Most children aren't weak at the subject. They're stuck at one specific thing nobody has found yet.
brainflex runs a five-stage teaching cycle designed to find that one thing, teach directly to it, and then check that it actually closed.
Assess → Diagnose → Target → Practise → Review
It's a cycle, not a checklist. Review feeds straight back into the next assessment.
Stage 01
Assess
"Before we teach, we first understand."
A teacher works through problems with your child and watches how they think — where they hesitate, what they skip, what they guess.
Stage 02
Diagnose
"We don't just look at the wrong answer, we look for the reason behind it."
Two children can get the same question wrong for completely different reasons. The reason is what we teach to.
Stage 03
Target
"We focus on the gap, not just the syllabus."
Lessons start at the specific thing that broke, rather than re-running a chapter your child has already sat through twice.
Stage 04
Practise
"We don't stop when your child understands the explanation, we practise until they can apply it."
Understanding in the moment is not the same as being able to do it alone, under time, in an exam.
Stage 05
Review
"We check whether your child has actually improved, not just whether the lesson was completed."
If the gap hasn't closed, the cycle runs again on the same gap. Completion is not the measure — improvement is.
What the cycle looks like on one P6 student and successive percentage change
A composite walkthrough written to show how the method runs in practice on a real P6 syllabus skill. It is not a case study of an identifiable child.
Assess
The pattern wasn't 'bad at percentage'
A P6 student comes to us because percentage questions keep coming back marked wrong. In the assessment, the teacher gives him a mix. Single-step questions — find 15% of $80, express 12 out of 50 as a percentage, increase $60 by 20% — he gets right, quickly and confidently. The errors cluster in one place only: multi-step questions involving successive percentage change. Not percentage generally. Successive percentage change specifically.
Diagnose
He applies both percentages to the original price
The teacher asks him to talk through one he got wrong: a bag costs $80, its price is reduced by 15% in a sale, and 9% GST is then added to the sale price. He works out 15% of $80 = $12 and the sale price of $68 correctly. Then, instead of finding 9% of $68, he finds 9% of $80. He isn't confused about percentage at all — every single calculation is done correctly. What he loses is the base: after the first percentage change, the quantity the next percentage acts on has changed, and he keeps using the original. It's a well-known error pattern in successive percentage change.
Target
We target the base, not percentage itself
So we don't send him back to revise percentage. His teacher targets one habit: write out the new value after each percentage step before applying the next one. "Step 1: 15% discount → new price = $68. Step 2: 9% GST of $68 = $6.12. Final price = $74.12." Same questions, but the base for every step is written down in full before a single new calculation is made.
Practise
Running base written out, then without
In the same session he does six successive-percentage problems — discounts then GST, a price increase then a further increase, a quantity decreased twice — with the running base written out at every step. The next session, the written base comes off. Two he still rushes; the teacher stops him mid-question and asks, "9% of what?" — and he catches it himself.
Review
Closed within two sessions
At the end of the second session he's given a fresh set of successive percentage change problems, with no scaffolding. He gets them right, and more importantly he gets them right for the right reason — he recalculates the base each time. The parent sees the note: gap identified as base not recalculated in successive percentage change, closed in two sessions, monitoring for recurrence under time pressure.
The point of the example: he never had a percentage problem. If we'd drilled percentage harder, he'd have stayed stuck. Finding the actual reason — the base he wasn't recalculating — is the whole job.
The gap-finding happens live, in conversation with a real teacher
A teacher watching the working
A teacher sees the hesitation before the wrong answer, hears "I just took 9% of the $80," and asks the follow-up question that exposes the real cause. That's where diagnosis actually happens.
Not an algorithm scoring a quiz
A quiz score can tell you a child got percentage questions wrong. It can't tell you every calculation was correct and only the base was wrong. Only someone watching him work can tell you that.
Where AI fits — and where it doesn't
AI helps us create and personalise learning content. Teachers remain at the centre of your child's learning.
We use AI the way a teacher uses a good photocopier and a good question bank: to build targeted practice faster once the teacher has decided what the child needs. The deciding, the diagnosing, the teaching and the judgement of whether your child has really improved all stay with the teacher.
We will never tell you AI teaches your child. It doesn't. A person does.
See the method run on your own child
The $20 Learning Gap Assessment is stage one, done properly — a real live session inside a small group, with a report of exactly what we found.