Peak Goods Forward, Slow Goods Back: The Real Math of Seasonal Slotting Rotation

Let me be blunt: seasonal slotting is a trade, not a tidy-up. You pay once to move a pallet, and in return you buy back a few steps on every order for the next few months. When that trade pays, it pays quietly, day after day. When it does not, you have just paid good money to make your own pickers walk further. I have watched too many warehouses treat a pre-peak move like a moral duty, forklifts running all weekend, only to find pick rates worse than before they started. The fault was rarely effort. Nobody had asked the one question that matters: how many orders will this pallet actually produce in the next ninety days?

Seasonal Slotting Is Not Spring Cleaning, It Is a Moving Bill

Here is a mistake I lived through. In the fall of 2019 I was helping a third-party warehouse near Ontario, California, roughly twenty thousand square meters, get ready for the holiday gift season. The two rack rows closest to dispatch had spent all summer holding garden tools. By September, Christmas gift sets were arriving and landing in high, far-back locations. Pickers were walking an extra hundred meters or so for a single gift-set order. The site manager made the obvious call: clear those two front rows over a weekend and fill them with gifts.

On paper it looked right. In practice it was expensive in ways nobody had priced. Moving more than four hundred pallets of garden tools took two weekends, overtime, and something on the order of twenty thousand dollars in labor and equipment time. Worse, a clearance wave of garden orders had not finished yet. For the first two weeks of peak, pickers were sent to the back of the building to fetch the very stock they had just moved out of the front. Average pick time did not fall. It rose by more than ten minutes per order compared with August. The lesson was plain: judging a move only by whether the incoming product is hot, without checking whether the outgoing product is truly cold, just relocates the problem from the front row to the back row.

Every move has two sides. Pushing peak stock forward is right, because every meter saved repeats on hundreds of orders. But moving stock costs forklift hours, labor, recounting, system updates, and the lost picking capacity of any location locked during the move. Moving one pallet is cheap. Moving four hundred is a different decision entirely. The question is never simply whether a product is seasonal. It is whether it is seasonal enough to justify the cost of moving it.

The counter-example comes from a pool-supply warehouse in the summer of 2022. Inflatable pools and water toys were at full peak in June, and the manager planned to push all winter heaters to the back to make room. Before the forklifts rolled, we checked three years of outbound history. Heaters did slow down in July, but distributors reliably placed autumn pre-orders in late August, and by the first week of September heater shipments were back to roughly half their June level. A full move in July would have meant a full move back in September. We moved only a third of the heaters, the bulkiest and slowest lines, and left smaller heater models in the middle rows. Peak picking stayed clean, and the September pre-orders shipped without a second round of moving. Clearing the front row completely would have felt decisive. Leaving part of it alone was cheaper.

When to Move and When to Leave It Alone: A Seasonal Slotting Decision Table

My working method is deliberately simple: forecast the next ninety days of shipments, price the move, then weigh how hard the current location is to pick from. The table below is a rule of thumb I use with mid-sized warehouses. Treat the figures as planning ranges to test against your own labor rates, not as a vendor standard.

Behind that table sits a judgment most managers get wrong: a bigger golden zone is not a better golden zone. Add another front row and pickers walk further sideways inside the zone. Replenishment traffic gets denser. I have seen front zones grow until pickers spent their day crisscrossing them, with steps per order flat or worse. A golden zone earns its value only when it holds the highest-frequency stock for the next ninety days. Put the wrong products there and a larger zone just wastes more space, faster. Before any move, rank front-zone stock by forecast shipments for the next ninety days. Anything outside the top third does not belong there, however busy it looks today.

There is another cost people forget: the data. Every location change ripples into pick paths, replenishment suggestions, and cycle-count plans in the WMS. In a warehouse still running on paper or spreadsheets the risk is worse, because a printed location list goes stale the moment a pallet moves. My rule is strict: the system change and the physical move finish in the same shift, and ten locations get spot-checked against the system before the next shift starts. If the check fails, you recount that night. You do not send the morning crew out with wrong location data and hope for the best.

How to Sequence the Moving Week Without Breaking Peak Picking

Never move during a picking peak if you can avoid it. I sequence moves in three waves, one zone at a time, and the next wave does not start until the previous one is verified. Wave one moves slow stock that is already out of stock or barely shipping, which is the lowest-risk work. Wave two moves peak stock forward, and that work goes on nights or weekends so daytime picking never stops. Wave three tidies the middle rows and closes the gaps. After each wave, the supervisor signs off on two facts: the stock is physically in its new location, and the system shows the new location correctly. Miss either one, and the next wave waits.

Timing matters as much as sequence. Move peak stock forward four to six weeks before peak, and leave at least a week for pickers to learn the new layout. Move too early and your forecast is still guesswork. Move too late and pickers are still learning locations when error rates matter most. Slow stock can move later, even after peak stock is in place, as long as front space opens at least as fast as peak stock arrives.

Then watch the result for two full weeks. If average pick time has not fallen, something is wrong: the wrong products moved, or the front-zone layout itself is the bottleneck. In that case, admit it and move individual pallets back. Slotting is not a one-shot decision, it is something you test. The steadiest warehouses I know do two major rotations a year, one for spring and summer, one for autumn and winter, with only small adjustments in between. Low frequency keeps the data clean and the locations memorable. Warehouses that reshuffle every month usually end up with location data that is always half old and half new, and pickers who have to check the system at every stop.

So here is my advice, and I will leave it as a question you can take back to your own floor: the last time you did a big rotation, did the picking time you saved actually cover what the move cost? If you cannot answer that with numbers, do not move the next pallet until you can.

By Lao Mi, Supply Chain Veteran (20 yrs)