Abstract During the last decade completion axillary lymph node dissection (cALND) was gradually omitted in sentinel lymph node positive (SLN+) breast cancer patients. However, adoption varies among hospitals. We analyzed factors associated with the omission of cALND in all Dutch SLN+ patients. As one of the focus hospital‐related factors we defined “innovative” as the percentage of gene‐expression profile (GEP) deployment within the indicated group of patients per hospital as a proxy for early adoption of innovations. cT1‐2N0M0 SLN+ patients treated between 2011 and 2018 were selected from the Netherlands Cancer Registry. Hospitals were defined to be innovative based on their GEP use. Multivariable logistic regression (MLR) was performed to assess the relationship between innovative capacity, patient‐, treatment‐ and hospital‐related characteristics and cALND performance. 14 317 patients were included. Treatment in a hospital with high innovative capacity was associated with a lower probability of receiving cALND (OR 0.69, OR 0.46 and OR 0.35 in modestly, fairly and very innovative, respectively). Other factors associated with a lower probability of receiving a cALND were age 70 and 79 years and ≥79 years (ORs 0.59 [95% CI: 0.50‐0.68] and 0.21 [95% CI: 0.17‐0.26]) and treatment in an academic hospital (OR 0.41 [95% CI: 0.33‐0.51]). Factors associated with an increased probability of undergoing cALND were HR−/HER2− tumors (OR 1.46 [95% CI: 1.19‐1.80]), macrometastatic lymph node involvement (OR 6.37 [95% CI: 5.70‐7.13]) and mastectomy (OR 4.57 [95% CI: 4.09‐5.10]). Patients treated in a hospital that early adopted innovations were less likely to receive cALND. Our findings endorse the need for studies on barriers and facilitators of implementing innovations.