A retail buyer sends a purchase order for 40 cases, to ship the week of the 10th. Your web store is selling the same product tonight, six units at a time, off the same shelf. Neither system knows about the other, and nothing stops those 40 cases from being sold twice.
This page covers what changes when a wholesale order lands on a shelf the web store already sells from, how to hold stock back for it, and how to keep the two channels in agreement week to week. It assumes you already run the storefront well; the operator's playbook for CPG brands covers that starting point.
What changes when a wholesale order and the web store share one shelf?
The storefront's count stops meaning sellable. Once you accept a purchase order, part of the shelf belongs to a buyer who will not take it for two weeks, and the web store cannot see that. Three numbers now matter per SKU: what the shelf holds, what open orders have promised, and what is left for the storefront to sell.
| Line | Units | Where it comes from |
|---|---|---|
| On hand | 600 | The shelf count |
| Promised on open orders | 480 | 40 cases × 12, due the week of the 10th |
| Sellable online | 120 | On hand minus promised |
| Web run rate | 6 a day | Last eight weeks |
| Days of cover | 20 | Sellable online ÷ run rate |
The shelf count is the one number you already have, and it is the one the storefront shows. The promised figure is the one nobody writes down: the buyer's 480 units sit on the shelf looking exactly like stock for sale. Sellable online is the difference, and it is the only number the web store should ever be allowed to sell against.
Direct sales take units, daily, in ones and twos. Wholesale takes cases, in bursts, against a date. Both pull on one shelf, so the fix is not a second shelf; it is writing the promised figure down the moment it exists.
How do you hold stock back for a purchase order?
Reduce what the storefront may sell by the order quantity on the day you accept the order, not the day you ship it. Keep the promised figure per SKU in the same sheet as the shelf count, with the order number and the ship window beside it, so sellable online is always on hand minus promised.
Do the hold-back in units. The buyer orders cases, but the shelf is counted in units and the web store sells units, so the conversion happens once, on the order line, and never in the on-hand column. Write the case pack on the product record as well as on the order: a product that shipped in twelves last year and ships in sixes now still reads "40 cases" on the buyer's email.
If sellable online goes negative on paper, an oversell is waiting, and the sheet has told you before the buyer or a web customer does. The honest moves are to buy or make ahead, to short the web store on purpose for a few days, or to ask the buyer about a split shipment before the window opens. All three are cheaper than a short ship.
How do you plan the pick around the ship window?
Put every open order on a calendar with its ship window and its case count, and read it against the deliveries or production runs on their way. An order you can fill today but must ship in three weeks ties up stock for three weeks. An order you cannot fill today is a buy or make decision now, not in the window.
The window is the buyer's rule, and shipping late or short usually costs a deduction from your invoice. In the example above, the web store has 20 days of cover and the window opens in 12, so the order fits. A second order of the same size would not, and the calendar shows that before you say yes to it.
What the buyer expects in that window, from case labels to the routing guide, comes with the vendor setup packet. The calendar is how you avoid earning the chargebacks the packet describes.
How do you keep the storefront, the shelf and the open orders in agreement?
Once a week, for each SKU that moves, compare the three numbers: what the storefront shows as sellable, what the shelf holds, and what open orders have promised. Shelf minus promised should equal storefront sellable. When it does not, the usual causes are an order accepted without the hold-back, or a web return put back onto promised stock.
Keep the log short: SKU, expected, found, cause. Fast movers weekly, the rest monthly. After a month it shows which step leaks: an oversell is the symptom, a missing hold-back is nearly always the cause.
Where the wholesale stock sits in a second building, the same three numbers apply per location, and every transfer between the two buildings has to move the promised figure with the units.
What should you do next?
Set up
- List every open order with SKU, cases, case pack and ship window.
- Write the case pack on every product record.
On the day you accept an order
- Subtract the promised units from sellable online.
- Flag any SKU where sellable online goes negative.
Each week
- Compare storefront sellable, the shelf, and promised units for the SKUs that move.
- Log every gap with a cause before correcting it.
Frequently asked questions
What is wholesale inventory management?
It is running one stock count for a brand that sells both to consumers and to stores or distributors, so that the units a purchase order has promised are held back from the storefront and both channels sell against a number that is true.
What if a purchase order does not fit the stock on the shelf?
Decide before you accept it: buy or make the shortfall ahead of the window, short the web store for a few days on purpose, or ask the buyer for a split shipment. Finding out at pick time is the expensive version of all three.
Should wholesale and direct stock be kept in separate locations?
Not usually. A second physical pool doubles counts and transfers. One shelf with a promised figure per SKU does the same job. Separate locations earn their place when a 3PL or a co-packer ships wholesale for you.