Most Shopify inventory problems are not caused by bad luck or market forces. They are caused by the same ten operational mistakes, repeated across thousands of stores, that erode margins slowly enough that most merchants never connect cause and effect.
Inventory management does not fail dramatically. It fails gradually.
A few phantom units here. A stockout there. A purchase order placed too late. An MOQ accepted without analysis. A supplier change that was never reflected in safety stock. Each failure is small enough to shrug off — until the aggregate damage shows up in your margins, your cash flow, and your customer reviews, and you are not sure where it came from.
These are the ten mistakes that cause it.
Shopify's inventory count reflects transactions processed through Shopify. It does not reflect physical reality unless every stock movement — returns, damage write-offs, theft, receiving discrepancies, manual adjustments — is logged accurately every single time.
In practice, it is not. The gap between Shopify's count and physical stock grows with every unlogged event. Merchants who trust Shopify's number without periodic physical counts are operating on fiction.
Fix: Count your physical stock against Shopify's record regularly. Treat any discrepancy as a signal to investigate, not a number to overwrite.
"Looks like about three weeks of stock left" is not a reorder point. It is a guess — and guesses average out to stockouts roughly half the time when demand or lead time varies.
A reorder point is a number: (Average Daily Demand × Lead Time in Days) + Safety Stock. When stock hits that number, an order is placed. Not when it "looks low." Not when someone happens to check.
Fix: Calculate reorder points using the formula for your top 20 SKUs. Set system alerts to fire when stock crosses the threshold.
Most merchants use their supplier's quoted lead time. The quoted time is the average in normal conditions. It is not the worst case — and the worst case is when a late delivery causes a stockout.
If your supplier quotes 14 days but sometimes delivers in 21, planning around 14 days means you will stockout every time they deliver at the slow end of their range.
Fix: Measure your actual historical lead times. Use the 80th or 90th percentile — not the average — for safety stock and reorder point calculations.
Safety stock is calibrated to your current demand pattern and your current supplier's reliability. Both change. A product that used to sell 10 units per day now sells 18. A supplier who used to deliver in 14 days now averages 20 after a warehouse move.
Safety stock set 18 months ago is not protecting you against today's risks. It is protecting you against risks that may no longer exist, while leaving you exposed to risks that are now very real.
Fix: Review and recalculate safety stock quarterly for Class A SKUs. Update immediately when demand pattern or supplier lead time changes significantly.
"I always order 200 units" is not inventory management. It is a habit. Sometimes 200 is right; often it is too many (tying up capital in slow months) or too few (creating stockouts in peak months).
Order quantities should be driven by the Economic Order Quantity formula for each SKU — the quantity that minimises the combined cost of ordering and holding — adjusted for seasonal demand variations and current stock level.
Fix: Calculate EOQ for your top SKUs. Use it as the default order quantity, adjusted upward before peak season and downward when entering a slow period.
Store-level metrics — total inventory value, overall stockout rate, aggregate turnover — are useful for board reporting and investor updates. They are useless for operational decisions.
A store with healthy aggregate metrics can simultaneously have 15 SKUs at critical stockout risk and 30 SKUs sitting on 12 months of excess stock. The aggregate hides both.
Fix: Review inventory health at SKU level. At minimum, weekly for Class A SKUs. Your store-level number is the sum of SKU-level decisions — manage at the level where decisions actually get made.
A product generating £40,000 per year in revenue and a product generating £400 per year should not receive the same attention, the same safety stock calculation, or the same reorder frequency.
Most merchants apply one-size-fits-all inventory logic because it is simpler. The result is that high-value SKUs are under-managed (stockouts on your best sellers) and low-value SKUs are over-managed (too much time and capital invested in C-class products).
Fix: Run an ABC analysis. Apply high-touch inventory management to Class A SKUs and minimal-touch (or discontinuation) logic to Class C SKUs. Match management intensity to revenue contribution.
If Component A sells 15 units per day individually and is included in a bundle that sells 8 units per day, the real demand for Component A is 23 units per day — not 15.
Merchants who forecast and reorder based only on individual product sales consistently under-order components used in bundles. The result is component stockouts that make bundles unfulfillable while the individual product still shows as in stock.
Fix: Aggregate all demand streams for each component SKU — individual sales plus bundle contribution — before calculating safety stock, reorder points, and order quantities.
Your supplier ships 95 units. Your team receives the delivery, notes it looks right, and logs 100. Five phantom units are created instantly — and no one will know until a customer order cannot be fulfilled.
Supplier short-ships are common. They are not always intentional — picking errors happen. But they only get resolved if they are detected and logged at the point of receipt.
Fix: Count every inbound shipment against the PO before logging receipt. Record any discrepancy as a receiving variance. Follow up with the supplier for credit or replacement on short-ships. This one process eliminates an entire category of phantom stock creation.
The most expensive way to find an inventory problem is a customer order that cannot be fulfilled. By that point, the stock has already been oversold. The customer is already disappointed. The refund or emergency restock is already required.
Inventory problems should be detected by your systems before they reach customers — through reorder alerts, anomaly detection, variance flags on stock counts, and demand spike monitoring.
Fix: Build detection into your operations, not your customer service queue. A reorder alert that fires three weeks before stockout is free. A customer complaint about an order that cannot be fulfilled costs the order, the customer, and the review.
Looking across these ten mistakes, the pattern is consistent: reactive instead of proactive, aggregate instead of SKU-level, gut feel instead of formula, occasional instead of continuous.
The fix for each mistake is not complicated. What makes it hard is that fixing all ten simultaneously requires building new habits and new systems — and most merchants are too focused on the next sale to rebuild the operational foundation underneath it.
Start with the mistake that is costing you most. For most merchants, that is Mistake 2 (no real reorder points) or Mistake 1 (trusting Shopify's count without verification). Fix those two and you eliminate the most common sources of stockouts and phantom stock. Work through the rest systematically.
None of these problems are permanent. All of them are fixable with the right data and the right processes.
CoreCaptain detects phantom stock, sync errors, and inventory discrepancies automatically. 14-day free trial, no credit card required.
Start Free Trial