Competency-Based Training Examples: 8 Worked Examples With the Standard and the Evidence
Eight competency-based training examples, each with the task, the standard and the evidence used to judge it. Start with a real Australian unit, then adapt the workplace illustrations.

Competency-based training is training where a person is signed off only when they show they can do a defined task to a stated standard, judged on evidence, rather than when they have finished a set number of hours or modules.
- Every usable example names three parts: the competency as an observable task, the written standard and the evidence an assessor sees or collects.
- "Not yet competent" is a result with a next step, such as more practice and another attempt, not a final fail.
- Test any example against the four rules of evidence: is the evidence valid, sufficient, authentic and current?

Set up your branded LMS and deliver training with AI
I've been in enough kickoff meetings with L&D managers to recognise the moment someone says "we want this to be competency-based" and means "we want a quiz at the end". The course gets built, people pass the quiz, and six months later nobody can say what "competent" looked like for that job.
Competency-based training is training where a person is signed off only when they show they can do a defined task to a stated standard, judged on evidence, rather than when they have finished a set number of hours or modules.
So a competency-based training example is only an example when you can see three things: the competency written as something you can watch someone do, the standard it is judged against and the evidence that proves it. I start from a real Australian unit of competency, HLTAID011 Provide First Aid, and write seven workplace examples after it in the same format so you can copy them.
I'm an instructional designer at Coursebox, where we build an AI course builder for registered training organisations, and I test every release. Near the end I build the first aid unit in our product and show what it got right, what it got wrong and what I fixed.
What is competency-based training?
Competency-based training is training where a person is signed off only when they show they can do a defined task to a stated standard, judged on evidence, rather than when they have finished a set number of hours or modules.

The idea is older than corporate L&D. In medical education, Jason Frank and colleagues describe competency-based education as organised around competencies, with less emphasis on time-based curricular design. Hours still happen, but they stop being the thing you certify.
The term is used in two senses:
- The formal vocational system. In Australia and several other countries, competency-based training is how national vocational qualifications work. The standard is a published unit of competency, and the training provider does not get to change it. Our guide to the unit of competency explains how Australian units are structured.
- Corporate L&D. Inside a company, competency-based training usually means you write the standard yourself, for a task specific to your team.
The logic is the same in both. What changes is who owns the standard. In Australia, the formal version is how a whole national system certifies people.
According to NCVER's 2025 figures, fee-for-service short courses in that system are mostly workplace compliance training, including first aid. Every unit in it is assessed against a published standard, which corporate programmes often skip.
Competency-based vs traditional training
| Question | Time/completion-led emphasis | Competency-led emphasis |
|---|---|---|
| What organises progress? | A scheduled course or sequence, with its assessments | Demonstrated achievement of the required competencies |
| What gets recorded? | Attendance and completion, plus assessment evidence where required | Judgements against the standard and the evidence supporting them |
| What happens when evidence is insufficient? | Additional support or reassessment should follow the programme rules | Record the gap, provide support and collect the missing evidence |
| Who sets the standard? | The institution, regulator or organisation | The published unit or the organisation, as applicable |
A quiz can provide knowledge evidence in either approach. Where the standard requires practical performance, it must be assessed through a suitable task or observation as well. Both time-based and competency-based programmes can use clear standards and careful assessment; the difference below is their organising emphasis.
The three parts every example needs
I write every competency in three parts. This is my working method for writing examples, not a rule from any standard, but it follows the way Australian units are built.
- Competency: the task, written as something you can watch someone do or see in what they produce.
- Standard: the written criteria and conditions the performance must meet. In a regulated unit, use its published requirements; in a workplace example, agree the standard with the people responsible for the task.
- Evidence: what the assessor sees or collects to judge the task against the standard.
The competency line is where programmes usually go wrong. "Communicates effectively" cannot be observed. "Summarises the agreed actions and confirms an owner for each before closing the meeting" can. Written that way, a "soft skill" can be assessed as directly as CPR.
The evidence line is where the quiz habit hides. Requests typed into our own course builder show how strong that habit is.
The AI classifier flagged 11 requests as asking learners to perform a task. When I read them, several were assignments with no criteria for judging them.
A completion record or attestation does not establish practical performance. Use knowledge questions for knowledge and suitable work samples or observation for performance, following the applicable standard. Record each requirement as met, not yet met or not observed; use not applicable only where the standard permits an exclusion, with the reason. Required evidence cannot be waived or hidden by a high average.
8 competency-based training examples
Example 1 is built from a real published unit. Examples 2 to 8 are mine: illustrations in the same format, not programmes from any company, so you can copy them and change the standard to fit your team. Each uses the same five labels: the three parts above, plus what "not yet competent" looks like and when to reassess.
1. First aid: HLTAID011 Provide First Aid
HLTAID011 is the Australian unit for providing a first aid response, used in workplace and community first aid courses. The unit on training.gov.au sets out four elements, which describe the outcomes, and 17 performance criteria, which describe the performance needed. Element 1, "Respond to an emergency situation", has four criteria:
- 1.1 Recognise and assess an emergency situation.
- 1.2 Ensure safety for self, bystanders and casualty.
- 1.3 Assess the casualty and recognise the need for first aid response.
- 1.4 Seek assistance from emergency services.
Element 2 opens with 2.1: perform cardiopulmonary resuscitation (CPR) in line with the guidelines of the Australian Resuscitation Council (ARC).
- Competency: Provide a first aid response, including CPR, as described in HLTAID011.
- Standard: All 17 performance criteria, plus the unit's assessment requirements and the ARC guidelines.
- Evidence: Among other items, at least 2 minutes of uninterrupted single-rescuer CPR on an adult manikin on the floor, at least one shock from an automated external defibrillator (AED) delivered by following its prompts, and at least one simulated incident where the candidate does not know the casualty's condition beforehand.
- Not yet competent looks like: The CPR is interrupted before the 2 minutes are up, or the AED prompts are not followed. A strong written answer does not make up for either.
- Reassess by: Arrange another assessment after the unmet requirements have been practised, before recording competence. Separately, follow the refresher and currency requirements that apply to the role and jurisdiction.
Those evidence items come from the HLTAID011 assessment requirements, which also list infant CPR and a range of injuries and illnesses. A multiple-choice quiz cannot produce any of them. For keeping the skill current after the first sign-off, see our guide to verification of competency.
2. Communication: closing a meeting with agreed actions
- Competency: A team leader summarises the agreed actions and confirms an owner for each before closing the meeting.
- Standard: Before the meeting ends, every action is read back with one named owner and a due date, and the written summary sent afterwards matches what was agreed.
- Evidence: Observation of one live meeting the learner chairs, plus the written action summary they send afterwards.
- Not yet competent looks like: An action left with no owner ("someone will look into it"), or a summary that lists actions nobody agreed to.
- Reassess by: The next meeting they chair after feedback.
I ask for both pieces of evidence because a summary that sounds right in the room can drift by the time it is written down.
3. Customer service: handling a complaint call
- Competency: A contact centre agent handles a complaint call so the caller knows what happens next.
- Standard: The caller hears three things: the complaint acknowledged, the facts confirmed back to them, and what happens next and by when.
- Evidence: Two recorded or role-played calls, scored against the same three-item checklist, using scripts like our customer service role play scenarios.
- Not yet competent looks like: One of the three items missing on either call.
- Reassess by: Two more calls after coaching on the missing item.
4. Sales: handling a price objection
- Competency: An account manager responds to a price objection by finding out what the buyer is comparing the price with before offering anything.
- Standard: The rep asks at least one question about the buyer's comparison or budget, restates the value in the buyer's terms and offers no discount they are not authorised to give.
- Evidence: A role play with a colleague playing a scripted buyer, plus one real call reviewed with the rep's manager.
- Not yet competent looks like: A discount offered before any question is asked.
- Reassess by: The next scheduled call review.
Giving every rep the same scripted buyer, as in scenario-based learning, is what makes their results comparable.
5. Leadership: delegating a task
- Competency: A new manager delegates a task with a brief the other person can act on without coming back to ask.
- Standard: The brief states the outcome, the deadline, the person's level of authority (decide, recommend, or check first) and the check-in point.
- Evidence: The written brief, plus the delegate's own account of what they were asked to do. The two must match.
- Not yet competent looks like: The delegate's account leaves out the deadline or the authority level.
- Reassess by: The next two tasks the manager delegates.
Part of the evidence comes from the other person, because a brief can look complete on paper and still not land. If you are unsure whether delegation needs training or coaching, our comparison of coaching, training and mentoring can help you decide.
6. Healthcare: hand hygiene at the right moments
- Competency: A nurse cleans their hands at each of the five moments for hand hygiene during patient care.
- Standard: The WHO "Your 5 Moments for Hand Hygiene": before touching a patient, before a clean or aseptic procedure, after a body fluid exposure risk, after touching a patient and after touching patient surroundings.
- Evidence: Direct observation during a shift, with each moment marked met, not yet met, not observed or not applicable.
- Not yet competent looks like: An observed moment that is missed.
- Reassess by: A further observed shift, then your health service's own audit schedule.
Naming the five moments on a quiz shows that someone knows them, not that they clean their hands at those moments during a shift.
7. Compliance: spotting and escalating a privacy breach
This example uses a hypothetical policy: staff must log any suspected privacy breach with the privacy officer on the same working day, recording what was sent, to whom and when. Your own policy and the law that applies to you set the real deadline.
- Competency: An office worker recognises a possible privacy breach and escalates it under the organisation's policy.
- Standard: The breach is identified and logged with the privacy officer on the same working day, with what was sent, to whom and when.
- Evidence: A scenario where the learner is told an email with customer details went to the wrong recipient, and completes the actual log entry.
- Not yet competent looks like: A late log, or one that leaves out what was sent or to whom.
- Reassess by: After feedback and further practice, before treating the person as competent on this task. Continue later checks under the organisation's compliance schedule and after policy changes.
Our guide to LMS compliance covers keeping the records that prove this kind of sign-off.
8. Technical: logging a reproducible bug report
- Competency: A support engineer logs a bug report another person can reproduce from the steps alone.
- Standard: The ticket gives the environment, numbered steps, the expected result and the actual result.
- Evidence: A real ticket that a colleague who has not seen the bug reproduces using only the ticket.
- Not yet competent looks like: The colleague has to ask a question before they can reproduce it.
- Reassess by: The next three tickets the learner logs.
A one-page card with the same five labels as the examples: Competency, Standard, Evidence, Not yet competent looks like and Reassess by. It includes the filled HLTAID011 card, the filled meeting-actions card and a blank card, plus a judgement record that marks each part of the standard met, not yet met, not observed or not applicable. It is a planning aid with no score, in an editable Word file, and no email is required.
How to check any example: the four rules of evidence
Australia's training regulator, ASQA, sets four rules of evidence that every judgement of competence must meet. They are in its 2025 practice guide on assessment, and they work as a test for any example, in any organisation. Here they are, each applied to the complaint call in example 3.
- Validity: the evidence lets the assessor be "reasonably assured" the person has the skills described. Example 3: a call scored on the three items shows the skill; a quiz about complaint handling would not.
- Sufficiency: the quality, quantity and relevance of the evidence must support an informed judgement. The two calls in example 3 are an illustrative starting sample, not a validated minimum. The assessor needs enough varied evidence for the standard and may need more calls.
- Authenticity: the evidence is the person's own original work. ASQA's guide now names AI, asking assessors to confirm evidence "has not been plagiarised or generated with artificial intelligence (AI) tools". Example 3: a recorded call in the learner's own voice, not a written script that anyone or any tool could produce.
- Currency: the evidence shows the person's "current skills and knowledge". Example 3: calls from this month, not a recording from last year's induction.
The same test catches problems with the first aid unit. If a learner uploads a video of CPR on a manikin, the assessor still has to know whose hands are on the manikin and when it was filmed. That is one reason to observe CPR live.
Benefits of competency-based training
- Training time follows the learner: someone who can already do the task gets signed off sooner, and someone who cannot gets more practice instead of a pass.
- A record you can show: each sign-off points to the evidence behind it, which is what an auditor or a client asks for.
- Managers see who is not there yet: a not yet competent result names the gap, so the next session targets it.
One documented example of the first benefit comes from surgical training.

Case study: Competency-based orthopaedic surgery training in Toronto
Surgical residency has traditionally been measured in years: a resident trains for a fixed period, rotating through placements, and finishes at the end of it. In July 2009, the Division of Orthopaedic Surgery at the University of Toronto started a pilot residency programme built on a competency-based framework instead.
- What they did: The programme was split into modules, and according to the eight-year review published in Medical Teacher it "dramatically intensified both the structured learning elements and the assessment processes". The team later refined the curriculum through curriculum mapping.
- Evidence and result: Feedback from residents, faculty and accreditation bodies was positive. The authors say shortening training was not the goal, but the modular, competency-based programme "resulted in shortened time to completion of residency training for some learners".
- Limitation: One programme in one specialty, reported by the people who ran it, with no comparison group in the abstract and no numbers for how many residents finished early. It shows the model can work at that level, not that it will save time in your programme.
- Lesson: When progress depends on evidence rather than the calendar, some people finish sooner. The cost is more assessment, so budget for assessors before you promise anyone a shorter course.
Limitations of competency-based training
- Writing standards takes time. Every task needs a standard and an evidence plan before content is built, which is slower at the start than writing slides.
- Assessors need calibration. ASQA's reliability principle asks assessors to interpret evidence consistently. Compare judgements on representative work, discuss disagreements and refine ambiguous criteria before relying on the results.
- Some people will not get there. ASQA's guide says some learners need more tasks to show competence and that some "may not be able to achieve competence despite repeated opportunities". Decide in advance what happens then: more support, a different route or a different role.
How to implement competency-based training
- Pick the task from a gap. Start with a task people get wrong at work, not a topic. A training needs analysis questionnaire and a skills matrix show where the gaps are, and competency mapping links tasks to roles.
- Write the competency card. Fill in the five labels before you build anything, starting with the standard.
- Decide the evidence before the content. Choose what the assessor will see: an observation, a work product, a recording or a scenario.
- Build the explanation and practice needed for the whole task, including prerequisites and reasons for the required actions. Criterion-referenced instruction aligns assessment with the performance standard; it does not mean stripping away useful context. If you draft a rubric with an AI rubric generator, check each required action separately and prevent an average from concealing an essential failure.
- Assess, record, reassess. Use practice checks along the way and a judgement at the end, as our guide to formative vs summative assessment explains. Record not yet competent results with the gap named, and book the reassessment.
Notes from testing: building HLTAID011 in Coursebox
I wanted to see whether our course builder keeps the evidence requirements of a real unit or turns them into a quiz, so I built one course in a test workspace.
What I asked for. I used the Create RTO Package option, which builds a course from a unit's page on training.gov.au, and pasted the HLTAID011 link. The settings came prefilled with a Certificate IV supervisor profile (Australian Qualifications Framework Level 4), although the first aid unit has no AQF level of its own. I changed the assessment types to practical observation, scenario and written knowledge, kept the rest, and built the course. Generation took more than 8 minutes.
What it produced. Eight sections and 28 lessons. Each of the 17 performance criteria had its own lesson, in unit order from 1.1 to 4.2, which I checked against the training.gov.au unit page. The knowledge questions sat in quizzes with an 80% pass mark. The performance evidence was assessed by practical tasks, not quiz questions: the "CPR and AED resuscitation" assessment asked for single-rescuer adult CPR "for at least 2 minutes (5 cycles...) on the adult manikin placed on the floor", at least one AED shock, a two-rescuer rotation and infant CPR on a firm surface. Another assessment opened with "Respond to a simulated workplace first aid incident".
What was wrong. Every practical was marked with a weighted percentage rubric. The CPR page used "Practical Skills 40%", "Teamwork and Communication 30%" and "Knowledge and Justification 30%", with the written answers scored inside the same rubric, so a learner could reach a passing total while weak on the CPR itself. The other four assessment pages used the same style. I found no mapping document: nothing in the course editor mapped each performance criterion to an assessment task by number. The task wording tagged elements, not criteria. One task asked the learner to state "ETA requests" when calling emergency services, which is not in the unit, and a wellbeing task asked for a report of 900 to 1,200 words, which is heavy for this unit. I did not open the knowledge quiz items, so I cannot say whether any criterion is covered only by multiple choice.
The fix, and how I checked it. I sent one message to the AI editor on the CPR page: replace the weighted rubric with an observation checklist, one Satisfactory or Not yet satisfactory row per observable action, every row needing Satisfactory, the written task marked separately, replacing rather than adding. The editor replied that it had made the change. A reply is not a check, so I reloaded the page and read the rubric. It now had seven action rows (adult CPR, infant CPR, AED operation and shock, safe AED use, response to regurgitation or vomiting, rescuer rotation and verbal handover) and an "Overall practical result" row: "Every observable-action row above must be marked Satisfactory for an overall Satisfactory result. Use no percentages and no averaging." The 40/30/30 rows were gone.
The fix has limits. The block is still a percentage-type rubric: the header says "Percentage", and each cell is cut off at "Satisfactory / Not-y". The wording now asks for an all-actions judgement, but I did not test how the system calculates or reports it. Nor did I check separate marking of the written task. I changed one of five assessment pages; the others still use weights, and the "ETA requests" line and long report remain. Without a mapping document, I could not verify full assessment coverage. What I verified is the new checklist wording on one page, not a completed competency-assessment workflow.
Where to start
Competency-based training connects a defined task, an agreed standard and suitable evidence. A quiz can assess relevant knowledge, but it cannot stand in for a practical demonstration where the standard requires performance. Check what a sign-off actually certifies.
This week, pick one task your team gets wrong and fill in a competency card, starting with the standard and evidence. Use the four rules to examine that evidence. If the task requires practical performance and you have only a quiz, add a suitable demonstration before treating it as a sign-off.
Frequently Asked Questions
Competency-based training is training where a person is signed off only when they show they can do a defined task to a stated standard, judged on evidence, rather than when they have finished a set number of hours or modules.
The Australian unit HLTAID011 Provide First Aid is one. A learner is signed off only after showing evidence such as at least 2 minutes of uninterrupted CPR on an adult manikin on the floor and an AED shock delivered by following its prompts, not after finishing a set number of hours.
Progress depends on evidence judged against a written standard, not on time spent in training. In Australia, that evidence must meet four rules: valid, sufficient, authentic and current. To make that workable, I write each competency in three parts: the observable task, the standard and the evidence.
Not yet competent is the result an assessor records when the evidence does not yet show the person can do the task to the standard. It is a result with a next step: more practice on the gap and another attempt. ASQA notes that some learners may not achieve competence despite repeated opportunities.
A time-led programme organises learning around a schedule or sequence; a competency-led programme organises progress around demonstrated achievement. Both can use clear standards and careful assessment. Attendance or a quiz alone cannot replace practical evidence where the standard requires it.

Carolina Martin
Carolina is a highly experienced learning designer with tech start-up experience. She is based in Sydney and previously designed courses alongside Coursebox founder Travis in a previous job. She is passionate about delivering engaging learning experiences and brings the same perspective to customer experience too. She manages our customer support documentation and product testing processes.


