The rapid growth of e-grocery has intensified the need for efficient and resilient fulfillment systems, with order picking remaining the most labor-intensive and costly process. This study investigates the potential of robotic pre-picking in an AutoStore-based fulfillment center, using a case study at Albert Heijn. A framework combining system analysis, key performance indicators, and scenario-based modeling is applied to assess how piece-picking robots can complement human operators. Tests with two robotic systems (Swisslog ItemPiQ and Sereact) showed SKU pickability rates of 74% and 90%, respectively. Using transformed operational data from 47,543 orders, hybrid scenarios were evaluated against a manual baseline. The results indicate that full robotic pre-picking can reduce order cycle times (OCT) by over 50% but requires substantial robot investment. Selective allocation strategies, particularly efficiency-ratio–based pre-picking, achieve similar time savings with fewer robots, enabling the closure of 2–3 pick stations per shift, saving up to 273 labor hours weekly. The findings show that hybrid human–robot configurations can reduce manual workload, lower bin presentations, and improve efficiency, offering both practical and scientific value. Future research should validate results in real-world settings and assess product pickability and cost implications. The proposed targeted hybrid approach offers a practical pathway toward more automated, labor-efficient, and future-proof e-grocery fulfillment operations.