• The trusted Australian LMS. Build your own learning. Only pay for active users!
STORE

|

LOGIN

Contact Sales

DEMO

Measuring training change and Kirkpatrick Level 3 Behaviour

Table of Contents

Reviewed by Tribal Habits learning specialists
This article is based on practical experience helping organisations measure training quality, learner response, and early feedback signals across onboarding, compliance, and workplace learning programs.

Published on: 17 September 2024
Last updated: 30 June 2026

Training has long worried about justifying its existence. How do we measure training success? What are the critical outcomes of training? How do we know if the training is any good and if it is making any impact? In the third in this series, let’s turn our attention to Kirkpatrick Level 3 behaviour – The degree to which participants apply what they learned during training when they are back on the job.

Most learning and development professionals will be familiar with The Kirkpatrick Model of evaluating the effectiveness of training. It’s a great model and serves as inspiration for the built-in reporting in the Tribal Habits platform. The model suggests four levels of training measurement.

  1. Reaction: The degree to which participants find the training favourable, engaging and relevant to their jobs.
  2. Learning: The degree to which participants acquire the intended knowledge, skills, attitude, confidence and commitment based on their participation in the training.
  3. Behaviour: The degree to which participants apply what they learned during training when they are back on the job.
  4. Results: The degree to which targeted outcomes occur as a result of the training and the support and accountability package.

Over a series of four articles, let’s examine The Kirkpatrick Model and how Tribal Habits can help any organisation with reporting on all four levels – automatically! We’ll see how organisations can select the appropriate level of reporting and utilise the information at each level to improve their training topics. And certainly, for any modern organisation, the ability to gather high-quality data that can demonstrate the learning and understanding outcomes from training is critical to supporting training budgets and initiatives.

Measuring training change and Kirkpatrick Level 3 Behaviour

What is Kirkpatrick Level 3 Behaviour trying to measure?

Now we get to the interesting measurements…where most training experiences fail to go! This is because most training experiences stop at the transfer of knowledge (Levels 1 and 2) and leave it to someone else to track any further outcomes. Perhaps many training experiences don’t want to know the results, as it may reflect poorly on the training!

Although, to be fair, often a Level 3 Behaviour measurement is not appropriate for every topic. Training involving regulations, technical knowledge or product knowledge may only require a Level 2 Learning measurement.

As a result, Level 3 Behaviour reporting is optional within Tribal Habits and should be considered carefully for any training program. Measuring behavioural change in training can be a complex and consuming task. It requires a mechanism to capture both quantitative and qualitative feedback from participants after they have left the training program – sometimes weeks or months later.

It requires commitment from all stakeholders – training provider, training admin, participant and the participant’s manager – to ensure enough data is collected and that the data is high quality.

Once you have behavioural change data, you must then do something with it. The training provider and training admin need to review the data for trends and outcomes. Perhaps the best use of behavioural data is by managers, who can use the data for 1:1 coaching with participants or to provide timely intervention when participants are struggling.

Measuring behavioural change through reflection and intent

First and foremost, asking participants to self-reflect and identify their own areas for change is a critical measurement of likely behavioural change. It is a massive leading indicator.

When a training participant is asked to compare their thoughts to others, identify areas of strength and improvement, and reflect on training ideas, their response is very telling. Participants who struggle to reflect are unlikely to see much behavioural change. If participants provide poor-quality reflections or no reflections at all, they are clearly not engaging in the training. If they cannot even identify interesting points or potential new ideas to implement, then no implementation will occur.

This is why training programs must ask participants to identify areas for change. When you simply give each participant the same ‘checklist’ of things to do, participants have not taken ownership of that list or envisioned how they may change. A better idea is to ask participants to build their own checklist of things to do or ask them to select key items from a larger list. Participants must take this first step towards behavioural change themselves.

Tribal Habits enables this process in two critical ways.

  1. Reflection questions. Creators can simply enable reflection questions in their topic, and participants will be automatically engaged with a series of reflection opportunities throughout the topic. Not only does this boost learning retention and commitment, but participants can also compare their reflections with those of their colleagues, providing a deeper level of self-reflection and awareness.
  2. Poll and Q&A answers. Creators can use polls (fixed response) and Q&A (free response) elements anywhere in their topic to ask explorers to provide a response to a question. Creators can also elect to have any of those responses copied to the participant’s journal. This allows creators to use polls and questions to ask participants to identify ideas for improvement, select next steps, highlight uncertainties or commit to action…and to have information recorded and available offline and for managers.
  3. Next steps. Creators can enable an optional final reflection question for participants, which asks them to write down their next steps regarding the topic. Once again, their answer is captured in their journal for later use.

In all cases, it’s easy to include reflection opportunities throughout a topic with just a few checkboxes. It challenges participants to think through the status quo and look for ways to change. It also creates a wealth of useful data for the firm on participant intentions.

Measuring behavioural change through on-the-job activities

The next step in measuring behavioural change is to move from intent to action. In this case, through on-the-job activity to demonstrate new behaviours.

For many training programs, this level of measurement falls outside the scope of the training. And that’s a shame. Demonstrating proficiency with a new skill or tool in a training experience – a controlled environment when the new knowledge is fresh – is very different from demonstrating proficiency in unpredictable real-world conditions at a later time.

For topics where behavioural change is important, setting up on-the-job activities for participants to demonstrate their new behaviours is the ultimate test. Measuring changes in behaviour is also often a great leading indicator of future improvements in business impact.

In Tribal Habits, creators can simply enable one or more on-the-job activities in any topic. These activities are unique to each topic. They can be designed to specifically test and measure new workplace behaviours related to that topic. Explorers have a deadline to complete the activity and must then report on their success and learnings, providing both quantitative and qualitative measures.

Activities can be very versatile. An activity might require the explorer to conduct a demonstration of a technique with their manager or another assessor. Or it might require them to build or implement a new tool or document in their role. Or perhaps download a case study and complete a technical assessment.

Good Kirkpatrick Level 3 Behaviour reporting promotes long-term change

Good Kirkpatrick Level 3 Behaviour reporting promotes long-term change

Most training programs that aim to improve business outcomes require some change in how people work. That means Level 3 isn’t optional for those topics — it’s the only way to know whether the training is doing anything useful beyond generating completion records.

The challenge is that behavioural data is harder to collect than reaction or learning data. It requires follow-up after the training ends, manager involvement, and a mechanism to capture what’s happening on the job. Without those things in place, organisations are left guessing whether the training translated into anything real.

Kirkpatrick Levels 1 - 4

Why this guidance matters

Kirkpatrick Level 3 is where training evaluation starts to move beyond course completion and knowledge checks into real workplace change. At Tribal Habits, we help organisations look at whether people are actually applying training on the job, not just finishing modules or passing assessments.

That matters because training only creates business value when people do something differently afterwards. For some topics, that change can be seen through self-reflection and intent. For others, it needs observable workplace activities, manager input, or follow-up evidence over time.

Kirkpatrick Level 3 Behaviour: Quick answers

What is Kirkpatrick Level 3 Behaviour?

Kirkpatrick Level 3 Behaviour measures whether learners actually apply what they learned once they are back on the job.

In simple terms, it asks whether training has changed what people do in real work situations. That is a very different question from whether they enjoyed the training or passed a quiz.

What does Level 3 actually measure?

Level 3 measures workplace application. It looks for signs that the learner has taken ideas, skills, knowledge, or commitments from training and turned them into observable action.

That might mean following a new process, using a tool correctly, changing how they communicate, improving judgment, or completing a task to a new standard. The key point is that the evidence sits in behaviour, not just in course activity.

When is Kirkpatrick Level 3 worth measuring?

Not every training topic needs Level 3 measurement.

It becomes most useful when the goal of training is to change how people work, improve consistency, reduce risk, strengthen decision-making, or lift performance in the workplace. For lower-risk or purely informational topics, Level 2 may be enough. But when behaviour change matters, Level 3 gives you much stronger insight into whether the training is actually sticking.

What are the best ways to measure Level 3 behaviour change?

The best Level 3 methods usually combine more than one source of evidence.

Useful options include:

  • Self-reflection and intent to identify commitment to change
  • Manager observation to confirm whether new behaviours are being used
  • On-the-job activities to test the application in real work conditions
  • Follow-up questions after training to capture progress and barriers
  • Practical outputs such as completed tasks, documents, demonstrations, or improvements

This aligns closely with the current article, which already focuses on reflection, intent, and on-the-job activities as practical Level 3 methods.

Example Kirkpatrick Level 3 evidence and follow-up questions

Here are a few practical examples of Level 3 evidence:

Self-reflection
What have you changed in your work since completing this training?

Manager follow-up
Have you observed this team member applying the process correctly in their role?

On-the-job action
Complete this workplace activity and report on the outcome, challenges, and learnings.

Observed behaviour
Has the learner demonstrated the expected behaviour in a real work situation?

Barrier check
What has made it easier or harder to apply this training in practice?

These questions are useful because they move the conversation beyond intention and into evidence of actual change.

Example Kirkpatrick Level 3 evidence and follow-up questions

How to interpret Level 3 data properly

Level 3 data becomes useful when you look for patterns across people, teams, and conditions.

For example:

  • If learners leave strong reflections but no real changes are visible later, intent may be high but workplace support may be weak
  • If one team applies the training well and another does not, the issue may sit with local leadership, systems, or priorities
  • If people understand the training but struggle to use it under pressure, the job context may need more reinforcement, tools, or coaching
  • If participants complete activities but behaviour still slips later, the training may need follow-up prompts or accountability over a longer period

This is where Level 3 becomes powerful. It helps you identify whether the problem is motivation, clarity, environment, manager support, or training design.

Common mistakes when measuring Kirkpatrick Level 3 Behaviour

1. Measuring intent but not action

It is useful when learners say they plan to change, but intention alone is not proof of behaviour change.

2. Following up too early or too late

If you check too soon, learners may not have had a chance to apply the training. If you check too late, the signal may be weaker and harder to interpret.

3. Relying on self-report alone

Self-reflection is useful, but it is stronger when supported by observation, outputs, or practical workplace evidence.

4. Measuring behaviour without manager support

Behaviour change often depends on reinforcement, feedback, and accountability after training, not just the training itself.

5. Collecting workplace evidence and doing nothing with it

Level 3 data only matters if it leads to better coaching, improved training design, or changes to support systems around the learner.

How Tribal Habits supports Kirkpatrick Level 3 measurement

The Level 3 features described in this article — reflection, polls and Q&As, next steps, and on-the-job activities — are all built into the Tribal Habits platform and available in any topic.

Reflection questions are enabled with a single toggle. Once on, Tribal Habits automatically generates reflection prompts throughout the topic, captures responses in each learner’s journal, and makes them available to admins and topic creators for review. Learners can also see how peers responded, which adds a comparison layer without any extra admin work.

Polls and Q&A elements can be placed anywhere in a topic and configured to copy responses directly to a learner’s journal — creating a record of intended next steps that managers can access and use in 1:1 coaching conversations.

On-the-job activities are set up by the creator with a task description, optional files, and a deadline. Tribal Habits then handles reminders, submission notifications to managers or assessors, and qualitative reporting on outcomes. All of this sits on the same platform as the topic itself — no separate tool, no manual follow-up coordination.

If you want to see how this works across a real training program, book a demo, and we’ll show you what Level 3 tracking looks like in practice.

Book a demo with Tribal Habits

FAQs about Kirkpatrick Level 3 Behaviour

What is the difference between Level 2 and Level 3?

Level 2 measures whether someone has learned something. Level 3 measures whether they actually used that learning back in the workplace. The current page already positions Level 3 as an application on the job after training.

Can self-reflection count as Level 3 evidence?

It can be an early indicator, especially when a learner identifies intended changes, barriers, and next steps. The current article already treats reflection and intent as important leading indicators, so this supports the existing page rather than changing it.

How long after training should Level 3 be measured?

It depends on the topic and the speed of application. In many cases, a follow-up after a few weeks is more useful than checking immediately after completion, because Level 3 needs time to appear in real work behaviour. The existing article also notes that data may need to be captured weeks or months later.

Does every training topic need Level 3 evaluation?

No. The current article already says Level 3 is not appropriate for every topic and may be optional for some training programs.

What makes good Level 3 evidence?

Good Level 3 evidence is tied to a real workplace behaviour, collected at the right time, and useful enough to support coaching, improvement, or intervention.

What should you do if behaviour is not changing?

Look beyond the course itself. Review whether learners had the chance to apply the training, whether managers reinforced it, whether systems supported the new behaviour, and whether the training gave clear enough examples of what good looked like in practice.

Final word

Kirkpatrick Level 3 Behaviour is where training starts to prove its real value. If people are not doing anything differently afterwards, it is very hard for training to create a meaningful business impact.

That is why Level 3 matters so much. It helps you move from learning in theory to change in practice, and from course completion to workplace improvement.

Further Reading