Examining ecosystem functioning of logged-over forests requires a quantitative trait-based monitoring approach, in addition to a taxonomic approach, which allows for evaluation of the accompanying shifts in species traits and functional composition. Our study evaluated forest regrowth after different logging and thinning techniques applied to 60 one-hectare forest subplots in East Kalimantan that were logged using selective conventional and reduced-impact techniques. We used seven functional traits to compare the short-term (∼1-year) and longer-term (∼10-year) changes in functional diversity and composition of the forest. Both conventional and reduced impact logging resulted in higher functional diversity after 10 years, but forest functional traits shifted away from those observed in unlogged (control) subplots. Reduced-impact logging led to smaller deviations in functional composition compared to unlogged forest. We found that low-intensity reduced-impact logging, i.e., <3 m2 ha−1 (equivalent to 5 trees ha−1 or approximately 42 m3 ha−1), maintained functional diversity equivalent to unlogged forest measured in 12 one-hectare control plots. Selectively logged forest subplots with liberation thinnings increased functional diversity while leaving the trait composition unchanged. Thinned subplots exhibited higher functional diversity than the un-thinned subplots and greater similarity to unlogged forest. However, high-intensity thinning, i.e., >10 m2 ha−1, resulted in large functional traits shifts away from unlogged forest. Our study suggests that preserving the original forest functionality, i.e., by applying low intensity, reduced-impact logging and thinning operations, represents a sustainable strategy for long-term tropical forest management.