The Hours a School Gets Back

Every education platform promises to save teachers time, and almost none of them say where, how much, or what they cost in return. So here is the count as we see it from the school portal, including the parts that go the other way.

Marking: this one is real

A quiz scores itself the moment a student finishes. Not later that evening, not on Thursday. The result is recorded against the topic, so a class result reads as "eleven of twenty-eight missed both questions on electric circuits" rather than as a column of percentages.

For a teacher setting weekly practice to four classes, this is the largest single block of time the system removes, and it is removed completely rather than reduced. It is also the block that was never teaching. Nobody became a teacher to tick boxes against an answer key.

The exception is written work uploaded as a file. Those are marked by hand, with a score out of a maximum the teacher sets. We have no interest in pretending a model can mark an essay to a standard a school should accept.

Building practice material: a large drop, not a disappearance

Before: a teacher writes a worksheet or a set of questions, types it, formats it, prints it, and writes an answer key.

After: they upload the chapter, get a draft, read every question, change the ones that need changing, and publish.

The drop is substantial and it is not to zero, because reading twenty drafted questions properly takes half an hour and that half hour is the part that must not be cut. Our own count from teachers doing this regularly is that three or four questions in twenty get rewritten or deleted on a good chapter, more on a vague one. If a teacher ever finds themselves changing nothing, that is a sign to read more carefully, not a sign of a better model.

Reporting: changes shape

The figures a head of department wants at the end of a term, which topics a cohort is weak on, which students are behind, how a class is trending, already exist as a consequence of students doing the work. Nobody assembles them.

So the saving here is not hours of spreadsheet work removed so much as hours of spreadsheet work that never need to start. The remaining work is reading and deciding, which is the work that was always worth doing.

Administration: front-loaded

Setting up a school is a day. Classes, teachers, branches, students imported from a spreadsheet, coupon codes for onboarding. That day is written up step by step.

After that the administrative load is close to flat. Adding a student is a row. Adding a class is a form. The ongoing cost is small because the data model matches how a school is actually shaped, with branches and classes and sections, rather than a flat list of users.

What it adds

Three things, and a school that is not told about them in advance feels misled.

The first fortnight. Setup, importing students, deciding which classes allow retries, which allow hints, which are full-screen for tests. Real work, done once.

Learning to review. The first few generated quizzes take longer to check than they should, because reading for subtle wrongness is a skill. It gets quick. It never gets free.

The monthly batch. If a school uses the monthly generated quizzes, someone publishes them. The review desk allows publishing a whole month at once, so this is minutes rather than hours, but it is not zero and it should not be.

The place the saving is bigger than expected

Not marking. Intervention.

The old loop is: set work, collect it, mark it over the weekend, hand it back on Tuesday, notice on Wednesday that a third of the class has the same misconception, re-teach it on Friday. Nine days from problem to fix, and by then the class has moved on and the misconception has set.

The new loop is: set work, read the topic breakdown that afternoon, start tomorrow's lesson with the ten minutes that fixes it. One day.

That is not a time saving in any spreadsheet sense. The teacher works the same hours. But eight days of a class carrying a misunderstanding is a real cost that nobody has ever put on a timesheet, and removing it is the change teachers mention first when we ask what actually improved.