More batches per teacher, with the teachers on board
A constraint-based timetable, and one-to-one conversations before any announcement, took the institute from 11 batches on 15 teachers to 18 batches on 10.
Batches per teacher0.7 to 1.8
The situation
After a co-founder transition I took charge of academics, operations and finance. Teachers carried fewer batches than the centres needed, and the timetable was reworked every week.
What I did
I wrote down more than 25 real constraints (who can teach what, where and when) and let optimisation software search for a timetable that people could not build by hand. A script then turned it into a one-click output for every batch and every teacher, replacing about half an hour of admin work each time.
The timetable was the easy part. Before any announcement I spoke to every teacher and centre head one to one, and framed the change around their own day and around student outcomes. Then I announced a shorter teaching day, from 6 hours 45 minutes to 6 hours, keeping the three-lecture structure. It was a small concession that showed teachers they had been heard.
Resistance still appeared after the timetable went live, as it usually does. I had already started a parallel hiring pipeline, with one or two alternates for each subject, as a contingency.
What changed
| Measure | Before | Now |
|---|---|---|
| Batches per teacherAbout 2.5 times | 0.7 (11 batches, 15 teachers) | 1.8 (18 batches, 10 teachers) |
| How often the timetable changes | Every week | Monthly or every two months |
| Adoption of the new timetable | Strong early resistance (50–60% of the team) | Near-full adoption within about four weeks |
| Admin time to publish the timetable | About 30 minutes each time | One click |
What it cost, and where it stops working
The teaching roster is smaller than it was. Optimisation gives a timetable. It does not give agreement. The one-to-one conversations did that work, and they took longer than the software.
What it shows
Change that sticks because the people affected helped shape it.
How this was measured
Internal timetable and roster records, comparing the roster before and after the redesign. Batches per teacher is active batches divided by teachers on the roster (11 of 15 and 18 of 10, rounded). Unaudited.
Kind of problem
Turnaround and change management
Related