Distribution hubs: siting the node your orders flow through

One central building or several regional ones

Centralised stock pools demand, so total inventory for a given service level falls, picking is concentrated and systems stay simple. The price is longer outbound distances and dependence on a single site. Regional buildings shorten the final leg and improve resilience, but each one needs its own safety stock, its own management and its own peak labour. The honest comparison includes obsolescence and handling, not just freight. Splitting stock multiplies the number of units that age in the wrong location, which for short-life or fashion-driven ranges can outweigh any transport saving.

When goods should not stop at all

Cross-docking moves inbound loads straight to outbound vehicles with no putaway and no picking. It works where volume is predictable, the inbound arrives pre-sorted or labelled for its destination, and the timing windows of the two flows overlap. Where any of those fail, the dock becomes a floor-stacked storage area with none of the controls of a racked one. Flow-through and put-to-store variants sit between the extremes, allocating stock on arrival without ever holding it. They demand tight upstream data, which is usually the real constraint rather than the building.

Returns are a second inbound stream

Reverse flow arrives unannounced, in mixed condition, in packaging that was not designed to survive a second journey. Sites planned only around forward volume end up processing returns in aisles, which slows the picking operation at exactly the point in the year when it can least absorb it. Give the reverse stream its own space, its own decision rules on disposition and its own staff. The financial argument is that unresolved returns are working capital sitting in a corner, not a customer service inconvenience.

Carrier access and cut-offs decide the geography

A building is only as good as the last vehicle that leaves it. Parcel and pallet networks work to fixed departure times feeding their own sorting operations, so a site far from a network hub effectively loses hours from its order-capture window. Two locations with identical rent can therefore offer very different order cut-offs to the same customer. Ask carriers for their collection times at each candidate site before shortlisting. That single question reorders most site comparisons more decisively than rent per square metre.

Cargo roles

• Order assembly and outbound despatch to customers
• Replenishment of downstream branches and stores
• Cross-dock transfer between inbound and delivery vehicles
• Returns receipt, inspection and re-entry to stock

Frequently asked questions

How many distribution sites should a growing business run?

As few as the service promise allows. Each additional site adds safety stock, management and fixed cost, so the case for a second building rests on a delivery commitment that a single location genuinely cannot meet, or on resilience that the business has decided to buy deliberately.

Is cross-docking always cheaper than storing?

No. It removes putaway and picking, but only when inbound volume is predictable and arrives ready for its outbound destination. With irregular arrivals or unsorted loads, the dock fills with floor-stacked pallets that need more handling than they would have done in racking.

What is the most common siting mistake?

Optimising on rent and road distance while ignoring collection times and labour supply. A cheap building that misses the parcel network's departure or cannot staff a second shift will underperform a dearer site that clears both, and neither factor appears in a straightforward freight model.