Most warehouse layouts are decided by the racking supplier, because they are the first party to draw anything. That drawing maximises pallet positions, which is a reasonable thing for a racking supplier to optimise and a poor thing for a warehouse to be built around.
Layout is a sequence of decisions, and they have an order. Get the order wrong and you spend years working around a building that fights you. This is the sequence, with the trade-offs stated plainly.
Start with flow, not with shelving
Before any rack is placed, decide how goods travel from the inbound door to the outbound door. Everything else follows from that line, because every metre of it will be walked or driven thousands of times a year.
Three shapes cover almost every Dutch warehouse:
I-flow (straight through)
Receiving at one end, dispatch at the other, goods travelling in one direction. Simple, no crossing traffic, and it scales cleanly. It needs doors on opposite walls and a deeper building than most units offer, and it doubles your dock infrastructure because inbound and outbound cannot share.
U-flow (in and out on the same side)
Goods enter and leave on the same wall, travelling a loop. This is the most common shape in Dutch distribution, and for good reason: one dock area, shared staff, shared equipment, and the fast-moving goods can sit near both doors at once. The cost is congestion — inbound and outbound traffic use the same space, and a busy morning receipt can block the afternoon dispatch.
L-flow
A compromise for buildings with doors on adjacent walls. It works, it is rarely a first choice, and it usually appears because the building came that way.
Zone by how often things move
Once the flow line exists, place goods against it by movement frequency rather than by product family. Grouping by supplier or by category is how most warehouses start, because it matches the way people think about the products, and it is almost always the wrong axis.
A small share of items generates most of the picking. Those belong closest to dispatch along the flow line, in the most accessible positions. Slow items go far away and high up, and it does not matter much that they are inconvenient, because they are rarely touched.
The mistake worth naming: fast-moving goods placed together in one dense aisle. It looks tidy and creates a traffic jam, because most of your pickers are in that aisle most of the time. Spread the fast movers across several aisles near dispatch instead of concentrating them.
The aisle width trade-off nobody states out loud
This is the decision with the longest consequences, because it is the one you cannot change later without emptying the building.
Narrow aisles fit more pallet positions into the same floor. Wide aisles let a counterbalance forklift turn, let two people pass, and let one person work without blocking another. Racking suppliers quote positions per square metre, so their proposal trends narrow.
What the position count does not show is that very narrow aisles require specific truck types, which are more expensive, need trained operators, and — the part that hurts — mean one broken truck can strand an entire zone because nothing else fits down there. Decide the truck type and the aisle width together, and treat them as one decision rather than two.
A practical check: walk the drawing and ask what happens when a picker and a pallet truck meet in the middle of the longest aisle. If the answer is that one of them reverses the full length, the aisles are too narrow or too long.
Build the pick face around the human
For hand-picked goods, the height a picker reaches decides both speed and injuries. Work between roughly knee and shoulder height is fast and safe. Below the knee and above the shoulder is slower and, repeated hundreds of times a day, is where back injuries come from.
That comfortable band is scarce. Give it to the items picked most often, and put slow-moving stock in the positions that require bending or a step. This single principle usually produces more measurable improvement than any amount of route optimisation, because it changes the cost of every individual pick rather than the sequence of picks.
Where an item is picked in cases rather than units, the same logic applies to weight: heavy items low and close, light items higher.
Leave room at the doors
Staging space next to the docks is the most commonly under-sized area in a new layout, because it looks like empty floor on a drawing and empty floor looks like waste.
It is not waste. It is the buffer that absorbs a truck arriving early, a load that must be checked before putaway, and an outbound order built two hours before the vehicle appears. Without it, that work happens in the aisles, which is exactly where it costs the most.
A useful sizing question: what is the largest single shipment you receive or dispatch, and can it sit on the floor without blocking a route? If not, the staging area is too small no matter what the drawing says about efficiency.
Address every location before you store anything
Give every location a code that a person can read from a distance and a scanner can read up close, and do it before the first pallet lands. Retro-fitting an addressing scheme onto a working warehouse is a weekend nobody enjoys.
What makes a scheme work in practice:
Make it read in the order someone walks: aisle, then bay, then level. A picker holding a list should be able to tell from the code alone whether the next location is ahead or behind them.
Number aisles so that odd and even sit on opposite sides, which lets a picker work down one side and back the other without crossing.
Leave gaps in the numbering. If you number bays 1 to 40 with no space, the day you add racking you get bay 12A, and every list involving it becomes ambiguous.
Label at eye level and again high up. High labels are what make a reach truck operator able to confirm a level from the seat.
Measure before you draw
Four numbers determine most of the layout, and all four are knowable before anyone quotes racking:
How many order lines you pick per day, and what share comes from how few items — this sets how much prime space you need near dispatch.
Your largest and heaviest handling unit — this sets aisle width and rack specification more than average dimensions do.
Peak inbound and outbound volume on the busiest day, not the average day — this sets dock and staging size.
Whether you must separate anything by rule: batch-tracked goods, temperature ranges, hazardous materials. These constraints shrink the usable space and are cheaper to design around than to retro-fit.
Where software fits
A good layout makes work shorter. Software decides which work happens in which order, and it can only choose among the paths the building allows. Both matter, but the sequence is one-directional: a warehouse management system placed on top of a badly zoned building will run an efficient version of unnecessary walking.
running a WMS selection covers what to score, what to demand in a demo, and when scanning on your existing system is the better spend.
Frequently asked questions
Where do you start when laying out a warehouse?
With the flow line from inbound door to outbound door, before any racking is placed.
- Every metre of that line is walked or driven thousands of times a year, so it sets the cost of everything else.
- U-flow is the most common shape in Dutch distribution: one dock area, shared staff, at the cost of congestion.
- Racking suppliers draw first and optimise pallet positions, which is not the same as optimising your operation.
How wide should warehouse aisles be?
It is not a width decision on its own — decide the truck type and the aisle width together.
- Very narrow aisles need specific trucks that cost more and need trained operators.
- One broken truck can strand an entire very-narrow zone, because nothing else fits down it.
- Test the drawing: if a picker and a pallet truck meeting mid-aisle means one reverses the full length, it is too narrow or too long.
Should goods be grouped by product family or by movement?
By movement frequency along the flow line. Grouping by supplier or category matches how people think about products and is almost always the wrong axis.
- Fast-moving items belong closest to dispatch, in the most accessible positions.
- Do not concentrate all fast movers in one aisle — it creates a traffic jam. Spread them across several aisles near dispatch.
- Slow items can be far away and high up; inconvenience costs little when nobody goes there.
What should we measure before designing the layout?
Four numbers, all knowable before anyone quotes racking.
- Order lines picked per day, and what share comes from how few items — this sets how much prime space you need near dispatch.
- Your largest and heaviest handling unit, which drives aisle width and rack specification more than averages do.
- Peak inbound and outbound volume on the busiest day, not the average — this sizes docks and staging.
- Any mandatory separations: batch tracking, temperature ranges, hazardous materials.