Knowing the formula is the easy part. The actual work of calculating EOQ for a real product is finding three numbers that are almost never sitting in one clean report waiting for you — annual demand, ordering cost, and holding cost. This is a walkthrough of where to actually find each one, the judgment calls involved, and what to do once you have a result.
What You Need to Gather
Three numbers, and only three:
- Annual demand (D) — how many units you expect to use or sell this year
- Ordering cost (S) — what it costs, in dollars, to place and receive one order
- Holding cost (H) — what it costs to hold one unit in inventory for a year
Everything else is judgment about whether these numbers are trustworthy enough to build a decision on.
Step 1: Estimate Annual Demand
Start with your actual sales or usage history for the past 12 months, not a guess. If you have inventory or point-of-sale software, this is usually the easiest of the three numbers to pull — export units sold or units consumed for the SKU over the last full year.
What to watch for:
- Trend. If demand is climbing or falling meaningfully month over month, last year’s total understates or overstates where you actually are now. Weight recent months more heavily, or use a short-term average instead of the full year (see Choosing a Demand Forecasting Method for specific techniques).
- Seasonality. If a big chunk of annual volume happens in a few months, an annual-average demand number will mislead you — EOQ assumes roughly steady demand, and seasonal items break that assumption. EOQ Reality Check covers what to do instead.
- New products. No history means no reliable D. Use a comparable product’s early performance, a supplier or sales forecast, or hold off on EOQ until you have a few months of real data.
Step 2: Find Your Ordering Cost
This is usually the hardest number to pin down, because “ordering cost” rarely exists as a line item anywhere. It’s an aggregate of several smaller costs that finance and procurement track separately, if they track them at all.
What to include:
- Purchasing staff time spent creating and sending the order
- Receiving and inspection labor when the shipment arrives
- Freight or expediting fees tied to placing the order (not the per-unit shipping cost — the fixed part)
- Payment processing and invoice-matching overhead
A reasonable starting formula: take your total purchasing-plus-receiving department cost for the year and divide by the number of purchase orders placed in that same year. That gives you an average cost per order across your whole catalog, which is a fine starting point even if it’s not perfectly precise for any one SKU. If a specific product has unusually high or low ordering complexity — say, it always requires extra customs paperwork — adjust from that baseline rather than starting from scratch.
Common mistake: using only the line-item freight cost and ignoring labor. Labor is often the larger share of ordering cost, and skipping it makes your ordering cost look artificially low, which pushes your calculated EOQ artificially small.
Step 3: Calculate Holding Cost
Holding cost has four components, and most people only remember the first one.
- Storage — the actual warehouse space cost, whether that’s rent, a share of a mortgage, or a per-pallet fee at a third-party facility
- Insurance — coverage cost scaled to the value of inventory held
- Capital cost — the return you’re giving up on cash tied up in inventory instead of invested or paying down debt; a common shortcut is your company’s cost of capital or borrowing rate
- Obsolescence and shrinkage risk — the probability-weighted cost of inventory that goes unsold, expires, gets damaged, or walks off
Add these up as a percentage of the unit’s value, then multiply by the unit cost to get a dollar figure per unit per year. As a reasonable industry starting point, total holding cost usually lands between 20% and 30% of a unit’s value annually — higher for perishable or fast-obsolescing goods, lower for stable, durable, cheap-to-store items. The Carrying Cost calculator breaks this out on its own if you want to isolate just this piece.
Common mistake: using only the storage-rent portion and ignoring capital cost. On a moderately expensive item, the capital-cost component is often larger than physical storage, and leaving it out understates H — which pushes your calculated EOQ artificially large.
Step 4: Plug Into the Formula
With D, S, and H in hand, the math itself takes seconds. Say you’ve gathered:
- Annual demand: 8,000 units
- Ordering cost: $120 per order
- Holding cost: $6 per unit per year
EOQ = √((2 × 8,000 × 120) / 6) = √320,000 ≈ 566 unitsThat’s your answer in isolation: order roughly 566 units at a time. Orders per year would be about 8,000 / 566 ≈ 14, meaning you’d place a new order every 3-4 weeks. If you want to see the full mechanics of this calculation, including the formula’s derivation, see EOQ Formula Explained.
Step 5: The Reality Check
This is the step people skip, and it’s the one that actually determines whether the number is usable.
- Does this match your supplier’s terms? If your supplier has a minimum order quantity above your EOQ, the math result isn’t achievable as-is. Compare the extra holding cost of ordering the minimum against the alternative of finding a different supplier or negotiating.
- Does it fit your storage? A calculated EOQ that exceeds your available space for that SKU isn’t a number you can act on directly — you’ll need to either free up space, split deliveries, or accept a smaller, more frequent order at a slightly higher blended cost.
- What’s actually driving the number? If EOQ came out surprisingly large or small, trace it back — is D unusually high this year because of a one-off bulk sale? Is S inflated because you included an unusually expensive rush order? A result that looks off is often telling you something about your inputs, not the formula.
What Changed? EOQ Decision-Making in Practice walks through exactly this reality-check process with a full worked scenario, including what to actually do when the “right” answer and the “possible” answer disagree.
Common Data-Gathering Mistakes
- Using last year’s costs without checking if they’ve changed. Freight rates, warehouse rent, and interest rates all drift; a holding cost calculated two years ago is probably stale.
- Averaging demand across dissimilar periods. Twelve months that include one abnormal quarter (a promotion, a stockout, a one-time bulk customer) will skew your average in a way that doesn’t reflect ordinary demand.
- Forgetting that “ordering cost” and “holding cost” need to be in the same units. If S is a one-time cost per order and H is meant to be annual, mixing up the time horizon will throw off the whole calculation.
- Recalculating once and never again. Revisit D, S, and H whenever any of them shifts meaningfully — a new freight contract, a warehouse rent increase, a real change in sales velocity.
Where to Go From Here
Once you’ve gathered real numbers, the EOQ Calculator does the arithmetic instantly and the Excel template lets you save your inputs and revisit them later without redoing this whole process. If you’re managing more than a handful of SKUs, an ABC Analysis first will tell you which products are worth this level of data-gathering effort and which aren’t.