The Second Cohort Is the Real Test — and It's the One NDX Is Built to Pass
Almost any programme can train its first group of teachers well. The systems that last are the ones that keep the next teacher — the one who arrives after the launch — just as capable, with no visiting trainer, no repeat budget, and no signal required. That's a design choice, and it's the one NDX makes.
Every digital education programme trains its first cohort of teachers well. The launch depends on it. The devices arrive, the trainers fly in, the first group learns the system in a room built for the occasion, and the classroom that results is real. That part is not where programmes fail.
They fail eighteen months later, with the second cohort — the teachers who were not in that room. Someone transferred. Someone retired. A new graduate took the post. The school is the same, the platform is the same, but the person now standing in front of the class never received the training that made the launch work — and there is no longer a budget line, a trainer, or an event to give it to them.
This is the quiet way good programmes decay. Not through faulty technology, but through turnover the design never planned for. A system is only as capable as the teacher using it this term, not the teacher who was trained two years ago and has since moved on.
So the honest test of any national rollout is not how the first cohort performs. It is what happens to the third. And this is the point NDX is built around, so it's worth being concrete about what that means in a school.
The training lives inside the system, not alongside it
With NDX, becoming capable on the platform is not a one-time event that happened at launch — it is part of the product a new teacher opens on their first day. Guidance, worked examples, and support are built into the tool itself, so the teacher who arrives in year three learns it the same way the first cohort did, in the moment they need it.
It works on the hardware already in the school, offline
Because NDX runs on the devices physically present in the classroom rather than a distant server, that onboarding doesn't wait for a workshop to be funded or a connection to hold. A teacher in a school four hours past the last cell tower gets the same start as one in the capital, the week they arrive — not the next time a trainer can reach them.
It assumes people come and go, because they do
NDX is designed for turnover from the start, so capability renews locally instead of depending on the same few trained individuals staying in post. The ministry isn't re-buying the launch every eighteen months; it owns a system that keeps making teachers capable on its own.
For a ministry, this changes the arithmetic of the whole investment. A programme that only trains the first cohort has costed the launch and quietly left the maintenance to chance — and to next year's budget. A system that renews teacher capability locally protects the money already spent, keeps the standard steady as staff move through, and reaches the last school on the road instead of the first.
That is where the dignity of a national education system actually lives: in a child able to count on the same standard of learning next year, and the year after, regardless of which adults have moved through the staff room. A rented, launch-and-leave programme can promise that. Only a system that keeps making the next teacher capable can keep the promise. That is the standard NDX is built to hold — and to hold long after everyone has stopped talking about the launch.
