"72%" answers
no question anyone asked.
Most progress reports lead with a number because a number is what the system produces and because it looks objective. Consider what a parent can actually do with it. Seventy-two per cent of what? Is that good? Better than last term? Better than the other children, and should that even matter? Does it mean my daughter can order a coffee in Spanish or not? The number has the form of information and carries almost none, and its main achievement is to prompt an email asking what it means.
The deeper problem is that a percentage describes performance on the tasks the school happened to set, which is a proxy for the thing the family cares about — whether the money is producing a person who can use the language. Language learning is particularly badly served by scores, because the visible gains arrive in bursts after long invisible stretches, and a flat percentage across a plateau tells a worried parent nothing about whether the plateau is normal. It usually is. Say so.
Four questions,
and only four.
The third row is the one schools soften into meaninglessness, and softening it is a false economy. A report in which everything is going well and nothing is difficult reads as a form letter, and it removes the school's ability to say anything serious later — if every report was positive, the first honest one arrives as a shock. Naming one real difficulty, with what you are doing about it, is what makes the positive parts believable.
The fourth row is the one to answer out loud. Parents rarely ask "should I be worried?" because it feels like an admission, and they are asking it anyway. A sentence that says plainly whether this is normal, and that a plateau in month four is what learning a language looks like, is often the most valuable thing in the document.
Three sentences
and one thing to do.
The report that works is much shorter than the one most schools produce. A line on where the student sits against the level, in the framework's own language rather than a score: working comfortably within B1, close to moving up in speaking. A concrete gain: in September she could not sustain a conversation past a couple of exchanges; last week she argued with a classmate about a film for five minutes. A named difficulty with a response: past tenses are still unreliable in speech — we are drilling them in pairs rather than on paper, which is where the gap shows.
Then one thing the family can do, if there is one, and only if it is genuinely small. "Fifteen minutes of a series in Spanish with subtitles, twice a week" is usable. "Please practise more at home" is not advice, it is a transfer of responsibility, and parents recognise it as such.
Keeping this to a screen is a discipline rather than a limitation. A long report is read less carefully than a short one, and the length tends to grow in inverse proportion to how much the writer has to say.
Nothing significant
should arrive here first.
This is the rule that matters more than the format: a family should never learn something important for the first time in a formal report. If a student is struggling badly, or is going to be moved, or has stopped attending, that is a conversation on the day it becomes clear — and by the time the report arrives it should be a written confirmation of something already discussed. A report that breaks news is a report that says the school noticed and did not tell you, which is a worse message than the news itself.
Cadence follows from that. Twice a term is usually right for a regular course — one brief mid-term note, one fuller report at the end — plus immediate contact whenever something changes. The mid-term note is the one that earns its keep, because it arrives while there is still time to act and it prevents the end-of-term report from being the first time anyone raised anything. For short intensive courses, a single close-out report plus same-day contact is enough.
Most of what
the system can generate.
The third row deserves particular care in a language school. Comparing a learner to their class produces a ranking, and a ranking makes half the class feel behind by definition, including students who are progressing perfectly well. The comparison that means something is with the same student in September, and that is the one a family can act on rather than merely feel something about.
There is also a line here about what is reasonable to measure and report about children at all. That argument belongs to the learning data article, which covers it properly, and the short version is worth carrying into every report you design: the fact that a system can report something is not a reason to report it, and some of what it could report should not be collected in the first place.
On what a platform can do here honestly: it can hold attendance, assessment results and the level a student sits at, so that whoever writes the report is not reconstructing a term from memory. It does not write the report. The three sentences that make it useful are a teacher's judgement, and they are the part a family is paying for.