Chess Academy Batch Scheduling: Coach Load Grid

Split students by rating range first and age second. Rating decides what a group can be taught; age decides whether the group can sit together. A batch is teachable when the gap between its strongest and weakest player is small enough that one lesson lands for everyone, which in practice means keeping the rating spread inside roughly 300 points. The Elo system was designed to express relative strength between players, so it is the natural banding axis [1].

One trap to avoid: online and over-the-board ratings are different currencies. Lichess and most online platforms run Glicko-2 style ratings, which sit on a different scale from FIDE Elo and move differently, especially for new accounts [2]. Pick one reference number per band (most academies use the online rating because every student has one) and convert mentally for the few students with FIDE ratings, rather than mixing both in one column.

Age is the second filter, and it overrides rating at the edges. A sharp 6-year-old rated 1100 does not belong in a Development batch full of 12-year-olds, even though the number says so; attention span, writing speed, and social comfort all differ. The clean fix is a junior section inside a band (Development Junior, Development Senior) rather than promoting young students into older groups.

Six to ten students per batch is the working range for most academies, and the right number shrinks as the level rises. Beginners are doing pattern drills and guided games, which scale well; a Foundation batch of ten or twelve runs fine with one coach. Tournament players need their own games reviewed and their own questions answered, which does not scale; past six students, individual attention per session collapses.

• Foundation: 8 to 12 students. Group drills and minigames carry the session, so headcount is cheap.
• Development: 6 to 10 students. Enough for pair work and internal tournaments, small enough for individual feedback.
• Tournament prep: 4 to 6 students. Every student needs game review time inside the session.
• Advanced: 2 to 4 students, or one-to-one. At this level the batch is closer to a training squad than a class.
• Hard ceiling: never open a second section late; split a batch the moment it crosses its band's ceiling, not the term after.

Batch size is also a revenue decision, so set it before you set fees, not after. A Tournament prep batch of five at a higher fee usually out-earns a crowded batch of ten at a discount, and the parents can see the difference in attention. How that maps to your fee table is its own topic, covered in our guide to academy fee structure.

Build the grid in a fixed order: coach availability first, rooms second, batches last. Most timetable conflicts come from doing this backwards, promising parents a slot and then discovering the coach or the room is already spoken for. Start by writing down each coach's true weekly availability, including commute gaps between branches, and treat those windows as the canvas everything else is painted on.

Then apply a coach load cap before you place a single batch. A coach teaching chess well is preparing positions, reviewing student games, and talking to parents around every session; 20 to 24 contact hours a week is the realistic ceiling before preparation quietly stops and sessions turn into babysitting. Schedule each coach in contiguous blocks (three batches back to back beats three batches scattered across a day) and leave at least 15 minutes between batches for handover, parent questions at the door, and resetting the room. If demand exceeds the cap, the answer is hiring or a waitlist, not a 28-hour week.

The conflict matrix above covers the five collisions that account for nearly every timetable fire. Run a change through it whenever anything moves: a new batch opening, a coach's college schedule changing, a room being repurposed for a weekend camp. The check takes two minutes against the grid; the same conflict discovered on a Tuesday evening, with eight families standing outside a locked room and a coach stuck across town, takes a week of apologies and at least one lost enrolment.

Make-up slots live inside the weekly grid as standing reservations, not as favours arranged over the phone. Block one or two fixed windows per week (a common choice is Saturday late morning and one weekday evening) and route every missed-class recovery into them. The slot exists whether or not anyone books it, the way an airline holds seats.

The alternative, scheduling make-ups ad hoc, is how grids rot. Every individually negotiated catch-up class either overloads a coach past their cap or quietly steals a room from a regular batch, and within a term nobody can say what the real timetable is. Standing slots also make the policy fair: every family gets the same two windows, so there is no perception that pushier parents get private rescheduling. The policy side, including how credits are earned and expire, is its own system; our guide to make-up classes covers the credit ledger in full.

Reshuffle at term boundaries only, on a fixed calendar the parents already know. Mid-term moves break social bonds, confuse fees, and create a steady drip of one-off timetable changes. A student who has clearly outgrown their band mid-term can be given harder material inside the same batch for a few weeks; the move itself waits for the term break.

1. 1Two weeks before term end, pull every student's current rating and rating trend alongside their attendance.
2. 2Mark promotion candidates: students sitting above their band's ceiling for four or more consecutive weeks.
3. 3Check age fit and batch capacity in the receiving band before confirming any move.
4. 4Message each affected family individually before the new grid is published, with the reason for the move.
5. 5Publish the new term grid at least one week before classes resume, and freeze it.

The individual message matters more than the grid itself. A promotion framed as recognition (your child has outgrown the group) lands well on WhatsApp, where the family will actually read it; a business messaging channel that parents already use beats a notice board for this kind of news [4]. A demotion should almost never happen by message at all: that is a conversation, and usually a sign the original banding was wrong.

Online batches change three grid variables: session length, room logic, and coach geography. Screen sessions hold attention for 60 to 75 minutes against 90 to 120 in person, so an online batch takes a shorter slot; rooms stop mattering, which removes one whole class of conflicts; and a coach can take an online batch from home, which turns dead commute windows into teachable hours.

Tooling for the online slot is cheap. A Lichess study works as the shared demonstration board, with the coach moving pieces while students annotate, and it costs nothing [3]; Lichess itself is free and open source, which is why it has become the default classroom server for academies [5]. The harder problem is hybrid batches, where some students sit in the room and some join remotely: the remote students consistently get less attention. The cleaner pattern is separate online-only batches in their own grid rows, usually in the early-evening window that physical rooms cannot serve. Running a fully online academy stretches this further, and our guide to starting an online chess academy covers that model end to end.

One scheduling bonus: online slots make excellent make-up windows. A student who missed a physical Tuesday class can join an online recovery session without anyone unlocking a room, which is often the cheapest way to add the second make-up window when the first fills up.