Factorising Harder Quadratics
Real quadratics rarely arrive tidy. Model the arc of a long throw, the profit of a business, or
the strength of a loaded beam and you often get an x^2 term with an
awkward number stuck in front — the point where the simple factorising trick stops working and you
need a sturdier method.
You already know how to
factorise a quadratic
when the x^2 coefficient is 1: hunt two
numbers that multiply to c and add to
b, and they drop straight into the brackets. For
x^2 + 7x + 10 the numbers are 2 and
5, and the answer is (x+2)(x+5). Done.
Now try the same trick on 2x^2 + 7x + 3. Two numbers that multiply to
3 and add to 7? There aren't any — and even
if there were, they couldn't just drop into brackets, because the brackets now have to start
(2x \;\; )(x \;\; ) to produce that leading
2x^2. That 2 scrambles the "inner" and
"outer" terms: the numbers in the brackets no longer simply add to give the middle. The old rule
needs an upgrade.
The upgrade is the AC method, and the beautiful thing about it is that it turns
the new, harder puzzle back into the one you can already solve. A quadratic in the form
ax^2 + bx + c
is factorised in three moves: find two numbers that multiply to
a \cdot c and add to b, use them to
split the middle term into two pieces, then factor by grouping.
Same hunt as before — just with a \cdot c as the target product
instead of plain c.
The two numbers
For 2x^2 + 7x + 3 we have a = 2,
b = 7 and c = 3, so
a \cdot c = 2 \times 3 = 6. We need two numbers that multiply to
6 and add to 7. Be systematic: list the
factor pairs of 6 and check each sum.
1 \times 6 = 6, \quad 1 + 6 = 7 \;\checkmark \qquad\qquad 2 \times 3 = 6, \quad 2 + 3 = 5 \;✗
The pair is 1 and 6. Listing pairs in order
like this beats staring and hoping: for a product like 6 there are only
two pairs to try, and even for a \cdot c = 36 there are just five. One
of them either adds to b or none does — and either way you know within
a minute.
(Notice that when a = 1 this is exactly the rule you already use —
multiply to c, add to b — because
a \cdot c = c. The AC method isn't a second method to memorise; it's
the same method, told in full.)
Splitting and grouping
Now spend the pair. Rewrite the middle term 7x as
1x + 6x — that changes nothing, since
1x + 6x = 7x — but it leaves four terms instead of three.
Four terms can be grouped into two pairs, and a common factor pulled out of each pair. Here is
the magic the AC pair guarantees: after factoring each pair, the same bracket appears in
both. That shared bracket is itself a common factor of the whole expression, so it comes
out the front — and the factorisation falls into your lap. Step through it below.
The working, line by line:
2x^2 + 7x + 3 = 2x^2 + 1x + 6x + 3
= (2x^2 + x) + (6x + 3) = x(2x + 1) + 3(2x + 1)
= (2x + 1)(x + 3)
A detail worth knowing: the order of the split doesn't matter. Write
7x = 6x + 1x instead and the grouping still works — the pairs are
different, but the same bracket still pops out:
2x^2 + 6x + x + 3 = 2x(x+3) + 1(x+3) = (x+3)(2x+1). Same answer,
different route. If your grouping ever fails to produce a shared bracket, you haven't
broken mathematics — you've mis-picked the pair. Go back and re-check the multiply-and-add.
Make checking a habit
Factorising has a superpower most topics lack: you can always mark your own work.
Factorising and
expanding
are inverse operations, so multiplying your answer back out must recreate the original — every
term, every sign. For our example:
(2x + 1)(x + 3) = 2x^2 + 6x + x + 3 = 2x^2 + 7x + 3 \;\checkmark
Ten seconds, and the check either passes or catches you. Suppose you'd rushed and written
(2x + 3)(x + 1) — a very tempting wrong answer, since it uses all the
right numbers. Expanding exposes it instantly:
(2x + 3)(x + 1) = 2x^2 + 2x + 3x + 3 = 2x^2 + 5x + 3 \;✗
The middle term is 5x, not 7x — wrong
answer, caught before it cost any marks. Do this check every single time. In an exam it
converts "I think that's right" into "I know that's right."
When c is negative: 3x^2 - 5x - 2
Negative coefficients change nothing about the method — but they demand care with signs. Here
a = 3, b = -5,
c = -2, so a \cdot c = 3 \times (-2) = -6.
We need two numbers that multiply to -6 and add to
-5. A negative product means the pair has opposite signs —
one positive, one negative — and a negative sum means the negative one is bigger. List the
candidates:
1 + (-6) = -5 \;\checkmark \qquad (-1) + 6 = 5 \;✗ \qquad 2 + (-3) = -1 \;✗ \qquad (-2) + 3 = 1 \;✗
The pair is 1 and -6. Split the middle term
-5x as x - 6x and group:
3x^2 - 5x - 2 = 3x^2 + x - 6x - 2
= x(3x + 1) - 2(3x + 1)
= (3x + 1)(x - 2)
Watch the second group closely: to make the bracket come out as (3x+1)
— matching the first — we pull out -2, not 2,
because -6x - 2 = -2(3x + 1). Pulling out a negative flips the signs
inside the bracket; that flip is exactly what makes the brackets match. And of course, the habit:
(3x+1)(x-2) = 3x^2 - 6x + x - 2 = 3x^2 - 5x - 2. ✓
Step zero: hunt a common factor first
Before any of this cleverness, always ask the simplest question first: do all the terms
share a common factor?
Sometimes that's the whole job. Take 4x^2 + 10x: there's no constant
term at all, and every term is divisible by 2x, so
4x^2 + 10x = 2x(2x + 5)
— fully factorised, no AC method required. Reaching for the fancy tool here would be like using a
crane to lift a teacup.
Even when the AC method is needed, a common factor taken out first makes everything that
follows easier. Look at 2x^2 + 8x + 6: every term is even, so take out
the 2 —
2x^2 + 8x + 6 = 2(x^2 + 4x + 3) = 2(x + 1)(x + 3)
Once the 2 is out, the quadratic left inside has
a = 1, and you're back on the easy road: two numbers that multiply to
3 and add to 4 — that's
1 and 3 — done. The common factor didn't
just save effort; it downgraded the problem to one you mastered ages ago. So the full
recipe is: common factor first, then AC, then check by expanding.
-
Missing the common factor. Attack 2x^2 + 8x + 6 with
the AC method directly (a \cdot c = 12, pair
2 and 6) and you'll grind out
(2x + 2)(x + 3). Expand it — it's correct! But it's not
fully factorised: (2x+2) still hides a factor of
2, and "factorise fully" questions dock the mark. The answer they
want is 2(x+1)(x+3). Common factor first — always.
-
Getting the pair's signs wrong. The pair must multiply to
a \cdot c including its sign, and add to
b including its sign. For
3x^2 - 5x - 2 the pair is +1 and
-6; pick -1 and +6
instead (same product!) and the sum comes out +5, the grouping falls
apart, and nothing matches. Rule of thumb: negative a \cdot c means
opposite signs, negative b means the bigger number is the negative one.
-
Assuming it must factorise. Not every quadratic factorises over the integers.
Try 2x^2 + 3x + 4: a \cdot c = 8, and no
pair of factors of 8 adds to 3 — because
there is no integer factorisation to find. There's a forward-looking test: the quantity
b^2 - 4ac (here 9 - 32 = -23) must be a
perfect square for whole-number brackets to exist. You'll meet it properly, as the
discriminant, with the
quadratic formula.
For now: if every pair fails, suspect the quadratic, not yourself.
Here's the payoff that makes all this hunting worthwhile. A product is zero exactly when one of
its factors is zero — so the moment you write
2x^2 + 7x + 3 = (2x + 1)(x + 3), the equation
(2x+1)(x+3) = 0 hands you its solutions at a glance:
2x + 1 = 0 gives x = -\tfrac{1}{2}, and
x + 3 = 0 gives x = -3. No formula, no
rearranging — the roots were sitting inside the brackets all along. That's also exactly where the
parabola
crosses the x-axis, which is why factorised form is the form engineers,
economists and game programmers actually want: it shows where things hit zero.
And a bit of classroom anthropology: maths teachers worldwide wage a gentle, decades-long war over
how to factorise these. One camp teaches guess-and-check — write
(2x \;\; )(x \;\; ) and audition numbers until the middle term works.
The other camp teaches the AC method you've just learned. The truth? They always
reach the same brackets. Guess-and-check is faster once you're fluent; the AC method is a
machine that never leaves you stuck, even on ugly coefficients. Learn AC first, and let
guess-and-check grow out of it — after enough splits, you'll start seeing the pair before you've
written anything down. That's not cheating; that's fluency.
See it explained
Another voice, same method: Khan Academy works through factoring a harder quadratic by grouping.
Watch for the two moves you now know by name — the multiply-to-ac,
add-to-b hunt, and the shared bracket appearing after grouping.