What actually decides the quality of a lesson?
We analyzed 4,796 lessons in one private school, every lesson from February to May rather than a sample, and checked the usual explanations for a weak lesson one by one.
- 4,796 lessons
- 51 teachers
- 22 classes
- 35 subjects
- February–May 2026
“It was the last lesson on a Friday.” We checked.
When a lesson goes badly, the timetable or the class usually takes the blame. We tested each usual explanation against the scores on the same 25 criteria.
- Day of the week
- Class size
- Time of day
- The teacher
The timetable moves the score by a few points. Between two teachers the gap reaches 28
Lesson score: the total on the same 25 criteria, as a percentage. Gaps are in percentage points. Teachers of one subject: the largest gap. Same teacher across classes: the median.
The day of the week changes nothing
Friday averages 77%, Tuesday and Thursday 79%, on about a thousand lessons each.
Every lesson from February to May, 25 criteria
A smaller class does not mean a better lesson
Almost every school believes it. The data does not bear it out.
CaveatThis is a private school. Most classes have 11 to 15 pupils, so very large classes are not represented.
point between the smallest classes and the largest
Six points between the first lesson and the seventh
The only outside condition that moves the score noticeably. It is a timetable decision, not a teaching one: teach the hardest new topic earlier in the day.
Every lesson from February to May, 25 criteria
Two teachers, same subject, 28 points apart
Same school, same subject, and the gap between the strongest and the weakest teacher is larger than anything the timetable explains. So the gap sits with the teacher.
points: the largest gap between two teachers of one subject
The same teacher is not the same in every class
So this is not a teacher league table either. The gap shows up at the level of this teacher in this room, and that is something you only see by watching every lesson.
The numbers16 teacher-and-class pairs with at least 12 lessons each. Median gap 7.6 points, largest 17.9.
points between a teacher’s own classes, typically
One school, four months, every lesson in it
Not a benchmark across schools, and not a full school year. It is what you see when nobody picks the lessons to look at.
The sample
- Lessons
- 4,796
- Teachers
- 51
- Classes
- 22
- Subjects
- 35
How to read it
- 4,796 lessons with a full analysis, February to May 2026.
- The same 25 criteria on every lesson. Nobody chose which lessons went in.
- One private school: 51 teachers, 22 classes, 35 subjects.
- Teacher and pupil data is anonymized.
- It shows where the gaps are, not what causes them.
Questions about the study
4,796 lessons in one private school, February to May 2026: 51 teachers, 22 classes, 35 subjects. Every lesson of that period went in, each assessed on the same 25 criteria.
Barely. Average scores ranged from 77% on Friday to 79% on Tuesday and Thursday: two points between the best day and the worst, on about a thousand lessons each.
Not in this school. The gap between the smallest classes and the largest was one point. Most classes have 11 to 15 pupils, so very large classes are not represented.
A little. The first lesson of the day averaged 81%, the fourth 78%, and the seventh 75%. It is the only outside condition that moves the score noticeably, and it is a timetable decision: teach the hardest new topic earlier in the day.
The biggest gaps are between teachers and between classes, not in the timetable. The strongest and the weakest teacher of the same subject were 28 points apart, and the same teacher was typically 7.6 points apart between their own classes. The study shows where the gaps are, not what causes them.
Yes. Lansy reviews every recorded lesson against your school’s own criteria or the UAE School Inspection Framework, so leaders see every lesson instead of a handful of observed ones. A working pilot takes one to two days.