Practice that explains the "why" — free
Our practice portal pairs questions with worked reasoning and visual models, so students build understanding alongside fluency rather than collecting hollow rules.
Open the practice portal →A child sails through early maths by learning the rules off by heart — and then, somewhere around algebra, the wheels come off. It looks like a sudden decline. It's usually a delayed bill. This is the research on conceptual understanding versus rote procedure: why memorised rules are brittle, how meaning and fluency are supposed to grow together, and how to teach for understanding without abandoning the basics.
For years, the strategy worked beautifully. Keep your head down, learn the rule the teacher gives you — "to divide fractions, flip and multiply"; "to solve, do the same to both sides" — apply it to the worksheet, collect your ticks. Our student, let's call him Adam, was "good at maths" all through primary school on exactly this method. Then came algebra, then functions, then a GCSE paper that asked him to do something he'd never seen drilled — and the whole structure came down at once. Adam hadn't got worse. The bill for years of memorising-without-meaning had simply come due.
Adam's story is one of the most common in secondary maths, and one of the most misdiagnosed. Parents see a sudden slump and assume the new content is "too hard" or that motivation has dropped. The real cause is usually older and quieter: a tower of remembered rules with no foundation of understanding underneath — fine while the questions stay familiar, fatal the moment they don't.
The pattern has a signature. Strong early results, then a sharp decline around the point where maths turns abstract (typically algebra and beyond). The child can do textbook exercises that match a taught example, but freezes on anything phrased unusually. They can't explain why a method works. They don't notice when an answer is obviously wrong. And re-teaching the new topic barely helps, because the missing piece isn't the new topic — it's the understanding the new topic assumed was already there.
To see why rote knowledge fails, picture two ways of knowing a city.
One person has memorised a single route from the station to the museum: turn left, third right, past the bakery. They get there perfectly — until a road is closed, and then they're completely lost, because they never understood how the city fits together. Another person has a map in their head. They can take detours, find new places, spot when they're heading the wrong way. Rote maths is the memorised route. Conceptual understanding is the map. Both reach the museum on a normal day — but only one survives a closed road, and secondary maths is nothing but closed roads.
That's the brittleness in a nutshell. A memorised rule can't transfer to an unfamiliar problem, can't be adapted when conditions change, and — crucially — gives the child no way to detect their own errors, because there's no underlying sense of what the answer should look like. Worse, disconnected rules pile up as separate items in memory, overloading the very working memory that understanding would have freed by tying everything into one connected picture.
Researchers separate two strands of mathematical knowledge. Procedural knowledge is knowing how — the steps, the method, the algorithm. Conceptual understanding is knowing why — what the ideas mean and how they connect. You can have either without the other: a child who can execute long division flawlessly but has no idea what division is has procedure without concept; a child who grasps that division is sharing but can't reliably carry out the method has concept without procedure.
The old debate framed these as rivals — "drill the procedures" versus "teach for understanding" — and asked which should come first. The modern research gives a more interesting answer: that's the wrong question.
In a landmark study, Rittle-Johnson, Siegler and Alibali (2001) followed children learning decimals and found that conceptual and procedural knowledge develop in an iterative loop: gains in understanding drove gains in procedural skill, and gains in skill fed back into deeper understanding. Neither simply comes "first". They build each other, like two legs walking. This is why pitting fluency against understanding is a false choice — starve either leg and the child limps.
The National Council of Teachers of Mathematics puts it plainly: conceptual understanding should "precede and coincide with" instruction on procedures, and effective teaching builds fluency on a foundation of understanding. The point isn't to drop procedures — they're essential — but to make sure they're anchored in meaning, so they bend instead of break when problems get unfamiliar.
The rote child writes ½ ÷ ¼ = ½ × 4/1 = 2, because "flip the second one and multiply". Correct — but if you ask "why?" or change it slightly, they're stranded. They also can't tell that the answer ought to be bigger than 1.
The child who understands reads ½ ÷ ¼ as "how many quarters fit into a half?" — pictures it — and sees the answer is 2 before doing any flipping. The procedure still helps them go fast, but it's now backed by a meaning that catches errors and survives new variations. Same method, completely different sturdiness.
Rote learning doesn't just leave gaps; it actively breeds misconceptions — wrong rules that happen to work for the simple cases a child first meets, then mislead them ever after. These are some of the most studied phenomena in maths education, because they're so predictable and so sticky.
"Longer decimal = bigger number." So 0.45 > 0.5, because 45 > 5. The rule borrowed from whole numbers seems to work… until it doesn't.
A child who pictures 0.5 as five tenths and 0.45 as forty-five hundredths can see that 0.5 is larger. The misconception dies because the meaning makes the right answer obvious.
"Multiplication always makes things bigger." True for whole numbers, so it gets memorised as a law — then × by ½ "breaks" it and the child is baffled.
Understanding multiplication as "groups of" (or scaling) makes "half of 8 is 4" feel natural. Multiplying by less than one should shrink things — no contradiction, no panic.
Notice the common thread: each misconception is a rule that overran its territory. Rote learning, by handing children rules without the meaning that marks their boundaries, practically manufactures these traps. Teaching for understanding is, in large part, the work of preventing and unpicking them.
If understanding matters so much, how is it actually built? The most influential answer comes from psychologist Jerome Bruner, who proposed that we grasp new ideas through three modes: enactive (doing, with physical objects), iconic (seeing, with pictures), and symbolic (abstract symbols). Maths teaching built on this idea is known as Concrete–Pictorial–Abstract, or CPA, and it's the engine of Singapore-style mastery.
The sequence is simple but powerful. First a child meets, say, fractions through physical objects — folding paper, sharing counters. Then through pictures — bar models, diagrams. Only then through symbols — the notation ½, ¼, the rules. Because each stage grows out of the last, the abstract symbols arrive already loaded with meaning, rather than as arbitrary marks to be memorised. The "flip and multiply" rule lands on a child who has already seen why it's true.
Singapore is the standard-bearer for building understanding and fluency together via CPA and the bar model. Concepts are introduced concretely, deepened pictorially, then formalised — and only then drilled. The result is a system famous for both deep understanding and strong procedural skill, not one at the expense of the other.
Japanese primary maths often spends a whole lesson on a single rich problem, with children inventing and comparing methods before the teacher formalises the idea. It's understanding-first by design — closely related to the "productive struggle" tradition — and it produces flexible, transferable knowledge.
High-performing Chinese classrooms use "teaching with variation" — systematically varying problems to expose the underlying structure and head off misconceptions. It looks procedural on the surface but is carefully engineered to build conceptual understanding through well-chosen examples.
The US has swung between extremes in its "math wars" — heavy procedure, then reform-era discovery, then back. The lesson many drew is that either pole alone fails: pure rote is brittle, pure discovery overloads novices. The settled middle is understanding and fluency, deliberately built together.
England's "maths mastery" reforms imported the Singapore balance: small steps, shared concrete and pictorial representations, and not racing to the abstract. It's an explicit national move away from the rote-rules tower toward fluency grounded in meaning.
Refuse the false choice. The research is clear that understanding and fluency are partners, not rivals — so build both, iteratively. Introduce a concept with meaning, practise the procedure to fluency, then return to deepen the concept. Each pass strengthens the other.
Teach through representations. Concrete materials and pictorial models (bar models, number lines, area diagrams) aren't just for strugglers or young children — they carry the meaning that makes abstract notation stick at every level. Fade them as understanding secures, exactly as with worked examples.
Hunt misconceptions on purpose. Anticipate the predictable wrong rules ("longer decimal is bigger", "multiplying makes bigger"), surface them deliberately, and confront them with a representation that makes the truth visible. Unaddressed, they don't fade — they go underground and resurface years later.
Our practice portal pairs questions with worked reasoning and visual models, so students build understanding alongside fluency rather than collecting hollow rules.
Open the practice portal →"As long as they get the right answers, understanding doesn't matter."
Right answers from rote rules are brittle — they don't transfer, adapt or self-correct. Understanding is what makes the right answers survive harder maths.
"Teaching for understanding means dropping times tables and procedures."
Not at all. Fluency and understanding grow together. The goal is procedures anchored in meaning, not procedures abandoned.
"Manipulatives and pictures are just for little kids who can't cope."
Representations carry meaning at every level — they're how abstract notation gets understood, not a crutch for the struggling.
Procedural knowledge is knowing the steps — how to carry out a method. Conceptual understanding is knowing why it works and how ideas connect. A child can have one without the other, but maths only becomes robust and transferable when the two grow together, each strengthening the other.
No — fluent facts and procedures are essential, and we make the case for them in our fluency article. The problem is memorising instead of understanding. Rules without meaning are brittle: they don't transfer, and the child can't tell when they've misapplied them. Memory and meaning should be partners, not substitutes.
Usually because early success rested on memorised rules that worked for simple cases but don't generalise. When maths becomes abstract and multi-step, those rules collapse, with no understanding to fall back on or adapt. It reads as a sudden decline; it's really an old gap surfacing. Rebuilding the concepts typically restores progress.
Drawing on Jerome Bruner's work, CPA teaches a concept first with physical objects, then with pictures, then with symbols — each stage scaffolding the next. The abstract notation arrives carrying real meaning rather than as a rule to memorise. It's central to Singapore-style mastery and works at every age, not just the early years.
Ask them to explain why a method works, to try a slightly unusual version, or to estimate whether an answer is reasonable. A child who only memorised handles the standard exercise but stumbles on the "why", on variations, and on judging whether an answer makes sense. Those three probes reveal the difference quickly.
The free assessment looks underneath the visible topic for the understanding gaps that cause secondary slumps — then shows you exactly what to rebuild, and in what order.
Book the free assessment