Factorising Quadratics
The height of a kicked football, the area of a rectangular garden, the point where a business
stops losing money and starts making it — a huge range of real problems boil down to a
quadratic, and cracking one usually starts with factorising it into brackets.
Expanding brackets is easy — almost boring. Hand anyone
(x + 3)(x + 4), and as long as they follow the recipe from
multiplying polynomials,
out comes x^2 + 7x + 12, every time, no thinking required.
Factorising is the same journey run backwards — and that makes it
detective work. Someone hands you the finished product,
x^2 + 7x + 12, and asks: which two brackets were multiplied
to make this? The evidence has been scrambled together — the
7x is really two terms merged into one — and your job is to
un-scramble it:
x^2 + 7x + 12 = (\,x + \boxed{?}\,)(\,x + \boxed{?}\,)
Happily, the culprits always leave the same two fingerprints. In this example, the numbers
hiding in the brackets multiply to 12 and
add to 7 — and once you know to look for that
pair of clues, every quadratic of this kind cracks the same way.
This lesson covers the cleanest case: a quadratic
x^2 + bx + c whose x^2 has
coefficient 1. (If every term shares a number first — like
2x^2 + 14x + 24 — pull out the
common factor
before you start; here that leaves 2(x^2 + 7x + 12), and the
bracket is exactly the case we're about to crack.)
The two-number trick
Why do the clues work? Expand a general pair of brackets
(x + p)(x + q) and
collect like terms:
(x + p)(x + q) = x^2 + px + qx + pq = x^2 + (p + q)\,x + pq
Look at where p and q end up. Their
sum becomes the number in front of x, and their
product becomes the constant on the end. Expanding hides the two numbers —
but it hides them in a completely predictable place.
For any numbers p and q:
- x^2 + bx + c = (x + p)(x + q) exactly when
- p + q = b (the two numbers add to the middle coefficient), and
- p \times q = c (they multiply to the constant term).
So to factorise x^2 + bx + c, run the identity backwards:
find two numbers that multiply to c and add to
b. That single sentence is the whole method — everything
below is just practising the hunt in different disguises.
Worked example: the full hunt
Factorise x^2 + 7x + 12. We need two numbers that
multiply to 12 and
add to 7. Don't guess at random — be systematic.
The product clue is the stronger one, because 12 has only
a few factor pairs. List them all, then check each pair's sum:
\begin{array}{ccc}
\textbf{pair} & \textbf{product} & \textbf{sum} \\
1,\ 12 & 12 & 13 \\
2,\ 6 & 12 & 8 \\
3,\ 4 & 12 & \boxed{7}
\end{array}
Only 3 and 4 pass both tests, so
p = 3, q = 4:
x^2 + 7x + 12 = (x + 3)(x + 4)
Now check — always. Factorising has a built-in answer key: expand your
brackets back out. Here
(x + 3)(x + 4) = x^2 + 4x + 3x + 12 = x^2 + 7x + 12 — exactly
what we started with, so the factorisation is right. Ten seconds of expanding catches nearly
every slip, especially the sign slips coming up next. Make it a reflex.
(The order of the brackets doesn't matter, by the way:
(x + 4)(x + 3) is the same answer, just as
4 \times 3 = 3 \times 4.)
See it as an area model
Here's what factorising really does, drawn as a picture. Think of each term of
x^2 + 7x + 12 as a piece of area: the
x^2 is a square of side x; the
7x splits into two strips (a 3x strip
and a 4x strip — that split is the two-number trick, since
3 + 4 = 7); and the 12 is a
3 \times 4 block of ones.
Slide the four pieces together and they tile one perfect rectangle,
(x + 3) wide and (x + 4) tall. A
factorisation is literally a rectangle whose side lengths are the two brackets. And notice
why the corner closes up: the gap left between the strips is exactly
3 \times 4 = 12 — the product clue. If the constant term were,
say, 11, the block wouldn't fit and no rectangle (no
factorisation with whole numbers) would exist. Step through it below.
The area picture above isn't a modern teaching gimmick — it's how quadratics were
originally done. Around 820 CE in Baghdad, the mathematician
Muhammad ibn Musa al-Khwarizmi wrote a book on solving equations whose
method was entirely geometric: he drew an actual square of side x,
glued rectangular strips onto it for the x-terms, and then worked
out what block of units was needed to complete a perfect shape — squares and rectangles,
no minus signs, no symbols, just areas.
His book's title contained the word al-jabr ("restoring" — moving a term across an
equation to balance it), which became our word algebra. And his own name,
Latinised, became algorithm. One author, two of the biggest words in
mathematics — and the diagram you just stepped through is essentially page one of his book.
When c is negative: the signs hunt
Factorise x^2 + 2x - 15. Same trick: two numbers that multiply to
-15 and add to +2. But a
negative product is a big clue in itself: two numbers only multiply to a
negative when they have opposite signs — one positive, one negative. So one
bracket will hold a + and the other a -.
And when the signs are opposite, adding the numbers really means taking their
difference. So hunt through the factor pairs of 15
(ignoring signs for a moment) and look for a pair whose difference is
2:
\begin{array}{ccc}
\textbf{pair} & \textbf{product} & \textbf{difference} \\
1,\ 15 & 15 & 14 \\
3,\ 5 & 15 & \boxed{2}
\end{array}
The pair is 3 and 5. Which one goes
negative? The sum we need is +2 — positive — so the
bigger number keeps the plus: +5 and
-3 (check: 5 \times (-3) = -15,
5 + (-3) = 2). So:
x^2 + 2x - 15 = (x + 5)(x - 3)
Check by expanding: (x + 5)(x - 3) = x^2 - 3x + 5x - 15 = x^2 + 2x - 15. ✓
Had the middle term been -2x instead, the same pair would flip:
(x - 5)(x + 3) — the bigger number takes the sign of
b.
When b is negative but c is positive
Factorise x^2 - 8x + 15. Two numbers that multiply to
+15 and add to -8. Read the clues in
order. A positive product means the two numbers have the
same sign — both positive or both negative. A negative sum
then settles it: two positives can't add to a negative, so both numbers are
negative.
Now the hunt is the familiar one — factor pairs of 15 summing to
8 — with minus signs stuck on at the end:
1 + 15 = 16 (no), 3 + 5 = 8 (yes).
So the numbers are -3 and -5:
x^2 - 8x + 15 = (x - 3)(x - 5)
Check: (x - 3)(x - 5) = x^2 - 5x - 3x + 15 = x^2 - 8x + 15. ✓
Notice the two minus signs multiplied together to give the plus
15 — which is exactly why a positive c
can hide either two pluses or two minuses, and only b tells you which.
Putting the three cases side by side, the constant term is your first witness and the middle
term your second:
\begin{array}{ccl}
\textbf{c} & \textbf{b} & \textbf{signs in the brackets} \\
+ & + & (+)(+) \quad\text{e.g. } x^2 + 7x + 12 = (x+3)(x+4) \\
+ & - & (-)(-) \quad\text{e.g. } x^2 - 8x + 15 = (x-3)(x-5) \\
- & \pm & (+)(-) \quad\text{e.g. } x^2 + 2x - 15 = (x+5)(x-3)
\end{array}
Three traps catch almost everyone at first:
-
The signs are the whole game. Most factorising errors are sign errors.
Read c first: if it's positive, the brackets take the
same sign (and b tells you which one); if it's
negative, they take opposite signs (and the bigger number takes the sign
of b). A student who writes
x^2 - 8x + 15 = (x + 3)(x - 5) has matched the
15… until they expand and find -2x
in the middle. The expand-back check catches this every time.
-
Not everything factorises. Try x^2 + 4: you
need two numbers multiplying to +4 (same signs) and adding to
0 — but two same-signed numbers can never cancel to zero. A
sum of squares like x^2 + 4 simply doesn't factorise
(don't confuse it with x^2 - 4 = (x+2)(x-2), which does).
Likewise x^2 + 5x + 3 has no whole-number pair multiplying to
3 and adding to 5. If the hunt turns
up nothing, that can be the answer.
-
Factorising is not solving. Writing
x^2 + 7x + 12 = (x + 3)(x + 4) hasn't "found
x" — there's no equation here, just the same expression
re-written in a more useful shape. Solving comes next: once you set the expression
equal to 0, the factored form hands you the solutions almost for
free. But don't jump ahead and write "x = -3" under a
factorising question — no "= 0", no solutions.
A quadratic isn't just a string of symbols — its graph is a
parabola, a U-shaped curve. Ask "where does
y = x^2 + 7x + 12 cross the x-axis?"
and the expanded form keeps mum. But the factored form answers instantly: the curve touches
the axis exactly where a bracket hits zero, because a product is zero only when one of its
factors is. (x + 3)(x + 4) vanishes at
x = -3 and x = -4 — read straight off
the brackets, no plotting required.
That's why factorising earns its keep: it is the x-ray that shows a parabola's
skeleton — its roots — through the skin of the expanded form. Engineers finding where a
projectile lands, economists finding a break-even point, game programmers finding when two
objects collide: all of them are asking "where is this quadratic zero?", and the factored
form is usually the fastest honest answer.
See it explained