inventory control (order or not each period)
You sell 100 phone cases a week on average, but actual weekly sales swing between 60 and 140. Your supplier needs two weeks to deliver. Right now you have 250 cases in your warehouse. Do you place an order today - and if so, for how many?
Inventory Control is the recurring decision rule of whether to order and how much to order each period, balancing the Error Cost of tying up cash in excess inventory against the Error Cost of running out when Demand arrives.
Inventory Control is the decision rule you apply each period (day, week, month) to answer two questions:
You already know that inventory ties up cash as a Current Asset on the Balance Sheet. Inventory Control is the system that determines how much cash stays tied up. Too much inventory and you're bleeding opportunity cost - that cash could be deployed elsewhere. Too little and Demand shows up with no product to sell, which means lost Revenue.
The difficulty is that Demand is uncertain. If you knew exactly how many units you'd sell next week, Inventory Control would be trivial arithmetic. You don't, so it becomes a problem of making good decisions under Variance.
Inventory Control hits the P&L and the Balance Sheet simultaneously:
The core tension: you're choosing between two Error Costs - the cost of having too much versus the cost of having too little. Neither is zero, and they pull in opposite directions.
Every Inventory Control system, from a spreadsheet to an enterprise platform, rests on three inputs:
How many units do you expect to sell per period? This comes from historical data, and it's never exact. You need both the average and the Variance - a product that sells 100 +/- 10 per week is a very different problem than one that sells 100 +/- 50.
How long between placing an order and receiving it? If your supplier takes 2 weeks to deliver, you need enough inventory right now to cover 2 weeks of Demand - plus a safety margin for the Variance in that demand.
Two Error Costs fight each other:
The simplest effective approach:
If current inventory falls below (average demand per period x lead time in periods) + safety margin, then order.
The safety margin accounts for Variance in Demand. You calculate it by multiplying the Standard Deviation of demand (scaled to the lead time window) by a chosen multiplier. How much to order? At minimum, enough to bring your post-delivery inventory back above the threshold. In practice, you order enough to cover expected Demand through the next ordering cycle plus the safety margin. This is a cost minimization problem: you're minimizing the Expected Value of total cost (excess carrying + stockout losses) across periods.
The multiplier on Standard Deviation is a tunable parameter, not a fixed constant. It determines how often you accept running out of stock:
| Multiplier | Coverage | Practical Meaning |
|---|---|---|
| 1.0 | ~84% | You stock out roughly 1 in every 6 ordering cycles |
| 1.5 | ~93% | Roughly 1 in every 15 cycles |
| 2.0 | ~98% | Roughly 1 in every 50 cycles |
The right choice comes from comparing what a stockout costs you (lost Profit per unit) to what extra inventory costs you (carrying cost per unit per period). When Profit per unit is high relative to carrying cost, a higher multiplier is cheap insurance. When margins are thin or carrying costs are steep (perishable goods, high Obsolescence risk), a lower multiplier may be appropriate.
In the phone case example below, each unit earns $10 in Profit while costing $0.05 per week to carry. An extra safety unit that sits unsold for 10 weeks costs only $0.50 in carrying but prevents a $10 loss if Demand spikes. At that ratio, a multiplier of 1.5 or higher is easily justified. A grocery store selling produce at $0.50 Profit per unit with $0.10/week spoilage cost faces a much tighter ratio and should use a lower multiplier.
You need a deliberate Inventory Control system whenever:
You do not need a sophisticated system if:
You sell phone cases at $15 each. Cost Per Unit from your supplier is $5. Weekly Demand averages 100 units with a Standard Deviation of 25 units. Your supplier takes 2 weeks to deliver. The cost of carrying inventory is $0.05 per unit per week - covering warehouse storage, insurance, and opportunity cost of tied-up cash. For phone cases this rate also reflects Obsolescence risk, since new phone models can make your stock unsellable. You currently have 250 units on hand. No orders are in transit.
Expected demand during lead time: 2 weeks x 100 units/week = 200 units.
Calculate the safety margin. The Standard Deviation of demand over a 2-week window is 25 x √2 ≈ 35 units. We choose a multiplier of 1.5 (targeting ~93% coverage - stockout roughly 1 in 15 cycles). With $10 Profit per unit at risk from a stockout versus $0.05/week to carry an extra unit, generous coverage is cheap. Safety margin: 35 x 1.5 ≈ 53 units.
Decision threshold: 200 + 53 = 253 units. Current inventory is 250, which is below 253. Order now.
Order quantity. During the 2-week lead time, you expect to sell 200 units, leaving 250 - 200 = 50 on hand at delivery. You want post-delivery inventory to cover the next cycle: 200 (expected Demand) + 53 (safety margin) = 253. Order 253 - 50 = 203 units. If your supplier ships in case packs, round up to the next pack size. Never round down on an order triggered by being below your threshold - that defeats the purpose of the safety margin.
Cost check. Cash outflow: 203 x $5 = $1,015. After delivery, inventory goes from 50 to 253, then sells down to ~53 over 2 weeks. Average inventory during the cycle: (253 + 53) / 2 = 153 units. Carrying cost: 153 x $0.05 x 2 weeks = $15.30. Expected Profit from selling ~200 units: 200 x $10 = $2,000. The $15.30 carrying cost is less than 1% of the Profit at risk from running out.
Insight: Your current stock (250) was just 3 units below the threshold (253). Without a disciplined decision rule, you might have waited another week - by which point you'd have ~150 units, well below the 253 threshold, and your order wouldn't arrive for 2 more weeks. You'd almost certainly run out mid-cycle, losing roughly 50 units of sales: $750 in Revenue and $500 in Profit. The $15 in carrying cost is insurance against that outcome.
Your average weekly Demand of 100 phone cases at $15 each (cost $5) is forecast to jump to 250 units during a 4-week holiday season. Demand Variance is higher during holidays - Standard Deviation rises from 25 to 50 units per week. Supplier still takes 2 weeks to deliver. The holiday season starts in 3 weeks. Current inventory: 300 units. Carrying cost: $0.05 per unit per week.
Holiday demand over the 4-week season: 250 x 4 = 1,000 units. Your supplier's 2-week lead time means you can't order reactively during the season. You need to plan your inventory position before the season starts.
Current stock timeline. You have 3 weeks of normal Demand (100/week = 300 units) before the season. Your current 300 units will be entirely consumed by normal sales before the season begins. Any inventory you want at season start must come from a new order.
Pre-season order. If you order today, delivery arrives in 2 weeks - one week before the season. At that point you have 300 - 200 = 100 units remaining from current stock, plus the new order. During the last pre-season week, normal Demand of 100 draws from that remaining current stock. At season start, your inventory equals the order quantity. To cover the first 2 weeks of holiday Demand until a mid-season delivery can arrive: 250 x 2 = 500 expected Demand, plus safety margin of 50 x √2 x 1.5 ≈ 106 units. Target at season start: 606. Order 606 units. Cost: 606 x $5 = $3,030.
Mid-season order. Place this order at the start of holiday week 1; it arrives at week 3. Inventory at start of week 3: 606 - 500 (two weeks of holiday sales) = 106 units (your safety margin, as expected). Need for weeks 3-4: 500 + 106 = 606. Already have 106 on hand. Order 500 units. Cost: 500 x $5 = $2,500.
Totals. Capital Investment in holiday inventory: $3,030 + $2,500 = $5,530. Expected holiday Revenue: 1,000 x $15 = $15,000. Expected holiday Profit before carrying cost: 1,000 x $10 = $10,000. Carrying cost over the season: roughly $100 (average inventory of ~350 units for 4 weeks at $0.05/unit/week, plus a week of pre-season storage). Net holiday Profit: ~$9,900. ROI on the $5,530 is strong - but only if Demand actually materializes.
Insight: Seasonal Inventory Control forces you to commit cash before you observe the Demand. You're investing $5,530 based on a forecast of 250 units/week. Sensitivity Analysis matters here: if holiday Demand comes in at only 150/week instead of 250/week, you sell 600 units over 4 weeks and end the season with roughly 400 unsold. At $5 each, that's $2,000 in inventory facing markdowns or Obsolescence. Your ROI swings from +179% to potentially break-even or worse. The higher the seasonal multiple over your base rate, the more this becomes a Capital Investment decision under uncertainty - not a routine restocking.
Inventory Control is a recurring decision rule, not a one-time setup. Every period you're choosing between the Error Cost of excess (carrying cost, Obsolescence) and the Error Cost of shortage (lost Revenue and Profit).
The safety margin multiplier is a business decision you tune, not a fixed constant. It depends on Demand Variance, supplier lead time, and the ratio of what a stockout costs you to what extra inventory costs you. Higher Profit per unit relative to carrying cost justifies a higher multiplier.
Getting Inventory Control right directly improves Cash Conversion Cycle and Working Capital Management - you free up cash without sacrificing Revenue.
Ignoring Variance and ordering to the average. If you order exactly 100 units because you sell 100 on average, you'll run out roughly half the time. The safety margin exists because Demand is a distribution, not a point estimate - this is the same logic behind Expected Value thinking.
Treating inventory as free to carry. Every unit sitting in a warehouse has a real cost: storage, the opportunity cost of tied-up cash, and the risk of Obsolescence. Operators who only worry about running out tend to over-order and quietly bleed Profit through carrying costs that never show up as a single dramatic line item on the P&L.
Rounding order quantities in the wrong direction. When your decision rule says to order 203 units, do not round down to 200 for a clean number. You triggered the order because inventory was already below your threshold. Rounding down means your post-delivery position is still below the safety margin you calculated - you paid for the order and waited through the lead time but didn't restore your protection. If you round at all, round up to the next case pack or pallet size.
You sell widgets at $25 each with a Cost Per Unit of $10. Weekly Demand averages 60 units (Standard Deviation: 15). Your supplier takes 2 weeks to deliver. You currently have 140 units on hand. Should you order? If so, how many?
Hint: Calculate the threshold: (average demand x lead time weeks) + safety margin. For the safety margin, scale the Standard Deviation to the lead time window (multiply by √2 for a 2-week window) and apply a 1.5 multiplier. Compare the threshold against current inventory.
Expected Demand during the 2-week lead time: 60 x 2 = 120 units. Standard Deviation over 2 weeks: 15 x √2 ≈ 21 units. Safety margin at 1.5 Standard Deviations: 21 x 1.5 ≈ 32 units (round up). Threshold: 120 + 32 = 152. Current inventory is 140, which is below 152. Order now. Expected inventory at delivery: 140 - 120 = 20 units remaining. Order to restore post-delivery inventory to the threshold level: 152 - 20 = 132 units. Cash outflow: 132 x $10 = $1,320. Those 132 units generate Revenue of 132 x $25 = $3,300 and Profit of 132 x $15 = $1,980 as they sell through.
Same widget business. Your supplier offers a 10% discount if you order 200+ units at once. The cost of keeping inventory on hand - storage, insurance, and opportunity cost on tied-up cash - runs about $0.04 per unit per week, or roughly 21% of item value annually. Under your normal ordering pattern you order about 132 units each cycle. Should you take the bulk discount?
Hint: Calculate the discount savings first. Then calculate the incremental carrying cost - the difference between carrying cost under the bulk order versus your normal order, not the total carrying cost of 200 units against zero. Use the fact that each unit in a batch of Q sits in inventory an average of Q / (2 x weekly Demand) weeks before it sells.
Discount savings: 200 x $1 = $200 versus buying at full price. Now compare carrying costs. Under the normal pattern (132 units per order), each unit sits in inventory an average of 132 / (2 x 60) ≈ 1.1 weeks. Per-unit carrying cost: $0.04 x 1.1 = $0.044. Under the bulk order (200 units), each unit sits an average of 200 / (2 x 60) ≈ 1.67 weeks. Per-unit carrying cost: $0.04 x 1.67 = $0.067. Incremental carrying cost per unit: $0.067 - $0.044 = $0.023. Over 200 units: $4.60. The $200 discount dwarfs the $4.60 in extra carrying cost. Take the discount. At realistic carrying cost rates (15-35% of item value annually), the carrying arithmetic almost never flips against a 10% discount on non-perishable goods. The real reasons to decline are practical: (1) Cash Flow - the bulk order requires $1,800 upfront versus $1,320, so you need $480 more available cash. (2) Obsolescence - if the product faces a refresh before you sell through 3.3 weeks of inventory, unsold units may require markdowns. (3) Storage capacity - you need room for 200 units instead of 132. A student who internalizes this comparison method will also recognize that the important variables are often outside the formula.
You run an e-commerce store with 50 different products. Ten of them represent 80% of your Revenue. You have limited time to build Inventory Control systems. How do you prioritize?
Hint: Think about where Inventory Control errors have the highest Error Cost. Which products have the most Profit at risk from a shortage? Which have the most cash at risk from over-ordering?
Focus Inventory Control on the top 10 products first. They drive 80% of Revenue, so a stockout in any of them has a large Error Cost. For the remaining 40 products (20% of Revenue), simple rules work fine - order a fixed amount when stock looks low. This is a resource allocation and Triage problem: your time as an Operator is the Bottleneck, so you allocate it to the highest-Profit items. Within the top 10, prioritize the ones with the highest Demand Variance and longest supplier lead times - those are the products where getting Inventory Control wrong costs you the most. A product with steady, predictable sales and next-day delivery barely needs a system. A product with volatile weekly sales and a 4-week lead time can destroy your Cash Flow or your Revenue line depending on which direction you get it wrong.
Inventory Control sits directly on top of two prerequisites. Inventory taught you what inventory is and why it matters on the Balance Sheet - Inventory Control is the system you use to decide how much of it to hold. Demand taught you that the quantity customers want to buy is uncertain and external - Inventory Control is how you make ordering decisions despite that uncertainty. Looking forward, this connects to Cash Conversion Cycle (better Inventory Control shortens the time between paying suppliers and collecting Revenue), Working Capital Management (the safety margin directly sets how much cash is locked in Current Assets), and Cost Reduction (reducing carrying costs and stockout losses simultaneously is one of the highest-leverage moves in Operations). It also links to Sensitivity Analysis - every safety margin decision is implicitly a bet on the range of Demand, and you should stress-test what happens when your forecast is wrong.
Disclaimer: This content is for educational and informational purposes only and does not constitute financial, investment, tax, or legal advice. It is not a recommendation to buy, sell, or hold any security or financial product. You should consult a qualified financial advisor, tax professional, or attorney before making financial decisions. Past performance is not indicative of future results. The author is not a registered investment advisor, broker-dealer, or financial planner.