Product Pricing Calculator
Cost-plus pricing misses by the fee, every time. Take an 18.00 unit cost, 6.00 of ads, 2,400 of monthly overhead over 200 units and a 12% platform fee: a 30% target margin does not give 51.43, it gives 62.07. The difference is 10.64 a unit — 2,128.08 a month, 25,536.95 a year — and it is not a rounding error, it is the fee being left out of the divisor.
| How the price is built | Amount | Note |
|---|
| Target margin | Price | Profit / unit | Profit / month | Markup on unit cost |
|---|
Divide by what is left, not by what you want
Every cost-plus formula on the internet is missing a term. The price has to cover the product, the ads, the overhead and the fee — and the fee is a percentage of the price, so it cannot be added; it has to be divided out:
price = (cogs + cpa + fixed ÷ units) ÷ (1 − feePct ÷ 100 − margin ÷ 100)
With the defaults that is (18 + 6 + 12.00) ÷ (1 − 0.12 − 0.30), which is 36.00 ÷ 0.58, which is 62.07.
The denominator is the whole story. A zero margin does not mean "price equals cost" — it means price equals cost divided by 0.88, because the fee still has to be paid. That is why the break-even price here is 40.91, not 36.00, and why the 62.07 target is 1.52 times the break-even price rather than 1.72 times.
The naive version of this formula is worth stating explicitly, because it has an exact and slightly shocking result. Divide the cost by one minus the margin and you get a price whose real margin is target minus fee. Not approximately — exactly:
| Target margin | Price from cost ÷ (1 − margin) | Real margin after a 12% fee | Correct price | Underpriced by |
|---|
At a 30% target the two prices are 51.43 and 62.07. That is 10.64 a unit, and at 200 units it is 2,128.08 a month — 25,536.95 a year, which for most small sellers is the entire profit line and then some. The identity is the useful part: cost-plus pricing returns your target margin minus your fee, so the more fees you pay, the more confidently wrong the number is.
Markup is not margin, and 100% markup is 50% margin
Markup divides profit by cost. Margin divides profit by price. They are the same number only when profit is zero, and the conversion is margin = markup ÷ (1 + markup). Read down the markup column and the margin column never catches up:
| Markup on cost | Margin it equals | Price at a 36.00 unit cost | Margin after a 12% fee | Profit / unit |
|---|
A 100% markup is a 50% margin — everyone knows that one, and most people stop there. The row that matters is the bottom: 233% markup buys a 70% margin before fees and 58.0% after them. A seller who says "I mark up 233%" and a seller who says "I run 70% margins" are describing the same price, and neither of them is describing what lands in the bank.
Your own default price sits at a 72.4% markup on the 36.00 unit cost, which is how a 30% margin is spelled in markup terms. On the 18.00 product cost alone the same price is a 244.8% markup — same price, same profit, three different percentages. This is the mechanism behind most "my margins are great but there is no cash" conversations: the percentage being quoted is measured against whichever base makes it look largest.
The fee is a hard ceiling on margin, not a cost you can outrun
Because the denominator is 1 − fee − margin, the two percentages compete for the same 100 points. Push margin up and the denominator shrinks; when the two add to 100 it hits zero and the price stops existing. There is no very large number at the end of that road — there is no number.
| Fees → / Margin ↓ | 10% fees | 15% fees | 20% fees | 25% fees |
|---|
The same grid in denominators shows where the curve steepens — these are the divisors, and the price is just 36.00 divided by each:
| Fees → / Margin ↓ | 10% | 15% | 20% | 25% |
|---|
At 12% fees your margin ceiling is 88%, and it is an asymptote rather than a target: a 88% margin leaves a divisor of exactly 0.00, and 89% leaves −0.01. The last ten points of margin are the expensive ones — moving from 60% to 70% at these costs takes the price from 128.57 to 200.00, a 71.43 increase, against 6.79 for the first ten points from 10% to 20%. That is a 10.5× difference in price per point of margin, and it is why "just price for a better margin" fails at the top of the range: the arithmetic, not the customer, says no.
| Situation | Denominator | Result |
|---|
A 20% discount is a 200% volume promise
The recovery formula is volume increase = discount ÷ (margin − discount). It is short, it is exact, and it explains why discounting is the most expensive habit in retail. At a 30% margin a 20% discount divides 20 by 10 and demands 200% more volume — 600 units a month instead of 200, to end up exactly where you started:
| Margin | Volume needed to break even (classic) | Units / month | Same, with fees falling with the price | Units / month |
|---|
The two columns differ because your fees are a percentage of the price: when the price drops, the fee drops with it, so the discount costs slightly less than the classic rule assumes. At a 30% margin the honest figure is 141.9% — 484 units — rather than 200%. Both are quoted because the classic one is the number to use in an argument and the fee-aware one is the number that is true; either way a 20% discount at a 30% margin requires more than doubling the business to stand still.
At the default 30% margin, 20% off takes profit per unit from 18.62 to 7.70. You would need 484 units to make the same 3,724.14 a month, and 600 to make it under the conservative rule. At a 60% margin the same discount needs only 50% more volume — 300 units — which is the real reason high-margin businesses can run promotions and low-margin ones cannot.
Same percentage, not the same dollar
The claim that raising prices beats cutting costs is true per percentage point and false per dollar, and conflating the two is how the advice gets misapplied. Here is both, at this cost structure:
| Margin | Price | Profit / unit | +10% price → profit | As % of profit | −10% cost → profit | As % of profit | Ratio |
|---|
At a 30% margin, ten percent of price is 6.21 and ten percent of cost is 3.60 — the price move is simply a bigger number, because the price is bigger than the cost. After the 12% fee takes its cut of the increase, the price still adds 5.46 against 3.60, which is 29.3% of profit versus 19.3%, a ratio of 1.52. The fee-free version of that ratio is 1 ÷ (1 − margin), which is 1.43 at 30% and 2.50 at 60%; with fees in the divisor it is (1 − fee) ÷ (1 − fee − margin), which is larger, 1.52 and 3.14, because the cost base is smaller once the fee is stripped out.
Now the other direction, and it matters: one dollar of price adds 0.88 of profit, one dollar of cost adds 1.00. Per dollar, cost cutting wins by 12%, and you would need a 1.14 price increase to match a 1.00 cost cut. So the correct statement is narrow: a percentage improvement in price moves profit further than the same percentage improvement in cost, and the gap widens as margins rise. A dollar is a dollar and cost wins that comparison every time.
Fixed cost per unit is a volume bet you cannot see
The overhead term is fixed ÷ units, so it is not a property of the product at all — it is a property of your current volume, baked silently into a price you may keep for a year:
| Units / month | Fixed cost / unit | Cost base | Price at 30% margin | Share of price that is overhead |
|---|
At 100 units the price is 82.76 and 29.0% of it is overhead; at 800 units it is 46.55 and 6.4%. Nothing about the product changed. This is the mechanism by which a business that grew gets a windfall and a business that shrank gets a slow leak: the price was set at the old volume, and it is either too high for the new cost base or too low for it, with no line item anywhere that says so.
Where this model stops working
Everything above is arithmetic on your inputs. These are the cases where the arithmetic is right and the answer is still wrong:
- Competitive pricing. In most categories the price is set by the market and the cost is what you have to fit underneath it. If comparable listings clear at 49, then 62.07 is not a price, it is a wish — and at 49 this cost structure returns 7.12 a unit, a 14.5% margin, not 30%. Run the tool backwards in that case: the output you need is the cost that makes the market price work, not the price that makes your cost work.
- Price elasticity. The units input is treated as fixed, but it is the input most likely to move. At a 30% margin, the 10% price rise that adds 5.46 a unit is still profitable if volume falls by up to 22.7% — past that the extra margin is worth less than the orders it cost. The formula cannot tell you which side of 22.7% you are on; only a price test can.
- Tiered purchase discounts. The landed cost is not constant: the per-unit price you pay usually falls as the order quantity rises, which changes the selling price through a second channel that this page holds still. Model the price at each break quantity separately, or the volume decision and the price decision will contradict each other.
- Commission tiers and category caps. A flat fee percentage is an average, and averages hide steps. Where a marketplace's referral fee changes by category, by price band or with a per-item minimum, a small price rise can cross a threshold and cost more than it earns — at these costs, a fee that steps from 12% to 15% takes the required price from 62.07 to 65.45, a 3.39 jump for a change the customer cannot see.
- Returns and refunds. A return does not just reverse a sale: the outbound shipping is gone, the item often cannot be resold at full price, and the payment fee is frequently not returned. At a 10% return rate the effective cost per unit you actually keep is 20.00 rather than 18.00, which on its own moves the target price. This page has no returns input — the return rate calculator does, and the two should be read together.
- Subscription and lifetime value. If a customer buys three times, the 6.00 acquisition cost is 2.00 per order, the cost base falls to 32.00 and the price to 55.17 — a 6.90 difference on the same product. Pricing a repeat-purchase business on first-order economics either overprices it or makes every acquisition look unaffordable.
- Tax shown in the price. In VAT and GST markets the sticker price usually includes the tax, and the platform fee is charged on the tax-inclusive amount in some jurisdictions and not others. A 20% VAT included in a 62.07 sticker leaves 51.72 of actual revenue, which is a different business from the one this formula modelled.
Frequently asked questions
Why can't I just divide my cost by one minus my margin?
Because the fee comes out of the price, not out of the cost, and that formula never names it. Dividing 36.00 by 0.70 to target 30% gives 51.43; after a 12% fee you keep 9.26, which is 18.0%. The correct divisor is 1 − 0.12 − 0.30 = 0.58, giving 62.07. Written as an identity, cost-plus pricing delivers target margin minus fee — 30 − 12 = 18 — which is exact, not approximate, and gets worse with every fee you add.
What markup do I need for a 70% margin?
233%, which prices a 36.00 unit cost at 119.88 and leaves 70.0% before fees. After a 12% fee that same price leaves 58.0%, or 69.49 a unit. To actually keep 70% after fees you need a price of 200.00, which is a 455.6% markup. The conversion is markup ÷ (1 + markup), so the numbers diverge faster the higher you go: 100% markup is 50.0% margin, 150% is 60.0%, 200% is 66.7%.
My fees are 12% and I want an 88% margin. Why is there no answer?
Because 88% is the ceiling and the ceiling is not reachable. The divisor is 1 − fee − margin, so 12% fees with an 88% margin divides by exactly 0.00, and 89% divides by −0.01. Fees are charged on the price, so they cap margin at 100 minus the fee percentage — 88% here, 70% for a business paying 30% in fees. Above that line the only fix is a lower fee, not a higher price, and no amount of pricing skill substitutes for it.
Is a 20% discount really worth 200% more volume?
At a 30% margin, yes — discount ÷ (margin − discount) is 20 ÷ 10 = 200%, or 600 units instead of 200. Profit per unit falls from 18.62 to 7.70, so you need roughly three times the orders to make the same 3,724.14 a month. Fees falling with the price soften it to 141.9%, or 484 units, at a 12% fee. At a 60% margin the same discount needs only 50% more volume, 300 units — which is the whole difference between a business that can promote and one that cannot.
Should I raise the price or cut the cost?
Per percentage point, price. At a 30% margin a 10% price rise adds 5.46 a unit against 3.60 for a 10% cost cut — 29.3% versus 19.3% of profit, a 1.52 ratio that rises to 3.14 at a 60% margin. Per dollar, cost: one dollar of price adds 0.88 after a 12% fee, one dollar of cost adds 1.00, so you need a 1.14 price rise to match a dollar of cost. Quote the percentage claim only when you mean percentage — the dollar version runs the other way.
Does fixed overhead really belong in a unit price?
It has to be recovered somewhere, and putting it in the price is what makes the price cover the month. At 2,400 over 200 units it is 12.00 a unit, 19.3% of the 62.07 price. At 400 units it is 6.00 and the price falls to 51.72 — a 10.34 drop with the same product and the same fee. If you prefer to price on contribution margin and treat overhead as a monthly hurdle, set fixed to 0 and the tool will give you the contribution price instead; the two answers differ by exactly the overhead you removed.
What if I price below cost on purpose?
The arithmetic holds, it just produces a loss. A margin of minus 20% prices at 33.33 on this cost base, losing 6.67 a unit and 1,333.33 a month at 200 units. That is a defensible customer acquisition spend if you know the lifetime value it buys, but it is a spend rather than a margin: it needs a budget, a cap and an end date, and it should not be sitting in the same column as the prices that are supposed to make money.
Last reviewed: October 2026. Assumes a flat percentage fee on price, a constant landed cost per unit, a constant ad cost per order and fixed overhead fully allocated across the units you enter. Excludes returns, tax shown in price, tiered purchase discounts, repeat-purchase economics and demand elasticity, all covered under the limitations above. Figures are arithmetic on your inputs, not a forecast.