Summary1. Confirming the impacts of invasive plants is essential for prioritizing management efforts, but is challenging, especially if impacts occur below-ground and over long periods as hypothesized for woodland shrub invasions. 2. For this reason, we developed a novel 'paired-point' framework capable of detecting the below-ground impacts of slow-growing invasive plants from short-term data sets in an investigation into the below-ground consequences of European buckthorn Rhamnus cathartica L. (hereafter buckthorn) invasions into Chicago-area woodlands. We measured differences between the members of 97 pairs of buckthorn-invaded and buckthorn-free points in 10 below-ground response variables (RVs) that buckthorn is hypothesized to alter (listed below). We then modelled these differences (DRVs) in response to the severity of the invasion found at each invaded point. A relationship (linear or nonlinear) between a DRV and severity, that is slope 6 ¼ 0, suggests a buckthorn-induced change. An intercept (value of DRV where severity = 0) different from zero suggests a pre-existing difference. 3. Relating differences to an invasion gradient rather than simply noting their presence provided evidence that the higher levels of moisture, pH, total C, total N, NH þ 4 À N and Ca 2+ in invaded soils pre-date and possibly promote invasion (particularly Ca 2+ ) and that neither earthworm biomass nor soil C : N ratios are associated with buckthorn invasion, all of which suggest buckthorn to be less of a 'driver' of below-ground change than hypothesized. We did, however, find evidence that buckthorn establishes in areas having greater leaf litter mass and higher rates of decomposition, and then proceeds to accelerate decomposition further and to alter spring soil NO À 3 À N levels. 4. Synthesis and applications. Our findings suggest that decisions regarding regional buckthorn management should be less driven by concerns about buckthorn's below-ground impacts and that greater consideration of how variation in below-ground factors relates to invader establishment is needed, rather than simply assuming this variation to be invaderinduced. This latter consideration can help to design better targeted monitoring programmes, limiting the further spread of woodland invaders. These insights illustrate the utility of the paired-point framework both for investigating below-ground causes and consequences of slow-growing invasive plants and for guiding the management of these invasions.