When was the last time anyone reviewed your WMS putaway rules? Take a moment. Most warehouses configure them during implementation — a couple of checkboxes, "default to random putaway, prioritize empty locations" — and never touch them again.
I spent an afternoon in a 3PL warehouse in Fontana back in 2023, roughly 80,000 square feet, under 4,000 SKUs. The picking supervisor kept telling me they were short-staffed. I shadowed the floor for a few hours and the problem wasn't headcount at all: fast movers had been directed to remote upper-level slots, while slow-moving dead stock sat in prime low-level bays right next to packing. Pickers were hiking laps around the building, paying in footsteps for a rule set nobody had revisited in years.
This is the unglamorous truth about WMS putaway configuration: it looks like a one-time setup task, but it quietly shapes every pick path in your building. Get the product flow wrong and no amount of automation will bail you out.
Know What You're Actually Choosing Between
Every WMS vendor brands its putaway strategies differently, but underneath the marketing there are really just a handful of approaches. Here they are, simplest to most complex, with honest notes on who each one suits.
Fixed slotting. One SKU always lives in the same dedicated locations. Pickers learn it by heart, training is cheap, and replenishment is predictable. The downside is space utilization — slots sit empty in slow seasons, and a viral SKU overflows its fixed homes. This works for brand warehouses with stable assortments and reliable fast movers. My take: reserve it for your A-class SKUs, and revisit the assignment regularly. A slotting map built on 2021 velocity data is a liability in 2026.
Random putaway. The system fills whatever empty slot is available. It's the beloved default for new implementations because configuration takes ten minutes. The price is fragmentation — one fast mover split across four corners of the building. It fits warehouses with huge SKU counts and relatively flat velocity curves. I once saw an auto-parts operation in Moreno Valley where a single SKU had been scattered across 17 locations; picking one order meant running three aisles. They fixed it with constrained random putaway — same-SKU locations limited to two adjacent aisles — and pick paths shrank immediately.
ABC-directed (velocity-based) putaway. SKUs are tiered by outbound frequency, and A-class product gets the slots closest to packing at the most ergonomic heights. This is the most logical approach on paper, and the most dangerous one to neglect — because velocity is seasonal. Holiday decor is A-class in November and C-class in January. If your WMS doesn't recompute ABC tiers on a cadence, you're navigating with an expired map.
Replenishment-driven putaway. Inbound goes to a receiving or bulk area, and pick faces get topped up by rule. This is the playbook for high-volume e-commerce fulfillment, where the forward pick area holds roughly a week's worth of stock. Here the putaway logic hides inside the replenishment rules: trigger levels, refill quantities, source zones. Set thresholds too conservatively and pickers idle waiting for stock; set them too aggressively and product shuttles back and forth between bulk and forward areas.
My verdict for small and mid-size operations: don't try to run everything at once. Start with velocity tiering plus fixed slots for A-class SKUs. Layer in complexity only after the basics are running clean.
A Bad Priority Chain Is Worse Than No Rules
WMS putaway rules typically execute as a priority cascade — rule one fires, and if it doesn't match, the system falls through to rule two, and so on. If that chain is ordered wrong, detailed rules are decoration.
The classic mistake is putting exception rules too early. I audited a warehouse whose first rule was "fragile items go low," with ABC zoning second. A fragile but extremely high-velocity SKU got parked in a remote low-level bay — and got picked a hundred times a day. The correct order is: tier by velocity first, then narrow by physical attributes (fragile, heavy, temperature-controlled), then optimize for space utilization. Fast movers always claim the good real estate; exceptions refine within the good zones rather than surrendering them.
Exception SKUs deserve their own explicit rule list, not footnotes inside generic rules. The categories that need standalone handling: oversized or overweight items (exceeding rack load ratings), temperature-controlled or hazardous goods (compliance-driven storage requirements), high-value theft-prone items (monitored zones), and lot/expiry-sensitive product (strict FIFO). Give each category its own rule, slotted at the right level of the chain. The worst habit I see is burying an exception condition in a rule's comment field — three months later, nobody remembers what that note meant.
Then there are rule conflicts. The same pallet is both A-class and overweight; two rules match — who wins? Before go-live, build a conflict matrix and walk through the combinations that will fight each other. My personal ritual: have the warehouse manager hand-write expected locations for the top 20 SKUs, then compare against what the system actually produces. If fewer than 80% match, the rule chain has a problem.
When Pickers Are Walking Too Much, Follow This Trail
Broken rules don't report themselves. They show up as tired pickers. The diagnosis runs in three steps, from data to floor.
Start with pick-path data. Sort pick lines by the distance from each SKU's assigned location. If your A-class SKUs have longer average pick distances than B-class, the rule chain is suspect. While you're in there, check how many SKUs are fragmented across five or more locations — that's the signature of unconstrained random putaway.
Next, pull replenishment records. Frequent replenishments with chronic forward-pick stockouts mean inbound went to the wrong zones, or the replenishment thresholds are disconnected from the putaway logic. Conversely, a bulging bulk area and empty pick faces means the rules are stuffing product deep into storage and nobody is feeding the pick faces. Those two reports together tell you exactly which layer is broken.
Finally, walk the floor. If the data looks fine but pickers still complain, check whether location labels match reality. I've seen rules that were perfectly configured against location master data where the rack heights were wrong — the system thought a slot was at ergonomic low level when it actually required a ladder. Clean data doesn't guarantee a clean reality.
One more thing before I close: here's the go-live checklist I use with every client. Print it, check it off.
- Hand-wrote expected locations for the top 20 SKUs and compared them against system output
- ABC tiers are based on the last 3 months of data, not last year's
- Exception SKUs (oversize / temperature / high-value / expiry) have standalone rules — nothing buried in comment fields
- Rule conflicts walked through; you know who wins when A-class and overweight collide
- Random putaway is constrained: same-SKU locations limited to adjacent aisles
- Replenishment thresholds aligned with putaway rules; forward-pick safety stock isn't a guess
- Location master data (height, load rating, temperature attributes) spot-checked for at least 20 locations
- During week one after go-live, A-class pick distances reviewed daily — adjust the same day if anything drifts
