There's a version of the education story that gets told from San Francisco, and it's about American schools and American tools. The larger story, by population, is somewhere else entirely: in the countries that inherited British curriculum and a particular way of running a school, and that are only now getting the one thing that was always missing.
The shared inheritance
Walk into a secondary classroom in Malaysia, Nigeria, Kenya, India, or the Caribbean and you'll find structural cousins. Curricula descended from or aligned to the British model. An exam at the end that decides a lot. And underneath, the same engine: students advance by time served, not by mastery shown. You move up with your age cohort whether or not you understood last year's material, and the gaps compound — quietly at first, then all at once around the exams that matter.
This isn't a quirk of any one country. It's a design inherited at scale, which is why the same product logic can work across very different markets. That is unusual, and it's the first half of the opportunity.
The access fee, and the thing that removed it
For decades there was a release valve, and it was money. Families who could afford it bought their children out of the structural problem with tutoring — one-to-one or small-group teaching that does what a crowded classroom can't. The research on why this works is old and settled: good tutoring with mastery-based instruction produces results far above ordinary classroom teaching (the two-sigma finding). The problem was never that we didn't know what worked. It was that what worked cost more than most families had.
AI changed exactly that. The thing that cost forty dollars an hour in a person now costs a few dollars a month in software. The access fee didn't shrink — it collapsed. For the first time, the teaching that was the preserve of wealthy families is deliverable to everyone, in the one form that can equalise it: software, which has no marginal cost of a tutor.
Why these markets are underserved
If the opportunity is this clear, why isn't it crowded? Because capital builds where capital is. The default of global EdTech is to build for Tier-1 markets first — US and UK curricula, English-first, premium pricing, broadband assumed — and to treat everywhere else as a later export market. So these systems get served by tools built for someone else's: confident in the wrong syllabus, monolingual where classrooms are not, priced for a different economy, and assuming devices and bandwidth that aren't universal.
The result is a large, structurally similar set of markets that are either served badly by exports or not served at all. The regional players that do exist prove the demand is real — platforms across Southeast Asia, South Asia, and Africa have reached tens of millions of learners — but the field is young and the best products, the genuinely system-native ones, are mostly still to be built.
Why the winners will be system-native
The lesson of the exported tools is that fit beats polish. A learning product has to clear four things for the students it serves: the curriculum they're examined on, the languages they learn in, a price their family can pay, and the devices and connection they actually have. Miss one and it's product-export, however good the technology. The companies that win these markets will be built from those four constraints out, not shipped in and translated. We've written the test for that elsewhere — the Four Mismatches — and the thesis underneath it.
That's the shape of the opportunity: a shared structural problem across enormous markets, a cost barrier that just fell, capital looking the other way, and a clear design principle for what wins. It's not a niche. It's most of the world's students, finally reachable.
What has to be true to win
Three things. Build natively — from the local system out, on mobile, in the right languages, at the right price. Measure mastery, not engagement — the moat is whether children actually learn, which is also the only marketing claim that overcomes a sceptical parent. And find the demand that already exists — parents, schools, and, for training, the budgets employers already hold. The opportunity is large, but it rewards the builder who respects the system, not the one with the best demo.
Frequently asked questions
Isn't the Commonwealth too fragmented to be one market? It's many markets, but with a shared structural inheritance that lets the same product logic travel. You localise the syllabus and language per market; the underlying design — mastery-based, mobile-first, affordable — carries across.
Hasn't AI made EdTech a crowded space everywhere? At the tool layer, yes. At the system-native layer for these markets, no — most are still served by exports that don't fit. Crowded is not the same as well-served.
Why now and not five years ago? Because the access fee only just fell. The pedagogy was known for forty years; what changed is that AI made it cheap enough to reach the families who were always priced out.
A note on why we wrote this. This is the opportunity Addestra builds into: system-native education for the Commonwealth, starting in Malaysia. We co-build ventures aimed at it. How we work: editorial standards.
Sources
- Education Next — "Two-Sigma Tutoring: Separating Science Fiction from Science Fact": https://www.educationnext.org/two-sigma-tutoring-separating-science-fiction-from-science-fact/
- Our framework: System-Native EdTech, The Four Mismatches Test