction manifestation over isolated areas of the port. The two types of reclamations, comprised of end-dumped gravelly fills and hydraulically placed silty-sandy fills, are not well represented in current empirical simplified liquefaction procedures. Comprehensive CPT investigations are presented and used to characterize the fills, and then perform CPT-based liquefaction evaluation for the three earthquakes that affected the port in 2013 and 2016. The results of the analyses are compared to observations of liquefaction performance and are examined in relation to the soil and deposit characteristics of the fills. The CPT-based liquefaction assessment was generally consistent with observations, though clear shortcomings and limitations of these procedures were also evident. The paper scrutinizes conventional engineering procedures for liquefaction assessment when applied to non-conventional or problematic soils for such assessment, with reference to a well-documented case history.