An advertisement creates a promise: this product is available, this is what it costs, and this is when you can expect it. Inventory management helps your business keep that promise.
For marketers, this skill is about more than counting boxes. It connects demand generation with products the business can actually deliver. A successful campaign can become a customer-service problem if it sells stock that is damaged, reserved, or still in transit.
Know what the stock number means
Shopify’s inventory-state documentation distinguishes on-hand, available, committed, unavailable, and incoming stock. In that model, on-hand stock includes available, committed, and unavailable units; incoming stock is tracked separately.
For a fictional craft-kit store, 100 units physically present might include 20 committed to orders and five held because of damaged packaging. Only 75 are available under those assumptions. Another 40 units expected next week are not the same as 40 units ready to ship today.
Check your own platform’s definitions. In particular, confirm how reservations, bundles, transfers, returns, and location rules affect the number customers can purchase. Do not add every inventory column together: some quantities are components of others.
Track the variant that the customer actually buys
A product can appear healthy in aggregate while the most popular size or color is unavailable. Use stable identifiers for sellable variants and review stock at the relevant fulfillment location.
For bundles, identify the limiting component. If a starter box requires one hoop, two thread packs, and one fabric panel, having 200 hoops does not mean you can sell 200 boxes. The bundle’s availability depends on the full bill of materials and how your system reserves those components.
Returned stock also needs a decision. A returned item should only become available again when your inspection and restocking process says it is sellable.
Learn a simple reorder model
A starting formula is:
Reorder point = expected demand during replenishment lead time + safety stock.
Suppose a hypothetical kit sells four units per day, the replenishment lead time is ten days, and you choose 20 units as a planning buffer. The reorder point is 60 units. That buffer is an assumption to review, not a mathematically optimized recommendation.
In a simple continuous-review model, compare the reorder point with inventory position: usable on-hand stock plus confirmed incoming supply minus outstanding demand, without double-subtracting reservations already excluded from your chosen base. Align this calculation with your system’s definitions before using it.
The model becomes less reliable when demand is seasonal, lead times vary, promotions change sales velocity, or a supplier frequently misses dates. Sales during a stockout also understate potential demand. A forecast based only on completed orders can miss customers who could not buy.
Turn the model into a weekly review
For your most consequential variants, record usable stock, commitments, confirmed receipts, recent demand, supplier lead time, and planned promotions. Highlight exceptions requiring a decision.
| Situation | Question to ask | Possible response |
|---|---|---|
| Demand rising; replenishment uncertain | Can we fulfill the campaign promise? | Reduce promotion or redirect attention |
| Slow-selling stock | Is the issue discovery, fit, price, or seasonality? | Investigate before discounting |
| Stock counts disagree | Which transactions or adjustments are missing? | Reconcile and correct the cause |
| Supplier delay | What can we communicate accurately? | Update timing and affected campaigns |
A discount is not automatically the right answer to slow-moving stock. Check margin, positioning, and whether the product is reaching the right shoppers. A bundle may help customers use complementary items, but it needs honest value and accurate component tracking.
Make inventory part of campaign readiness
Before promoting a product, confirm stock by variant, replenishment timing, order capacity, and shipping promises. Decide what happens if it sells out: pause the ad, feature a suitable alternative, or offer a clearly explained preorder if the business supports one.
Ensure feeds, product pages, and campaign landing pages receive availability updates. A stock change in one system does not prove every connected channel has received it. Test the integration’s behavior and identify who responds to failures.
Use AI to explain scenarios, not place blind orders
Using this anonymized SKU table, calculate a simple reorder scenario with the stated demand, lead time, and safety-stock assumptions. Show the formula and units. Flag missing dates, stockout periods, uncertain receipts, and variants where the assumptions may fail. Recommend questions for the operator; do not place orders or invent supplier commitments.
Verify the arithmetic and inputs. Purchasing decisions still need cash constraints, minimum order quantities, shelf life where relevant, and supplier terms that may not appear in the table.
Your assignment: review ten variants, identify the three biggest fulfillment risks, and write one campaign rule for each. Use the PIM guide to separate stock management from product facts and the advertising guide to connect stock readiness with paid traffic.