“Our enrolments are down” — how to tell a bad year from a trend

“Our enrolments are down” is a trend only when the fall survives three tests: it repeats across two consecutive intakes, it opens a gap against your sector, and it shows up in enquiry volume rather than conversion. Australian school enrolments grew 0.7% in 2025 — but government schools fell 0.2% while non-government schools rose 2.3%. One soft intake is noise.

The short version, before you brief the board

• Compare against your sector, not against last year. In 2025 total enrolments rose 0.7% to 4,160,918 students across 9,673 schools, while government primary enrolments fell 1.1% (ABS, Schools, released 5 March 2026).
• Split the drop before you spend a dollar. Fewer enquiries and worse conversion of the same enquiries are different problems with opposite fixes.
• Accept the lag. A change you make this term cannot move the intake that is already choosing; it moves the one after it.
• Two quiet weeks is noise; two quiet intakes is a trend. Nothing structural should be rebuilt on a single cycle.

Is our enrolment drop normal, or is it just us?

The single most useful number is not your enrolment change — it is the gap between your change and your sector’s. The ABS Schools, 2025 release puts hard figures against that: nationally, enrolments grew 0.7% (28,912 students) in 2025. Underneath the national number, government schools lost 6,109 students (−0.2%) and non-government schools gained 35,021 (+2.3%). Over 2021–2025, independent school enrolments rose 15.3% and Catholic 5.7%, while government schools fell 0.4%. A government primary school flat in 2025 outperformed its sector. An independent school flat in 2025 lost share in a growing one.

The thresholds below are our working rule, not a published standard — but they stop a normal cycle being escalated into a strategy review.

The intake-lag rule: what a change made this term can actually move

The intake-lag rule: an enrolment decision is made months before the student appears in your census, so a marketing change made this term cannot move the intake that is already deciding — only the one behind it. This is the fact that reframes the panic, and it is checkable against published timetables. For NSW selective high school placement for Year 7 entry in 2027, the placement test sits on 1–2 May 2026, the last day for families to change school choices is 5 June 2026, and outcomes are released 19 August 2026 (NSW Department of Education key dates). Those families locked their decision roughly eight months before the first day of Term 1, 2027.

Say this out loud to your board before anyone approves a campaign: money spent in Term 3 will not appear in next February’s census, and a vendor who implies otherwise is selling you the calendar you want rather than the one that exists.

What to do in the next 24 hours

None of this costs money and none of it involves an agency. First, pull every enquiry from the last two intakes out of your SIS or CRM and count them by month — you are looking for whether the fall starts at enquiry or later. Second, list every family who received an offer and did not accept in the last cycle, with the date and the reason field if you capture one. Third, check your enquiry response time: how many hours between a form submission at 8pm and a human reply. Fourth, confirm nothing is broken — a dead enquiry form, an unmonitored inbox, or a booking link that expired is a distressingly common cause of a “marketing problem”.

What to do in the next 7 days: the enquiry-to-enrolment audit

Run the funnel as five ratios on data you already hold. Enquiry → tour or interview, tour → application, application → offer, offer → acceptance, acceptance → enrolled on census day. A worked example with illustrative numbers, so you can substitute your own: 420 enquiries, 190 tours (45%), 96 applications (51% of tours), 78 offers (81%), 61 accepted (78%), 57 enrolled (93%). That is 13.6% enquiry-to-enrolment. If last year the same school ran 430 enquiries to 74 enrolments (17.2%), enquiry volume barely moved and the loss is conversion — specifically the tour-to-application step, where 51% is doing the damage.

The arithmetic matters because it tells you where not to spend. Recovering the previous 17.2% on 420 enquiries returns 72 students without a single extra enquiry. Buying your way back with new enquiries at 13.6% would take roughly 530 enquiries to reach the same place — over 100 more, at whatever your cost per qualified enrolment call actually is. Conversion is cheaper than volume almost every time you have both options.

The list of past enquirers who never enrolled is usually the most valuable file a school owns and the least worked. As a reference point from outside education: across LeadsNow’s Colliers-era database reactivation campaigns, dormant commercial databases converted at 4.4% on average and 8.9% at peak. That is our record on a commercial database, not an education benchmark, and school enquiry lists behave differently — but the direction holds: people who once raised their hand answer at a rate cold audiences do not.

What to change this term, and what to leave alone

Change the things that act on families already deciding: speed and persistence of follow-up, the number of tour slots offered per week, who makes the call after an unaccepted offer, and whether an enquiry at 8pm on a Sunday gets a reply before Tuesday. Leave alone anything that requires a full cycle to show a result — a new brand, a new website, a new positioning statement — until you have two intakes of evidence that the fall is a trend. Rebuilding a brand to fix a one-year dip is the most expensive way to discover the dip was noise.

When it is not you: policy, demography and the things marketing cannot fix

Some enrolment falls are structural, and no campaign touches them. Local demography is the big one: government primary enrolments fell 1.1% nationally in 2025 (ABS), and a school in a suburb past its family-formation peak is fighting a birth-rate number, not a marketing number. For providers with international students, policy is currently the dominant variable — the Australian Government’s National Planning Level is 295,000 places for 2026, and the 2027 level was announced on 3 July 2026 at the same 295,000, so the 25,000-place rise that came with 2026 does not repeat next year. ASQA’s suspension on new CRICOS applications commenced 19 May 2026 and runs 12 months to 19 May 2027, blocking new CRICOS registrations and additions to scope for providers that lodge with ASQA as the National VET Regulator; government schools are exempt, and the ESOS agency for schools is the Secretary of the Department of Education rather than ASQA. Funding, registration and compliance questions belong with the regulator or the department — ASQA, the Department of Education, or your state education department — not with a marketing supplier, and any agency that answers a compliance question for you is a risk rather than a service.

What running the audit yourself actually costs

The audit above is genuinely doable in-house, and most registrars can run it. The honest cost is roughly a day of someone’s time to export and reconcile SIS and CRM records that usually disagree, a working knowledge of your enquiry data model, and then the recurring part — someone who reliably calls back every enquiry within minutes rather than days, including evenings and school holidays, which is where in-house follow-up quietly fails. That is the part that breaks at volume, not the spreadsheet.

Where an external partner is worth costing, judge it on the same funnel: enquiries created, calls actually booked, and enrolments at census. Our own model for education and student recruitment is pay-per-result — you pay on booked qualified appointments rather than a retainer — which exists precisely because the enrolment cycle is long enough that a retainer can bill through two quiet terms without a single student appearing.

Questions principals and registrars actually ask

How many intakes down does it take before enrolments are a trend, not a bad year?

Two consecutive intakes, on our rule. One intake down under 2% with flat enquiry volume is inside normal variation for most schools. The second consecutive fall is the one that justifies changing something structural, because by then the pattern has survived a full decision cycle rather than a single cohort.

Are school enrolments falling across Australia, or just at our school?

Neither, on the national data. Total Australian school enrolments grew 0.7% in 2025 to 4,160,918 students, but the growth was uneven: government schools fell 0.2% while non-government schools grew 2.3%, and between 2021 and 2025 independent schools grew 15.3% against a 0.4% fall in government schools (ABS, Schools, 2025). Compare yourself to your sector and state, not to the national headline.

Can marketing fix enrolments for the intake starting next term?

Only the part of it that is still deciding. Follow-up on live enquiries and on families who declined an offer can move the current intake; awareness work cannot, because in NSW selective placement, for example, families finalise school choices by 5 June for entry the following year (NSW Department of Education).

Our international student enquiries have dropped — is that policy or us?

Check policy first. The National Planning Level is 295,000 international student places for 2026, and the 2027 level was announced on 3 July 2026 at the same 295,000, so plan on flat national capacity rather than a further rise. ASQA’s suspension on new CRICOS applications runs from 19 May 2026 to 19 May 2027, so a provider that lodges with ASQA cannot add courses to its CRICOS scope in that window (ASQA); government schools are exempt, and the ESOS agency for schools is the Secretary of the Department of Education. Direct registration questions to your ESOS agency or designated State authority before you review any marketing.

What should we measure before we change anything?

Five ratios: enquiry to tour, tour to application, application to offer, offer to acceptance, acceptance to census. Whichever step moved most against last cycle is the only one worth funding this term. If none moved and enquiry volume fell instead, the problem is upstream and the current intake is already set.