Newsvendor Calculator
The math behind a one-time buy: too few costs you margin and customers, too many costs you the E&O meeting. Enter both costs and this gives the order quantity that maximises expected profit, plus the trade-off curve it sits on.
Built for buyers who commit to finished goods long before demand is known: technical and industrial distributors, importers and brand distributors of consumer durables, and private-label ecommerce sellers. One order, a supplier minimum, and a lead time you cannot shorten.
Your numbers
Results update as you type. Nothing is sent anywhere, the calculation runs in this browser tab.
Show the numbers
| Order | Expected profit | Stock-out risk | Expected leftover |
|---|---|---|---|
| 593 | €24,281 | 99% | 0 |
| 620 | €25,354 | 98% | 1 |
| 647 | €26,407 | 97% | 1 |
| 674 | €27,434 | 96% | 2 |
| 701 | €28,421 | 94% | 4 |
| 728 | €29,358 | 92% | 5 |
| 755 | €30,235 | 89% | 8 |
| 782 | €31,037 | 86% | 11 |
| 809 | €31,752 | 82% | 16 |
| 836 | €32,370 | 77% | 21 |
| 863 | €32,884 | 73% | 28 |
| 890 | €33,299 | 69% | 36 |
| 917 | €33,605 | 64% | 45 |
| 944 | €33,803 | 60% | 55 |
| 971 | €33,877 | 55% | 66 |
| 974 | €33,878 | 54% | 68 |
| 998 | €33,835 | 50% | 79 |
| 1,025 | €33,679 | 45% | 93 |
| 1,052 | €33,417 | 41% | 109 |
| 1,079 | €33,064 | 38% | 125 |
| 1,106 | €32,619 | 34% | 142 |
| 1,133 | €32,088 | 30% | 161 |
| 1,160 | €31,465 | 26% | 180 |
| 1,187 | €30,759 | 23% | 200 |
| 1,214 | €29,983 | 21% | 221 |
| 1,241 | €29,144 | 18% | 243 |
| 1,268 | €28,239 | 16% | 266 |
| 1,295 | €27,282 | 13% | 289 |
| 1,322 | €26,275 | 11% | 312 |
| 1,349 | €25,229 | 10% | 336 |
| 1,376 | €24,155 | 9% | 361 |
| 1,403 | €23,049 | 8% | 386 |
| 1,430 | €21,915 | 6% | 411 |
| 1,457 | €20,754 | 6% | 436 |
| 1,484 | €19,573 | 5% | 462 |
| 1,511 | €18,373 | 4% | 487 |
| 1,538 | €17,156 | 3% | 513 |
| 1,565 | €15,926 | 3% | 540 |
| 1,592 | €14,683 | 2% | 566 |
| 1,619 | €13,427 | 2% | 592 |
| 1,646 | €12,162 | 2% | 619 |
| 1,673 | €10,889 | 1% | 645 |
How this is calculated
| Critical ratio | 46% | Cu ÷ (Cu + Co) = €41 ÷ (€41 + €48) |
|---|---|---|
| Optimal quantity | 974 units | the demand quantile at that ratio, F⁻¹(Cu / (Cu + Co)) |
| Demand model | log-normal | 4,000 draws with your mean and spread; positive and right-skewed, the way demand behaves. Seeded, so the answer never moves on its own. |
The classic newsvendor result is a formula, not a simulation. This page simulates anyway, because the same fixed sample then produces the whole trade-off curve, and because it is the identical engine the core tool uses, so the two can never disagree.
The model in one paragraph
The newsvendor problem is the oldest question in inventory theory and still the most useful: you order once, demand is uncertain, and both mistakes cost money. Order one unit too few and you lose the underage cost, normally the gross margin you did not earn. Order one unit too many and you pay the overage cost, what the unit cost you, minus whatever you can still recover. Because the second unit you add to the order is less likely to sell than the first, there is a point where the expected gain from one more unit exactly equals the expected loss. That point is the optimum, and it is not the average demand.
The answer is a quantile, not an average. Order enough to cover demand up to its critical ratio, the underage cost divided by the sum of both costs. If a shortage costs three times what an unsold unit costs, the ratio is 75% and you deliberately plan to have stock left over most of the time. If unsold units are written off entirely while margins are thin, the ratio can fall below half, and running out regularly is the correct, profit-maximising behaviour. The counter-intuitive part is that the optimum tracks the balance of the two costs, not your comfort with either one.
In distribution, both numbers are usually worse than they look on paper. The underage cost is not only the margin on the missed line: fill rate is a top- three supplier-selection criterion for most B2B buyers, and customers who hit a stock-out cut their follow-up orders measurably. The overage cost is not just the unit price either, dead stock carries at roughly a quarter of its value per year and liquidates at ten to forty percent of cost. If you enter honest numbers for both, the optimum this page returns will usually sit higher than the quantity that feels comfortable.
Where these numbers come from
- Sector patterns are built-in assumptions, a launch curve shape and, where the sector has one, a seasonal pattern. Each one names the research it rests on, or says plainly that it is our judgment, on where the curves come from. How wide the range is comes from your own answer about how new the product is, not from the sector. They are not pooled customer data, and nobody else's numbers are mixed into your result.
- The economics are your inputs, applied openly, every calculation on this page is shown with its formula. There is no model that "knows better" than the numbers you entered.
- Nothing is uploaded. The calculation runs in this browser tab. If you supply your own launch history, it is parsed here and stored on this device only; there is no server to send it to, and no account to create.
SKUZero simulates one from launches like yours. Your own past launches if you upload them, or typical sector patterns if you have none. Same engine, same economics, applied to a real first-buy decision.
Related
- The newsvendor model explained: worked through with numbers, no calculus.
- Initial Order Quantity Calculator: the same maths applied to a product launch, demand included.