A single LLND result isn't a support plan: screening skill by skill

Every RTO screens language, literacy, numeracy and digital skills before a learner starts. The Standards for RTOs require it: you identify a student's support needs, and then you meet them. Most providers treat the screening itself as the obligation. It isn't. The obligation is the response, and a response is only ever as good as what the screening actually told you.

That's where most screening falls down. It hands back a single result: one "ACSF level," or a pass/fail. Enough to drop in a file. Almost useless for acting on.

A single number can't tell you what support to put in place

The Australian Core Skills Framework was never one number. It describes five core skills, Learning, Reading, Writing, Oral Communication and Numeracy, each set out across five levels of performance. A learner can read a workplace procedure confidently and lose the thread the moment they have to write a short report, or work out a measurement on the floor. Roll all of that into "Level 2" and you've thrown away the only information that tells you what to do next.

Digital skills belong in the same conversation now. Once delivery moved online, a learner who can't navigate the portal, an email, or an online assessment is a learner at risk, regardless of how well they read. A screening that doesn't look at digital capability is measuring the wrong decade.

What the Standards actually ask for

There are two ways to get this wrong, and they pull in opposite directions.

The first is to treat LLND as a pass/fail exam that locks a student out on a number. That isn't what the Standards intend, LLND screening is a pre-training review to surface support needs, not a barrier to keep people out.

The second is the mirror image: screen, file the result, and enrol everyone regardless. ASQA is just as wary of a provider that signs up a learner it has no realistic way to support, capacity to complete is part of the picture, and "we ran the check" is not the same as "we acted on it."

The balance the Standards expect sits between the two. Where a learner is below the level a course demands, you put support in place. Where the gap is wide enough that reasonable support can't close it, you make an informed decision with the student rather than taking the fee and hoping. You can only strike that balance if you can see, per skill, how far below the line each student actually sits. A single figure can't drive either response, because it never says what support.

Per-skill, against the level the course demands

LLND screening in RTObase is a short check drawn from a validated, ACSF-mapped question bank: Reading, Writing, Numeracy, Oral Communication and Digital Literacy, each measured against the level the course expects, with Learning available as a sixth. It's marked instantly against a fixed answer key, with no AI generating or grading anything, so the result is reproducible and the same input always produces the same output an auditor can check.

Two skills are handled with care rather than overclaimed. Writing and Oral Communication are screened by recognition proxy, choosing the correctly spelled or punctuated option, the correct meaning, rather than capturing free text or recording speech, and they're flagged indicative for a trainer to confirm. No speech is stored; nothing is graded by a model.

Update (July 2026): Oral Communication is no longer screened by written proxy. Students now complete a short recorded speaking task in the portal, answering spoken prompts in their own words, and RTO staff listen and rate the recording against the ACSF. The rating is always made by a person, never a model, and recordings are stored securely, visible only to your organisation, as the evidence behind the rated level. Writing remains a recognition-proxy screen flagged indicative.

The output isn't a verdict. It's a per-skill reading against expectation. Where a skill lands below the course's expected level, the enrolment carries a clearly-labelled support concern your team records against and acts on, the difference between "this learner needs help with numeracy" and "support" in the abstract. Screen a learner once and a later course that expects the same skills reuses the result; only the skills a harder course demands more of get topped up, and the reuse is recorded as evidence on the new enrolment.

Why this follows from one record

A standalone screening tool gives you a PDF. Then the support has to happen somewhere else, a note in the student management system, a reminder in someone's inbox, a follow-up nobody owns. The screening and the response live in different places, which is exactly how an obligation quietly goes unmet.

Because the screening, the enrolment, the trainer and the assessment all sit on one record, a below-level skill becomes a support action on the enrolment itself, visible to whoever picks the student up next, with the evidence frozen against the enrolment: which bank, which items, which level, on what date. Identify and respond stop being two systems with a gap between them and become one workflow with a trail.

Most RTO platforms don't screen for any of this, the check, where it happens at all, is a separate tool and a separate file. The case for doing it on the record is the same as the case for the record itself: an obligation you can see is an obligation you can actually meet.

If that's how you'd like to run LLND screening, come and have a look.