True Life School Why gaps are the real enemy
Mathematics or grammar failed in April is built on material half-learned in October. Mastery gates catch the October gap while it is still small.
Evidence
We did not invent this model. We assembled it from research education has known for decades.
Before we can show you our own results, we owe you the scientific foundation. Every claim below has an author and a year, so you can check us — and we say plainly where the evidence ends and our work begins.
+2σ 1:1 tutoring effect (Bloom, 1984)
1968 mastery learning first described
~70% of students report boredom in class (PISA 2022)
The 2 sigma problem: tutoring works absurdly well.
In 1984, educational psychologist Benjamin Bloom published a finding that still frames the whole field: students taught 1:1 with mastery methods performed about two standard deviations above a conventional classroom — the average tutored student beat roughly 98% of classroom students.
- Bloom called it a problem, not a triumph: a tutor for every child was economically impossible. His challenge to the field was to find group methods as effective as tutoring.
- For forty years the blocker was cost. Adaptive technology changes that arithmetic: personal pace, personal level and instant feedback stop being a luxury.
- This is the entire reason our platform exists: not to replace people, but to make Bloom’s 1:1 condition affordable — with humans doing what tutoring alone never covered.
Move on when you can do it — not when the bell rings.
Mastery learning (Bloom, 1968) flips one classroom assumption: time becomes flexible and understanding becomes fixed. A child moves forward after demonstrating mastery, so gaps stop compounding. Meta-analyzes through the decades (including Kulik, Kulik & Bangert-Drowns, 1990) consistently found meaningful positive effects.
What the class pace really is
One speed for 25 children is too fast for some and too slow for the rest — usually both in the same hour. PISA 2022 data shows how that feels from the desk: most students report being bored.
What this buys the day
A child working at the right level does not need six hours of seat time to cover the core. That is the arithmetic behind the two-hour block — focus replaces waiting.
Three memory mechanisms, wired into the software.
How practice is scheduled matters as much as what is practised. The platform implements three of the best-replicated effects in cognitive psychology.
Retrieval practice
Recalling knowledge strengthens it far more than re-reading it (Roediger & Karpicke, 2006). The block is built from answering, not from watching.
Spaced repetition
Forgetting follows a curve described by Ebbinghaus in 1885; spacing reviews against it (meta-analysis: Cepeda et al., 2006) is the difference between knowing for the test and knowing for years.
Immediate feedback
An error corrected within seconds does not get rehearsed into habit. A classroom returns homework days later; adaptive software answers in the same breath.
What the research proves — and what we still have to.
The studies above validate the methods, not our school. That distinction matters to us. Until we have our own cohorts and our own numbers, we will not borrow the word “proven” for ourselves.
- What we commit to measure: entry diagnosis, per-semester progress on knowledge atoms, exam results and child wellbeing.
- Where results will appear: in the parent dashboard for your child, and publicly per campus once cohorts exist.
Our promise: we will publish our data — including the numbers that are not flattering. A school built on measurement cannot be shy about being measured.
Parent questions
Two focused hours at the right level replace six hours built around the average of 25 children. The mastery structure means nothing is skipped — the pace is personal, the coverage is complete, and state exams verify it every year.
Correct — and that is their strength. Tutoring, mastery, retrieval and spacing were proven long before language models. Technology is the delivery mechanism that finally makes them affordable for every child, not the theory itself.
Yes. Measurement is built into the model — entry diagnosis, semester progress, exams, wellbeing — and campus-level results will be published once real cohorts exist. Including the uncomfortable ones.