The first article in this series covered the four figures that are fixed before opening day: capital cost per seat, fee level against operating cost, time to first student, and the funding needed to reach break-even. The four that follow are about whether anyone comes.
They are the figures that live in enrolment forecasts, and enrolment forecasts are where most school investments are quietly lost. A building can be costed to within ten per cent and an admissions curve cannot. The temptation to draw the curve the model needs rather than the curve a real school produces is the single most common failure GSE sees in the projects it is asked to rescue, and the reason most new school projects fail financially rather than academically.
As in Part One, each figure carries a line stating where it comes from, and none travels above its basis.
5. Year one enrolment
Expect 15 to 30 per cent of design capacity in the first year.
Anything above 35 per cent is either an exceptional market, a converted school with an inherited roll, or an optimistic forecast. Year one is won or lost twelve months before opening, through admissions presence, the appointment of the Head and community credibility.
The sequence matters more than the budget. A Head appointed a full year out, visible in the community and answering parents’ questions in person, converts enquiries that a marketing campaign cannot. An admissions office that opens before the building does, with a fee schedule the family can commit to and a founding-family offer that rewards the risk of joining first, turns interest into deposits. Schools that appoint late and open admissions late spend the same money and land at the bottom of the band.
One caution on the denominator. Most campuses are built in phases, and operators publish opening headcount rather than day-one design capacity, so a clean percentage is rarely available. What is visible supports the shape: Nord Anglia opened Abu Dhabi in 2023 with 350 students on a campus designed for several times that; GEMS opened Dubai British School Mira in 2025 with more than 700, which is what an established brand in a proven catchment can do.
Basis. GSE practitioner range. Opening headcounts as stated by the operators; Nord Anglia Abu Dhabi from its BSO inspection record, GEMS from company statements reported in 2025.
6. The ramp curve
Most schools reach maturity in year five to seven. Growth of 60 to 100 per cent in year two, more at a well-received premium opening, then 30 to 50 per cent in year three, then a flattening curve.
Year two can run past the top of that band. Nord Anglia Abu Dhabi went from 350 students in its first year to 861 on roll by October of its second, an increase of about 146 per cent. A strong brand in a market with pent-up demand can do that. A model should not assume it.
Year three is where models break. Most of what goes wrong there was decided in year one, usually a founding cohort admitted too loosely to fill seats, or a Head chosen for the opening rather than for the years after it. Early adopters have joined, the novelty has gone, and the school now has to win families who waited to see results. If enrolment growth stalls in year three, the cause is almost always reputational rather than commercial. Fix the school before increasing the marketing budget.
Basis. GSE practitioner range. The Abu Dhabi figures are from the school’s BSO inspection report, October 2024.
7. Break-even occupancy
Typically 55 to 70 per cent of design capacity, measured after rent.
A school designed for 1,200 that breaks even at 800 has a workable margin of error. One that breaks even at 1,050 has almost none. Ask for this figure early, and ask whether it is before or after the rent, because a leased school lives on the second number. It tells you more about the resilience of an investment than any projection of year ten profit.
The published anchors are utilisation at profitability rather than at break-even, which is the more useful side of the line anyway. Curro has run at 69 to 70 per cent of built capacity while comfortably EBITDA positive, and describes 85 per cent as the practical ceiling for a developed school. Nord Anglia reported 79 per cent utilisation in 2014 on a group margin near 25 per cent.
Basis. GSE practitioner range. Curro utilisation from company data in its 2022 initiation report; Nord Anglia from its F-1, 2014. Neither operator publishes break-even utilisation after rent.
8. Cost to recruit one enrolled student
USD 2,500 to 6,000 per enrolled student at a mature school, fully loaded with admissions staff. A new international school in its growth years should expect to sit above that.
The only rigorous benchmark comes from US independent schools. The National Association of Independent Schools put the median cost per enrolled student at about USD 3,700 in its 2022 study, lower for elementary programmes and nearer USD 5,800 for secondary, and found that roughly ninety per cent of that cost was salaries and benefits rather than advertising. No comparable dataset exists for international schools, so treat the range as a reference point and measure your own.
Mature schools with waiting lists spend very little on acquisition. New schools spend heavily and should. The useful discipline is measuring cost per enrolment rather than cost per enquiry, and counting the admissions team in the number, because most underperforming admissions functions have a conversion problem rather than a traffic problem, and a school that counts only its advertising spend will never see it.
Basis. Published, NAIS 2022, US independent schools. The international adjustment is a GSE practitioner view.
What to test
Two relationships settle whether a forecast is honest. Is the funding sufficient to reach break-even occupancy, not just to complete construction? And does the ramp assume a year three that a real school could actually deliver? A model that passes both is rare. The seven figures in the final article decide whether the school that fills stays full, and stays profitable, for the decade after.
Read next
- International Education by the Numbers, Part One. The four figures that decide the build.
- International Education by the Numbers, Part Three. The seven figures that decide whether it lasts.
- Why Most New School Projects Fail Financially. The enrolment assumptions that survive a model and not a school.
- How to Conduct a Feasibility Study for a New School. Demand analysis before specification, in that order.
- How to Value an International School. What a ramp curve is worth to a buyer.
To talk about your school project, reach us through the contact page.