Good inventory management means never losing a sale to a stockout, and never tying up cash in overstock. This guide connects the core ideas — from forecasting demand to knowing exactly what to reorder and when — with a deeper guide for each.
Start with forecasting
Everything rests on predicting demand. Seehow to forecast inventory on Shopifyfor the overview, and forecasting methods(moving average, weighted average, exponential smoothing) for the how.
Sales velocity
Sales velocity — units sold per day — is the number that drives everything else. Use a recent window so it reflects current demand.
The reorder point
Your reorder point is the stock level at which you place a new order: (sales per day × lead time) + safety stock. Use the free calculator to find it per product.
Safety stock
Safety stock is the buffer that absorbs demand spikes and late deliveries — enough to avoid stockouts, not so much it ties up cash.
How much to order — EOQ
Economic order quantity balances ordering cost against holding cost to find the order size that costs the least overall. The reorder point sayswhen; EOQ says how much.
Measure health: turnover and days on hand
The inventory turnover ratio shows how many times you sell through your stock in a period; days of inventory on hand flips that into a countdown of days left. High turnover without stockouts is the goal.
Prioritise with ABC analysis
You can't watch every SKU equally. ABC analysissorts products into A, B and C classes so your best sellers never run dry while you stop over-managing the long tail.
One product, worked through
The four numbers above are easier to trust once you've seen them applied to something concrete. Take a candle that sells steadily.
Sold in the last 60 days 300 units Sales velocity 300 ÷ 60 = 5 units/day Supplier lead time 14 days Stock on hand 62 units Days of stock left 62 ÷ 5 = 12.4 days
Lead time is 14 days and you have 12.4 days of stock. You are already late. Not "getting low" — late. Whatever you order today arrives roughly two days after you sell out. This is the gap a low-stock alert never shows you, because 62 units doesn't look like an emergency.
The reorder point
Work out the level you should have ordered at. Suppose your worst weeks hit 8 units/day and your supplier has stretched to 20 days before:
Cycle stock 5 × 14 = 70 units Safety stock (8 × 20) − (5 × 14) = 90 units ──────────────────────────────────────────────── Reorder point 160 units
So the moment stock dropped to 160 you should have placed an order. You're at 62. Thesafety stock component is larger than the cycle stock here, which surprises people — but that's the honest cost of a supplier who is sometimes six days late on a product that sometimes sells 60% faster than average.
How much to order
Cover the lead time plus however long until you'll next review. Ordering monthly:
Demand over 14-day lead time 5 × 14 = 70 units Demand over 30-day review 5 × 30 = 150 units Safety stock = 90 units − Stock on hand − 62 units ──────────────────────────────────────────────── = Order quantity 248 units
Round to the supplier's case size and place it. Note that nothing here required a forecasting model — just velocity, lead time, and an honest view of how bad each can get. That's the whole of it for most products; the maths only gets harder when seasonality is real.
Two traps worth naming. Exclude stockout days from velocity — a week where you sold nothing because you had nothing is not a week of zero demand, and leaving those days in guarantees you under-order the things you sell best. And use your observed worst lead time, not the quoted one; suppliers quote the good case.
Moving off Stocky
Stocky stops working after 31 August 2026. See the best Stocky alternative for what to look for and how to migrate your suppliers and lead times.
Choosing a forecasting tool
If you're comparing dedicated planning tools, our guide to picking anInventory Planner alternative covers what separates them in practice — whether recommendations show their working, how they treat promotion spikes and stockout days, and how much sales history each one needs before its forecasts mean anything.
Automate the whole thing
Doing this by hand for hundreds of SKUs isn't realistic. Foreshelf reads your Shopify sales history and turns it into plain-language reorder recommendations — what's about to run out, how much to order, and when — so you plan in minutes, not spreadsheets.