We study the optimal supply chain design for a dual‐channel retailer that operates physical and web‐based stores and has traditionally managed each one separately. With increased pressures from customers and decreased profits due to marked‐down leftover inventories in each channel, the retailer considers integrating the supply chain operations of both channels and using in‐store inventory more effectively across different channels. In contrast to the existing literature, we focus on the firm's choice of the best omni‐channel strategies considering demand segmentation, cost structure, and, more importantly, the inventory rationing ability of the firm. We formulate our problem as a two‐stage stochastic programming model and use first‐order optimality conditions to study the optimal inventory ordering decisions. Based on different omni‐channel strategy decisions, we explore four supply chain design options. We identify the optimal inventory ordering policy under each option and explicitly describe the optimality conditions on cost and demand under perfect and imperfect demand information. When the demand information is imperfect, the demand fulfillment decisions can be characterized as nested protection‐level policies. First, we show that omni‐channel strategies are not necessarily profitable under all settings as dictated by market conditions. Second, we demonstrate that when the company can allocate its inventory perfectly, running omni‐channel strategies may help better serve web‐based customers and in‐store customers. Finally, we show that, in the worst case, the imperfect demand information may result in losing all the claimed inventory integration benefits when running omni‐channel strategies.