Land-use change drives biodiversity loss, but some species suffer more than others. Indicators of global biodiversity change must attempt to summarise these impacts representatively and meaningfully to guide biodiversity recovery. Yet species that are hard to detect, and thus feature less in relevant databases, might possess traits that make them particularly sensitive to anthropogenic impacts. Using global data for plant, bird, and spider species, we developed a statistical approach to analyse and correct for the impact of excluding hard-to-sample species from global biodiversity indicators. Based on over 4000 species with abundance comparisons available, we found that species with fewer global occurrence records consistently decline more as land-use intensity increases, suggesting that hard-to-sample species are particularly sensitive to land-use differences. When we extrapolate this relationship to all plant, bird and spider species with valid occurrence records (0.27 M species), we obtain a more representative global indicator of overall land-use impacts for these entire taxonomic groups. Our estimates indicate a lower average abundance in anthropogenic land uses compared to results obtained when hard-to-sample species are excluded. For example, intensive agriculture only has 18% of the biodiversity level of primary vegetation, rather than the 47% estimated without extrapolation. We recommend that other existing indicators include an extrapolation solution based on ours, to incorporate the available data as effectively as possible. Using occurrence data to predict species\' sensitivity unlocks many possibilities to improve global biodiversity indicators, without demanding additional data on poorly known species.