The Percentage Mistakes That Cost Real Money
These are not random errors that cancel out. Every one of them is biased against the person doing the arithmetic.
Almost everyone believes that a 50% discount followed by an extra 20% off is 70% off. It is 60% off. A $100 item goes to $50, then to $40, and the second discount applies to the already-reduced price rather than the original one. The seven-dollar version of this mistake is harmless. The same arithmetic error, applied to a pricing sheet, a portfolio, or a loan comparison, is where percentages stop being a school subject and start costing money.
Percentages fail people in a small number of specific, repeatable ways. Each one has a direction — these are not random errors that cancel out over time. They are biased, and in every case below the bias runs against the person doing the arithmetic.
Stacked discounts multiply, they do not add
Successive percentage reductions compose multiplicatively. Twenty percent off then ten percent off leaves you paying 0.80 × 0.90 = 72% of the original, a 28% discount, not 30%. The gap widens as the discounts get bigger, because each one is applied to a smaller base.
A real example: a $129.99 jacket advertised at 40% off, with a coupon for an extra 15% at checkout. Reading that as 55% off predicts $58.50. The actual price is $129.99 × 0.60 × 0.85 = $66.29 — $7.79 more than expected, which is roughly the amount that makes a purchase feel worthwhile or not.
The Federal Trade Commission's Guides Against Deceptive Pricing (16 CFR Part 233) governs the related question of whether the advertised "former price" was ever real, requiring that it be "the actual, bona fide price at which the article was offered to the public on a regular basis for a reasonably substantial period of time." The guides also address "50% Off" and similar claims directly, noting that such offers become deceptive when the seller quietly raises the base price or reduces quantity. Worth knowing: the rules constrain the honesty of the reference price, but nothing stops a retailer from stacking discounts in a way you will mentally add rather than multiply.
The reliable fix is to skip the discount arithmetic entirely and compute the final price directly — multiply the original by the remaining fraction at each step in the Percentage Calculator and compare that number against what you were willing to pay.
Percentage points and percent are different units
When a rate moves from 5% to 7%, it has risen by two percentage points and by 40 percent. Both statements are true; they measure different things, and conflating them is the single most exploitable ambiguity in financial writing.
The Bureau of Labor Statistics deals with the same problem in its Consumer Price Index documentation, distinguishing between changes in index points and percent changes, and noting that "index points are affected by the level of the index in relation to its reference period, while percent changes are not." The BLS documentation includes a worked table showing two items with different index-point changes — 9.0 points and 18.0 points — that nevertheless represent the identical 8% price increase, purely because they started from different index levels.
Where this costs money: a fund whose expense ratio moves from 0.35% to 0.70% may be described as increasing "by 0.35%," which sounds negligible. The fee doubled. Over a multi-decade holding period that is a substantial share of your returns, and the phrasing was chosen to make it sound like nothing. Whenever you see a small percentage describing a change to another percentage, stop and ask whether it is a point change or a relative change. The answer frequently reverses your reaction.
Losses and gains are not symmetric
Lose 50% and you need a 100% gain to get back to even. This is not a trick; it is unavoidable arithmetic, because the gain is calculated on the smaller post-loss base. The asymmetry accelerates:
| Loss | Gain required to recover |
|---|---|
| 10% | 11.1% |
| 20% | 25.0% |
| 30% | 42.9% |
| 50% | 100.0% |
| 60% | 150.0% |
| 75% | 300.0% |
| 90% | 900.0% |
Small losses are nearly symmetric — 10% down needs 11.1% up, which is close enough that intuition survives. The curve bends hard past about 30%. That bend is the actual argument for caring about drawdowns more than about upside capture: avoiding a 50% loss is worth more than catching a 50% gain, and the two are not equally likely to reverse.
The same asymmetry corrupts a statistic you see constantly. A portfolio that falls 50% and then rises 100% has an arithmetic average annual return of +25% and an actual return of exactly zero. Any "average annual return" quoted as a simple mean of yearly figures overstates what an investor actually earned, and the overstatement grows with volatility. The honest measure is the compound annual growth rate, which is what a compound growth calculation gives you.
Markup and margin are not the same number, and the confusion only ever costs you
This one is specific to anyone who sets prices, and it is the most expensive item on the list.
Markup is calculated on cost. Margin is calculated on price. An item that costs $60 and is marked up 30% sells for $78, and the gross margin on that sale is ($78 − $60) / $78 = 23.08%, not 30%. To actually earn a 30% margin on a $60 cost, the price has to be $60 / 0.70 = $85.71, which is a markup of 42.86%.
IRS Publication 334, the small business tax guide, frames the underlying quantity the way an accountant does: gross receipts minus cost of goods sold equals gross profit. Margin is that gross profit expressed as a share of receipts. Markup is a pricing instruction expressed as a share of cost. They answer different questions and coincide only at zero.
Now the part that makes this worth an article rather than a footnote. The error is directional. Applying a margin target as a markup always produces a price that is too low, never too high, because cost is always smaller than price and so the same percentage yields a smaller dollar addition. A small shop doing $400,000 in annual revenue that believes it is running a 30% margin while actually running 23.08% is short by roughly $27,680 a year in gross profit — on paper it is profitable, in the bank account it is not, and the discrepancy is invisible in the pricing spreadsheet because every individual price looks correct.
Tips, taxes, and percentages of percentages
A $86.40 restaurant subtotal with 8.75% sales tax comes to $93.96. Tipping 20% of the subtotal is $17.28. Tipping 20% of the post-tax total is $18.79. The $1.51 difference is trivial per meal and about $151 across a hundred meals in a year — not a scandal, but worth knowing that most card readers compute the suggested tip on the total including tax, which is a percentage applied to a number that already contains a percentage.
Custom in the United States is to tip on the pre-tax subtotal, since the tax is not part of the service. The Tip Calculator lets you set the base explicitly rather than accepting whatever the terminal decided. Separately, the IRS treats tips as wages for reporting purposes, and large food and beverage establishments must allocate tips among employees when reported tips fall below 8% of gross receipts — which is why the tip line is not merely a courtesy from the establishment's point of view.
One last one: a raise and an equal cut do not cancel
A 10% increase followed by a 10% decrease does not return you to the starting point. $100 becomes $110, then $99. Reverse the order and you get the same $99, because multiplication commutes — but the base changed between the two operations, and that is what breaks the intuition. A store that cuts a price 20% for a sale and then "restores" it by adding 20% back has quietly reduced the price to $96 on a $100 item.
The unifying principle behind every mistake above: a percentage is meaningless without knowing what it is a percentage of, and the base almost always changes between one step and the next. When a calculation involves two or more percentages in sequence, convert to dollars at each step and check the result. It takes ten seconds and it catches all of these.
Related tools
- Percentage Calculator — percent of a number, percent change, and reverse percentages
- Tip Calculator — set your own base and split the bill
- Compound Interest Calculator — see how repeated percentage changes actually compose over time
Sources and further reading
Figures and definitions on this page are drawn from the following primary sources. If you find something out of date, tell us and we will correct it.