In short
A percentage is a fraction whose denominator has been agreed in advance to be 100 — so 30\% means \dfrac{30}{100}, no more and no less. A ratio is the same idea written as a comparison: a : b means "for every a of the first there are b of the second." Both notations describe proportional relationships, and you can move freely between fractions, decimals, percentages, and ratios. One technique — the unitary method, "first find what one unit is worth, then scale up" — solves nearly every shop discount, GST, profit-and-loss, recipe-scaling, and ratio-sharing problem you will ever meet.
A shop sign in Sarojini Nagar says "30% OFF". The kurta is marked at ₹800. What do you actually pay?
You probably already know the answer — ₹560 — but think about what just happened in your head. You took a number (₹800), found a piece of it (30% of it = ₹240), and subtracted that piece from the whole. The whole calculation rests on a single idea: the symbol "30\%" is shorthand for the fraction \tfrac{30}{100}, and "30% of ₹800" is the multiplication \tfrac{30}{100} \times 800 = 240. Once you see the hidden denominator, every percentage problem becomes a fraction problem — and every fraction problem becomes one of the patterns from Fractions and Decimals.
This article is the operating manual for the three notations that describe the same kind of comparison — fractions, percentages, ratios — and the one universal technique (the unitary method) that handles every proportional problem in school arithmetic.
What a percentage actually is
The phrase per cent comes from a Latin word meaning per hundred. A percentage is a fraction whose denominator has been agreed in advance to be 100. So:
That is the entire definition. Everything else about percentages — converting them, comparing them, calculating with them — follows from this one fact.
The reason the notation exists at all is that it makes comparison cheap. If one bank pays 6.5\% interest and another pays 7\%, you can compare them at a glance because the denominator is already the same. Without percentages, you would have to find a common denominator first — the work that the convention has done in advance, once and for all.
There are three conversions worth memorising, because they appear in every real-world calculation:
- Percentage to fraction. Divide by 100, then (optionally) reduce. 30\% = \tfrac{30}{100} = \tfrac{3}{10}.
- Percentage to decimal. Divide by 100, which moves the decimal point two places to the left. 30\% = 0.30.
- Fraction to percentage. Multiply by 100, with \% as the unit. \tfrac{1}{4} \times 100 = 25\%.
Computing a percentage of a number
The basic question is: what is X\% of Y? The answer is
For "30\% of 800":
So 30\% of 800 is 240, and the kurta you pay for is 800 - 240 = 560 rupees.
There is also a useful mental shortcut: 10\% of any number is just that number with the decimal point shifted one place to the left. 10\% of 800 is 80. So 30\% is three of those, or 240. 5\% is half of 10\%, or 40. Almost every quick percentage calculation in your head is a combination of "10\% of …" and "1\% of …", added or subtracted.
You can feel the relationship between the percentage and the answer directly. The figure below shows a bar of length \textsf{₹}1000. Drag the red marker. As you move it, the shaded portion shows the corresponding percentage of the bar, and the readouts show both the percentage and the rupee amount. This is the picture every "percentage of a number" question is secretly asking.
The reverse question — X is what percentage of Y? — uses the same formula, solved for X. If a discount of ₹240 has been taken from a price of ₹800, the discount as a percentage is \tfrac{240}{800} \times 100 = 30\%.
Ratios
A ratio is a comparison written as a : b, read as "a to b." If a class has 12 boys and 18 girls, the ratio of boys to girls is 12 : 18. Like a fraction, a ratio can be reduced by dividing both sides by their highest common factor — here both are divisible by 6, so the simplified ratio is 2 : 3.
The way to read a : b is: for every a of the first kind, there are b of the second. The total is a + b "parts," and each side is its share of those parts. So in a 2 : 3 ratio, the first quantity is \tfrac{2}{5} of the total and the second is \tfrac{3}{5} of the total. The denominator is the sum of the two ratio terms, not either one of them on its own — and forgetting this is the most common ratio mistake.
A ratio with three quantities works the same way: a : b : c has a + b + c parts, and each side gets its own fractional share.
Proportion and the unitary method
When two ratios are equal, the equation is called a proportion:
Cross-multiplying gives the equivalent form ad = bc — the product of the "outer" pair equals the product of the "inner" pair. This is the rule that lets you solve "if a of these costs b rupees, how much do c of them cost?" by setting up a proportion and solving for the unknown.
But the cleanest, most general way to handle proportional problems is the unitary method, which has just one rule: first find what one unit is worth, then scale up to whatever quantity you need.
Take a concrete example. Five mangoes cost ₹120. How much do eight mangoes cost?
Find one. If five mangoes cost ₹120, then one mango costs \textsf{₹}120 \div 5 = \textsf{₹}24.
Scale up. Eight mangoes cost 8 \times \textsf{₹}24 = \textsf{₹}192.
That is the entire technique. You went from "five for ₹120" to "one for ₹24" to "eight for ₹192" with no algebra. The unitary method works for any proportional relationship — speed and time, recipe ingredients, exchange rates, work and labour, fuel and distance — as long as the rule "twice as much input gives twice as much output" holds. Almost all real-world percentage and ratio problems are unitary-method problems in disguise.
Two worked examples
The two examples below cover the most common real-world patterns: a percentage chain (discount followed by tax) and a ratio split.
Example 1: A kurta marked at ₹1500 has a 20% discount, after which 18% GST is added. What do you pay?
This is a chain of two percentages, and the order matters: the discount comes off the marked price, and the GST is then added on top of the discounted price (not on top of the original).
Step 1. Apply the 20\% discount.
The discount amount is \tfrac{20}{100} \times 1500 = 300. So the discounted price is
Why: 20\% of 1500 is 300, found by the formula \tfrac{X}{100} \times Y. You then subtract the discount because the customer pays less.
Step 2. Apply 18\% GST on the discounted price.
The GST amount is \tfrac{18}{100} \times 1200 = 216. So the final price is
Why: GST is computed on the price you actually pay, after any discount — that is the standard tax convention. So the base for the tax is the discounted price ₹1200, not the original ₹1500. Then you add the tax because the customer pays more.
A faster way using multipliers. Each percentage step can be written as a multiplication. A 20\% discount multiplies the price by 0.80 (since 100\% - 20\% = 80\%), and an 18\% tax multiplies by 1.18. So the whole chain is a single product:
The order of the two multiplications doesn't matter (multiplication is commutative — see Operations and Properties), so you would get the same answer if the GST were applied first and the discount second. But the rupee amount of each step would be different, even though the final total is the same.
Result. You pay ₹1416.
Example 2: Three friends share ₹4500 in the ratio 2 : 3 : 4. How much does each get?
This is a clean ratio-splitting problem and the unitary method handles it directly. The two-line plan: find the total number of parts, then find the value of one part, then scale up to each share.
Step 1. Add the ratio terms to find the total number of parts.
So the ₹4500 is being split into 9 equal parts.
Step 2. Find the value of one part.
Why: this is the unitary step. The total amount divided by the total number of parts gives the value of a single part, and from there you can multiply up to any share.
Step 3. Multiply to find each friend's share.
Step 4. Check: the three shares should add up to the original total.
Result. The three friends get ₹1000, ₹1500, and ₹2000 respectively.
Common confusions
-
"A 20% discount followed by a 20% tax brings me back to the original price." Not even close. Take ₹100. After a 20\% discount, you have ₹80. A 20\% tax on ₹80 adds ₹16, giving ₹96 — not ₹100. The discount and the tax act on different bases (the original vs. the discounted price), and the percentages do not cancel because they are not measuring the same thing.
-
"If apples are 50% more expensive than bananas, then bananas are 50% cheaper than apples." No. Take bananas at ₹10. Apples are 50\% more expensive: 10 + (50\%\text{ of }10) = 15. Now go the other way: bananas are how much cheaper than apples? The reduction is ₹5 on a base of ₹15, which is \tfrac{5}{15} = 33\%, not 50\%. The percentage depends on the base, and "X more than Y" and "Y less than X" use different bases, so they almost never give the same percentage.
-
"In the ratio 2 : 3, one quantity is \tfrac{2}{3} of the total." No — it is \tfrac{2}{3} of the other quantity, but \tfrac{2}{5} of the total. The denominator for the share is the sum of the ratio terms, not either term on its own. This is the most common ratio mistake on exams.
-
"Percentages can never go above 100\%." They can, and they often do. Stock prices, growth rates, profit margins, returns on investment — all routinely exceed 100\%. A profit margin of 150\% just means the profit is 1.5 times the cost.
-
"20\% off then 30\% off is the same as 50\% off." No — it is 44\% off. Take ₹100. After 20\% off you have ₹80. After another 30\% off you have 0.70 \times 80 = 56. The total discount is 44\%, not 50\%, because the second discount applies to a smaller base. Successive percentages multiply, they don't add.
Going deeper
If you came here to handle the everyday percentage and ratio calculations of school and shopping, you have it. The rest of this section is for readers who want to see how the same patterns power some larger ideas — compound growth, comparative pricing, and financial returns.
Multipliers and compound growth
Every percentage operation can be rewritten as a single multiplication, and that rewriting is the key to handling repeated percentage changes cleanly.
- A p\% increase multiplies by \left(1 + \tfrac{p}{100}\right).
- A p\% decrease multiplies by \left(1 - \tfrac{p}{100}\right).
- An n-step chain multiplies by the product of all the multipliers.
So a 5\% annual interest rate compounded for 10 years multiplies your money by 1.05^{10} \approx 1.629 — your principal grows by about 63\% over the decade, not 50\%. The difference is the interest on the interest, and it is exactly the difference between adding percentages and multiplying multipliers.
The exponent in 1.05^{10} is the topic of Exponents and Powers, and the way that exponent shapes the long-run behaviour of compounding is one of the most important consequences of the laws of exponents.
Why "average percentage" is almost always wrong
If a stock goes up 50\% one year and down 50\% the next, what is your average return? Naively you might say "zero" — but the correct answer is a loss. Start with ₹100. Up 50\% takes you to ₹150. Down 50\% takes you to ₹75. You lost a quarter of your money, despite the two percentages "averaging out."
The reason is that the two percentages are computed on different bases. The right average for compounded multiplicative changes is not the arithmetic mean (the sum-divided-by-count) but the geometric mean (the product-to-the-one-over-count). For the example, the geometric mean of 1.5 and 0.5 is \sqrt{1.5 \times 0.5} = \sqrt{0.75} \approx 0.866, which corresponds to an average loss of about 13\% per year — and indeed, 0.866^2 \approx 0.75, recovering the actual outcome.
This is why finance, demography, and any field that deals with compounded change quotes "geometric returns" instead of arithmetic averages. It is also why the shopkeeper's "20% off and 20% off" isn't the same as "40% off" — the two percentages are operating on changing bases, and the right combination is multiplicative.
Ratios as fractions in disguise
Every ratio can be rewritten as a fraction, and vice versa. The ratio a : b corresponds to the fraction \tfrac{a}{a + b} for the share of the first part out of the total. The proportion equation \tfrac{a}{b} = \tfrac{c}{d} is just two fractions being equal — which is the same as the equation you would set up in Operations and Properties, with cross-multiplication being the do-the-same-thing-to-both-sides rule applied twice.
So when you learn ratios, you are learning fractions in a different costume. And when you learn the unitary method, you are learning a particular way to solve linear equations of the form \tfrac{x}{a} = \tfrac{y}{b} — which is the simplest possible algebraic equation. Almost every "real-world" arithmetic problem reduces to the same template.
Where this leads next
Percentages and ratios feed directly into the financial and quantitative chapters that follow.
- Fractions and Decimals — the foundation that everything in this article rests on, with the conversion rules between fractions and decimals.
- Exponents and Powers — the next chapter, including the law a^m \cdot a^n = a^{m+n} that powers compound interest formulas like \,P(1 + r)^n.
- Operations and Properties — the algebraic rules behind cross-multiplication and the unitary method.
- Roots and Radicals — the next chapter after exponents, where the geometric mean from the going-deeper section above lives.
- Number Systems — for the picture of where rationals (which are what fractions, percentages, and ratios all secretly are) sit in the number line.