Objectives To evaluate the dimensional accuracy of impressions taken by use of disposable stock plastic trays and to compare performance with that of metal trays. Materials and methods From a metallic model incorporating three precision balls and three abutment teeth, one-step dual-phase polyether (PE) and vinyl polysiloxane (VPS) impressions were taken using either metal or disposable plastic trays (n = 10 for each of the resulting four test groups). Respective plaster cast scans were aligned with the reference dataset to evaluate global (distance and angle deviations) and local (trueness and precision) accuracy. Analyses of variance (ANOVA) were conducted to determine group differences. Results For all impression tray and material combinations, global accuracy was good (mean distance changes < 100 μm) with greatest deviations being observed for distances exceeding one quadrant of the dental arch. In general, distances measured in the plaster casts were too short. Only VPS impressions with plastic trays showed a different behavior with a large percentage of cross-arch distances exceeding the reference value. Mean local accuracy ranged between 6 and 14 μm (trueness), and 6 and 16 μm (precision). On abutment tooth level, metal trays were associated with a significantly better precision (p = 0.015). Conclusions The observed distortions of the studied impression trays and materials are small and should enable satisfying clinical impression-taking. Clinical relevance Cleaning and processing of metal trays before re-use are time-consuming. Especially for patients' management with single crowns and small fixed dental prostheses, disposable plastic trays can be a viable and cost-effective alternative.