Ensuring the recruitment of desired tree species is a primary goal when considering successful regeneration of timber harvests. Deer populations that are overabundant relative to forage availability within their local landscapes pose a considerable challenge to successful forest regeneration. We evaluated the short-term (<10 years) impact of white-tailed deer browsing on woody structure and composition in northcentral Pennsylvania. Specifically, we compared woody plant species density, height, and cover within 10 pairs of fenced and unfenced regenerating timber harvests. Percent of browsed stems was higher in unfenced harvests for red maple (Acer rubrum), oaks (Quercus spp.), Rubus spp., sassafras (Sassafras albidum), and witch hazel (Hamamelis virginiana). Stem density did not differ between fenced and unfenced harvests for any species except Rubus spp., which was greater in the former. However, height of serviceberry (Amalanchier spp.), sassafras, northern red oak (Quercus rubra), Rubus spp., and red maple was taller in fenced harvests. Both vertical and horizontal vegetation density were greater in fenced harvests, suggesting a more structurally complex understory in fenced harvests. Rubus spp. and tall (>1.5 m) sapling cover were greater in fenced harvests, while unfenced harvests had greater forb, fern, and blueberry/huckleberry (Vaccinium spp./Gaylussacia spp.) cover. Our results suggest that browsing by deer in the years initially (<10 years) following timber harvest may not immediately impact woody species density, but can considerably reduce seedling height, overall horizontal and vertical structure, and composition of vegetation strata. Reduced sapling height and greater prevalence of blueberry/huckleberry, forbs, and ferns in unfenced harvests threaten the successful recruitment of desired species into the future overstory. Our results from large, operational-scale timber harvests combined with those from previous small-scale studies stress the potential value of deer exclusion fencing to achieve successful forest regeneration. Future research should examine the timing of fence removal to achieve forest regeneration success, while also providing deer access to available foraging opportunities during the later portion of stand initiation.