Trawling and dredge fisheries remove vulnerable fauna, homogenise sediments and assemblages, and break down biogenic habitats, but the full extent of these effects can be difficult to quantify in the absence of adequate control sites. Our study utilised rare control sites containing biogenic habitat, the Separation Point exclusion zone, formally protected for 28 years, as the basis for assessing the degree of change experienced by adjacent areas subject to benthic fishing. Sidescan sonar surveys verified that intensive trawling and dredging occurred in areas adjacent to, but not inside, the exclusion area. We compared sediment composition, biogenic cover, macrofaunal assemblages, biomass, and productivity of the benthos, inside and outside the exclusion zone. Disturbed sites were dominated by fine mud, with little or no shell-gravel, reduced number of species, and loss of large bodied animals, with concomitant reductions in biomass and productivity. At protected sites, large, rarer molluscs were more abundant and contributed the most to size-based estimates of productivity and biomass. Functional changes in fished assemblages were consistent with previously reported relative increases in scavengers, predators and deposit feeders at the expense of filter feeders and a grazer. We propose that the colonisation of biogenic species in protected sites was contingent on the presence of shell–gravel atop these soft sediments. The process of sediment homogenisation by bottom fishing and elimination of shell-gravels from surficial sediments appeared to have occurred over decades – a ‘shifting baseline’. Therefore, benchmarking historical sediment structure at control site like the Separation Point exclusion zone is necessary to determine the full extent of physical habitat change wrought by contact gears on sheltered soft sediment habitats to better underpin appropriate conservation, restoration or fisheries management goals.