Dark Store

A retailer converts a small shop into a local fulfilment site where employees arrange shelves for quick picking, prepare online orders and hand packed bags to couriers. Customers no longer browse the aisles, though some may collect at a designated counter.

Shopify describes dark stores as order-fulfilment locations without normal in-store shopping, and McKinsey describes a standalone dark-store model among several grocery fulfilment options, each with different costs and demand requirements. Choose the market and define the service area, because a local site needs enough nearby online orders to support rent, stock and staff, and fast promises are costly in a sparse area.

Distance and traffic affect courier time, so a map radius is a starting point, not a guaranteed arrival window. Decide the assortment too: limited space favours products customers buy often, but too narrow a range may lead to small baskets and weak economics.

Keep inventory accurate so that online checkout does not sell a product the picker cannot find, updating stock for damage, shrinkage and substitutions. Design the layout for pickers, with popular items, packing stations and temperature zones that support efficient movement rather than display aesthetics, and use safe storage because food and other regulated goods may need temperature control, hygiene and traceability under local rules.

Plan receiving so that deliveries from suppliers or a central warehouse fit available space and avoid blocking picking during busy hours. Measure picking through orders per labour hour or items per hour, but different basket sizes make raw comparisons misleading.

Control packing with correct labels, fragile-goods protection and cold-chain methods to prevent refunds, and check dispatch, since a packed order waiting too long for a courier may arrive late or damaged. A dark store can offer a small pickup zone without opening its warehouse area to the public, provided location and hours are explained.

Compare the model with alternatives: an existing supermarket can fulfil some online orders without a dedicated site, but shoppers and pickers may compete for the same aisles, while a large central warehouse may lower some storage costs but increase last-mile travel. Match the model to demand density.

Calculate unit economics including product margin, labour, premises, technology, packaging, promotions, refunds and last-mile fees, and watch fixed cost, because a dedicated site keeps costing money when order volume dips, so test the break-even scenario before opening another. Avoid speed promises unsupported by capacity, since delivering in minutes requires sufficient nearby stock, pickers and couriers, especially at peak demand, and track fill rate, order accuracy and delivery time to show whether the site meets the promise.

Segment stockouts, because a missing item can cause a substitution, partial refund or cancellation, each with a different cost and trust effect, and review food waste, as local grocery inventory can spoil if demand forecasts are wrong. Consider neighbourhood impacts such as loading, courier parking and operating hours, connect systems so orders, location-level stock and delivery status share reliable data, pilot before scaling, and remember that a dark store succeeds only when the specific neighbourhood and service promise support its costs.