“Our RTO student numbers have dropped” — the causes, in order of how often they’re the real one

If your RTO’s starts are down, check the sector baseline before you change anything. Government-funded VET students fell 6.8% nationally in January–June 2025 — 866,055 students, down 63,320 on the same period a year earlier. Below that, you are tracking the market. Above it, the cause is almost always one of four, and each has a test that isolates it.

• Cause 1 — funding, scope or policy changed. Test: split funded from fee-for-service starts, same intake month, year on year.
• Cause 2 — a competitor took your qualification and postcode. Test: count RTOs with that qualification code on scope in your state on training.gov.au.
• Cause 3 — enquiries held and contact rate fell. Test: median minutes from enquiry to first human contact.
• Cause 4 — the course mix stopped matching demand. Test: starts by qualification code, same period year on year.

“Our RTO student numbers have dropped” — is it us, or is it the sector?

Answer this before you touch the marketing budget. NCVER’s Government-funded students and courses, January to June 2025 recorded 866,055 students in government-funded training nationally, a decrease of 63,320 or 6.8% on the same period in 2024, with every state and territory down except South Australia, which rose 2,125 (3.8%) to 57,645 (record and abstract, NCVER via ERIC).

Two rules make the comparison honest. First, compare the same intake month year on year, never month on month — VET enrolment is seasonal and a February-to-May fall is a calendar artefact, not a trend. Second, run funded and fee-for-service cohorts as separate lines, because they respond to completely different forces. An RTO down 5% year on year in a national market down 6.8% does not have a marketing problem; it has a market, plus a rounding error.

The threshold worth writing on the wall: one soft intake is noise, two consecutive intakes below the same period last year is a trend. Act on the second one, not the first.

The four causes, in order of how often they’re the real one

This is the four-cause start check. The ordering is our own, from the RTO pipelines we work on — it is a rule of thumb for the order you should test in, not a measured distribution, and we have no published sample behind it. What is not a rule of thumb is the test column: each row can be settled with data you already hold in your student management system or that is public on training.gov.au.

Cause 1: funding, scope or policy changed — the part marketing cannot fix

Start here because it is the objection every RTO owner raises and it is frequently correct. If your funded allocation was cut, a subsidy rate moved, an eligibility rule tightened or a qualification came off your scope, no campaign restores those starts. A marketing budget aimed at a cohort that is no longer subsidised buys enquiries that cannot enrol.

Two policy facts are worth knowing before you diagnose an international shortfall as a marketing failure. ASQA has suspended new CRICOS provider registration applications and applications to add new courses to an existing CRICOS registration; the suspension commenced 19 May 2026 and runs 12 months to 19 May 2027, with published exemptions for certain provider types, training products already on scope, and new delivery locations for an already-approved course. Separately, the Department of Education has set a National Planning Level of 295,000 for 2026, 25,000 higher than 2025, and on 3 July 2026 announced that the 2027 National Planning Level is also 295,000, unchanged, with no active provider receiving a lower allocation in 2027 than in 2026. The planning level is not tightening. The practical consequence for an existing CRICOS provider is narrow and specific: growth cannot come from widening your course offer this year, so it has to come from converting the enquiry volume you already generate.

This is general information, not compliance advice. Registration, scope and marketing-claim questions go to ASQA; funded-place questions go to your state body — in Victoria that is Skills First, in New South Wales Smart and Skilled. If the funded side is where your volume went, the practical question becomes which eligible cohorts you are not reaching, which is the subject of our page on government-funded RTO lead generation in Australia, and what you may and may not say in the ads, which is covered on ASQA-compliant marketing for RTOs.

Cause 2: a competitor took your qualification and your postcode

Competitor entry looks identical to a conversion problem from inside the office — the phone is quieter — but it separates cleanly in the data. In competitor entry, enquiry volume falls and your enquiry-to-start conversion rate holds. In a conversion problem, enquiry volume holds and conversion falls. You cannot have both explanations at once.

The public test takes about twenty minutes. Search your qualification code on training.gov.au, list the RTOs with it on scope in your state, and compare that list with twelve months ago. Then check paid-search impression share for your three highest-intent course terms. If the RTO count rose and your impression share fell, you are in an auction you were previously winning by default.

Cause 3: your enquiries held and your contact rate fell

This is the cause that hides, because nobody reports on it. An RTO can lose a quarter of its starts without losing a single enquiry, purely because the median time from web form to first human contact drifted from minutes to the next business day. Here is the arithmetic, with placeholder numbers — substitute your own from your student management system.

Enquiry volume is down 3% and starts are down 23%: the fall entered the funnel at one stage, contact rate, 67% to 52%. Restore contact rate alone and hold every downstream rate exactly where it is today: 408 × 67% = 273 reached, × 42.9% consult rate = 117 consults, × 51.6% start rate = 60 starts. That is 47 starts to 60 — thirteen more, on the same enquiry spend. Nothing in that calculation requires a bigger budget, which is why it is the first thing to test and the cheapest thing to fix.

Across the campaigns we run, speed to lead on its own is worth roughly 3x, and doubling contact rate roughly 2x. The honest wrinkle: those numbers do not multiply. 3x × 2x is not 6x, because fixing speed to lead is part of how contact rate improves — they overlap heavily. This is our own operator experience from client work, not research, and there is no published sample behind it.

Doing it yourself is entirely possible and worth costing honestly. It means someone rostered to call every new enquiry within minutes across 8am–8pm weekdays and Saturday mornings, because that is when people research courses; a documented follow-up sequence of at least six attempts across phone, SMS and email; and a rule that intake peaks never compete with student-support duties for the same person’s time. At 400 enquiries a month that is roughly 20 first calls a day plus the follow-up backlog, and it breaks at exactly the moment it matters most — the fortnight before census. That is the workload our RTO lead generation service automates on a pay-per-result basis, where the charge is on booked qualified enrolment calls rather than a retainer, and what those calls tend to cost by qualification level is set out in our cost per enrolment call benchmarks for Australian RTOs.

Cause 4: the course mix stopped matching demand

Run starts by qualification code for the same period year on year and rank the deltas. If two or fewer qualifications account for most of the total fall, the RTO does not have a pipeline problem — it has two courses that stopped selling, and the other qualifications are quietly holding. Mix problems are commonly mistaken for marketing problems because the aggregate number falls at the same rate either way; the per-code table is what tells them apart, and it takes one query.

It is ranked last because it is the slowest to act on: replacing demand for a fading qualification means an addition to scope, trainer capacity and an employer story — a twelve-month project, not a campaign.

What to do in the next 24 hours, then the next 7 days

In the next 24 hours, and none of it costs anything: pull starts for the same intake month year on year, split funded from fee-for-service; pull median minutes from enquiry to first human contact for the last 90 days; pull the percentage of enquiries never reached at all. Those three numbers eliminate two of the four causes immediately.

In the next 7 days: run the training.gov.au scope count for your top qualification, run starts by qualification code, and call every enquiry from the last 30 days that was never reached — an unworked enquiry from three weeks ago is a warmer contact than anything you will buy this month. Then fix the one cause your tests confirmed and re-measure at the next intake, not next week — changing four things at once destroys your ability to tell which one worked.

Frequently asked questions

How do I know if our RTO enrolment drop is a bad term or a real trend?

Compare the same intake month year on year, twice. One soft intake is noise; two consecutive intakes below the same period last year is a trend. Check it against the sector baseline before concluding it is you: NCVER recorded government-funded VET students down 6.8% nationally in January to June 2025, to 866,055 (NCVER, Government-funded students and courses, January to June 2025; figures quoted from the ERIC record of that publication, where they can be read in full).

Are declining enrolments an ASQA problem?

Falling student numbers are not in themselves a compliance matter, but what you do in response can be. Scope changes, marketing claims and CRICOS registration are all regulated, and questions about them go to ASQA directly rather than to a marketing agency. This page is general information, not compliance advice.

Can we add a new CRICOS course to grow international starts right now?

Generally not during the current suspension. ASQA states that from 19 May 2026, for 12 months to 19 May 2027, new CRICOS provider registration applications and applications to add new courses to an existing CRICOS registration are suspended, with published exemptions including certain provider types, training products already on the provider’s scope, and adding a delivery location for an already-approved course (ASQA, Suspension on new CRICOS applications). Check your own circumstances against the exemptions with ASQA.

How much of a decline is normal for an RTO right now?

There is no normal for a single provider, but there is a national reference point: a 6.8% year-on-year fall in government-funded students for January to June 2025, with South Australia the only jurisdiction to grow, up 3.8%. If your fall is materially steeper than your state’s, the gap is the part worth investigating.

Will spending more on marketing fix declining enrolments?

Only if the test says the cause is enquiry volume. If enquiry volume held and contact rate fell, more spend buys more enquiries that also go unreached, and cost per start rises. Fix the conversion stage first, then buy volume — in that order, because the second one multiplies whatever the first one is worth.