Most eBay sellers set reorder points as a flat rule: "reorder when stock hits 5 units." Simple, but wrong for two reasons. First, it doesn't account for demand variance (some SKUs sell 20/day, some 1/month — same 5-unit rule breaks both). Second, it ignores supplier lead time (a 3-day-lead item at 5 units gives different runway than a 30-day-lead item at 5 units).
The correct formula
Reorder point = (Average daily sales × Lead time in days) + Safety stock
Where:
- Average daily sales = units sold over last 30 days / 30
- Lead time = days from order-to-supplier until stock is in your warehouse
- Safety stock = buffer for demand variance and lead-time variance
Applied example
SKU: 2013 F-150 headlight assembly. Last 30 days sold 60 units (avg 2/day). Supplier lead time: 14 days. You want 95% service level (never stock out).
Base reorder point: 2 × 14 = 28 units. If you reorder at exactly 28, you'll stock out about 50% of the time due to demand variance.
Add safety stock: for 95% service level, add ~1.65 standard deviations of demand during lead time. If your demand std dev is 1.5 units/day, safety stock = 1.65 × 1.5 × √14 = 9 units.
Actual reorder point: 28 + 9 = 37 units. When stock hits 37, order the next batch.
Safety-stock shortcuts
Full statistical calculation is overkill for most SKUs. Practical shortcuts:
High-variance SKUs (fashion, seasonal): safety stock = 50% of lead-time demand. Reorder at 1.5× the base point.
Medium-variance (auto parts, consumer goods): safety stock = 25% of lead-time demand. Reorder at 1.25× the base point.
Low-variance (staples, high-volume): safety stock = 15% of lead-time demand. Reorder at 1.15× the base point.
Lead time uncertainty
The formula above assumes lead time is fixed. Reality: lead times vary. Your supplier's 14-day promise is really "13-19 days depending on their upstream constraints."
Two adjustments:
1. Use worst-case lead time (95th percentile) in the formula, not average
2. Track supplier lead-time reliability. Suppliers who consistently hit their promise let you reduce safety stock. Suppliers who miss increase it.
When to hold zero safety stock
Some SKUs shouldn't have safety stock — the carrying cost exceeds the stockout cost. Examples:
- Discontinuation-risk items (supplier ending production)
- High-cost, low-volume (each unit ties up significant capital)
- Items you can substitute with alternative SKUs
For these, reorder at just-in-time (base reorder point, no buffer). Accept occasional stockouts.
Demand seasonality
The 30-day rolling average breaks around seasonal shifts. In November-December for consumer categories, adjust reorder points upward 30-50%. In January-February, downward.
Auto-parts often have counter-cyclical patterns (repair season peaks in winter for northern climates). Track your category's rhythm.
Software vs spreadsheet
Under 100 SKUs: spreadsheet is fine. Manual reorder-point calculation per SKU, updated monthly.
100-1000 SKUs: inventory management software calculates rolling averages and reorder points automatically from sales data. Setup once, review weekly.
Over 1000 SKUs: automation is essentially required. Manual calculation at this scale means either missing reorder signals or over-ordering.
Profitio's inventory module calculates dynamic reorder points from live eBay sales data. 14-day free trial.