During cementless total knee arthroplasty (TKA), an overlap between the resected tibia and the implant's geometry, termed interference fit, is introduced to facilitate primary stability and direct bone-implant contact. However, little is known about the actual interference achieved and the resulting mechanical response in the surrounding cancellous bone. The aim of this study was (1) to experimentally quantify the actual interference achieved for a commercially available cementless tibial implant and (2) to assess its effect on the post-impaction cancellous bone strain. Seven human cadaveric tibiae were micro-CT scanned intact (23 µm/pixel), once prepared for TKA (46 µm/pixel) and following implantation (46 µm/pixel). The actual interference across the entire bone-implant interface was quantified and, via digital volume correlation, the compressive strains of bone in contact with the implant and at increasing distance, were extracted. An inhomogeneous actual interference was found across the implant pegs and keel (median ± std dev:​ 0.70 ± 0.27 mm), which was lower than the intended. Limited interference (0.02 ± 0.12 mm) was found directly under the baseplate, with immediate bone-baseplate contact of 54%. The induced compressive strains were related to the actual interference within 3.14 mm from the bone-implant interface (R² = 0.269–0.450, p < 0.001), with higher compressive strains corresponding to higher interference, but not being related to the bone volume fraction. Clinical Significance:​ Insight is provided into the interaction between the variability of the resection and the resulting mechanical environment. A complex relationship is apparent, whereby the actual interference accounted for up to 45% of the variation in induced compressive strain magnitude.