Accurate inventory at a distributor comes from a boring routine: somebody counts the same fast movers every week and chases each variance back to the step that caused it. Most of distribution inventory management is that routine.
Distribution inventory management is how a wholesaler or distributor buys, stores, tracks and ships the stock it resells, so the right quantity is in the right place when an order arrives. Two records have to agree for it to work. The ERP holds the financial one: items, quantities, cost and open orders. The warehouse holds the physical one: which bin, which building, which pallet. Nearly every problem below is a gap between the two. More posts like this one are in our warehouse operations series.
Here is what I would do with the first thirty days if I took over inventory at a distributor on Monday.
Week one: find out how wrong the count is
Start by deciding what you hold. The IRS lists what goes into inventory as merchandise or stock in trade, raw materials, work in process, finished products, and supplies that physically become part of what you sell. A distributor holds mostly the first: stock bought to resell.
Then pick the items that matter. The 80/20 rule, or Pareto principle, says a small share of causes produces most of the effects. It is a rule of thumb, not a measured law, but it is a fair way to choose where to start. Rank your items by annual sales value or by picks, and call the small group at the top your A items.
Count the A items by location and compare the results with the ERP. Do not fix anything yet. For each variance, find where the stock was last recorded and what happened to it after that. The list of answers is your plan for the other three weeks.
Week two: fix the dock and the bins
Receiving is where the record starts. Check what arrived against the purchase order before it goes anywhere: right item, right quantity, no damage. Record the receipt as it happens, then put each item in a defined bin and record the bin. Stock that is received but not put away shows in the ERP and is missing from the place pickers look.
Give every bin a barcode label and record every move against the bin. That is what lets a picker go straight to the stock and a counter count one shelf at a time. Our zone-by-zone guide to warehouse inventory management goes through the labeling.
Week three: replenish and pick
Keep fast-moving items in pick locations near packing, and refill them from overstock before they run out so pickers do not wait on a forklift. Check each pick by scan, then pack and confirm the shipment so the ERP quantity drops when the stock leaves. Our post on why barcode picking cuts mispicks covers the checking.
This is also the week to set reorder points. A reorder point is the stock level that triggers a new purchase order. It is usually the demand you expect during the supplier's lead time plus safety stock, the extra you keep for late deliveries or a spike in orders. Both only work if the on-hand number they start from is right, so counting comes first.
Week four: count on a schedule and read the numbers
Cycle counting means counting one slice of the warehouse at a time on a schedule, often by ABC class, and fixing variances as you find them. A items get counted most often and C items, the long tail, least. Counting by scanner instead of clipboard removes the step where someone types the results in later. Our guide to cycle counting with barcode scanners covers setting one up, and cycle counts compared with a full physical count covers why not to wait for the annual one.
At the food and beverage distributor I counted the fast movers every Tuesday, same bins, same route. Nobody looked forward to Tuesday. The first month the count sheet was a list of explanations. By the third month the same few bins kept showing up, and the cause was nearly always one of two things: product put away to a bin that was not the one on the label, or a pick that took the next pallet over.
Five numbers tell you whether it is working. A SKU, or stock keeping unit, is the code for one sellable item in one form, so the same part sold singly and by the case is two SKUs. A KPI, or key performance indicator, is a measure you track over time.
| KPI | What it tells you | How to calculate it |
|---|---|---|
| Inventory accuracy | System matches the shelf | Locations counted with no variance, divided by locations counted |
| Fill rate | Share of ordered lines that ship from stock | Order lines shipped complete, divided by order lines ordered |
| Inventory turnover | How fast stock sells through | Cost of goods sold, divided by average inventory value |
| Order accuracy | Orders ship right | Orders shipped without error, divided by orders shipped |
| Dock-to-stock time | Wait before received stock is pickable | Time from receipt to recorded putaway |
Start with inventory accuracy. If the counts are wrong, fill rate and turnover are calculated from wrong numbers too.
With a second building, set counts and reorder points per building and treat each as its own set of locations in one system. Record transfers between buildings as moves, not as a sale in one place and a receipt in the other. The rest is in running inventory across two warehouses.
Where software fits
Your ERP plans purchases and values stock, and it is only as current as the transactions that reach it. Warehouse software records receiving, putaway, picking and counts with a scan as they happen and sends them to the ERP. Endpoint Cloud's storage workflow, for example, tracks stock by zone, aisle, rack and bin and runs counts continuously, so the warehouse does not shut down for a physical count.
If you are still deciding whether the ERP can run the warehouse on its own, start with five questions to ask before adding a WMS.
Day 31
Count the same A items again, in the same locations, and put the two counts side by side. The variances that show up twice are the ones to chase first.
