
General Manager

I would be cautious about celebrating an improved simulation score if the person returns to a workplace where they have no opportunity to use what they practised. The surrounding job can make the intended behaviour possible, irrelevant or difficult to attempt.
Practice-Native Learning is Altaius's methodology for putting realistic decisions, consequences and feedback at the centre of learning design. Taking that design into work means asking what happens after the practice ends.
Blume and colleagues' meta-analysis examined 89 empirical studies of training transfer. It found positive relationships with factors including motivation and a supportive work environment, and showed that measurement choices affect the reported relationships. The study does not supply a universal percentage of training that transfers or fails to transfer. [1]
I would therefore avoid claims such as “most training is wasted” unless a specific source and definition justify them. The more useful question is which conditions this programme needs and whether they are present.
Imagine a manager practising how to clarify an ambiguous customer request before committing resources. The transfer plan should identify a real upcoming conversation in which that behaviour is appropriate. It should also identify what the manager is allowed to change and who can support them.
If no such opportunity is likely during the evaluation period, a lack of observed application may be unsurprising. That is different from demonstrating that the learner cannot use the skill.
Here is the short transfer record I would prepare with the learner and their manager:
This is a proposed working record, not an instrument validated by the cited review.
In an illustrative example, a manager asks the clarifying questions they practised, but the customer still refuses the proposal because its price exceeds the budget. The business outcome was unfavourable. It does not follow that the target behaviour failed to occur.
The reverse can also happen: the customer accepts despite a poorly considered commitment. A favourable result does not by itself prove sound judgement. The evaluation needs enough context to distinguish how the decision was made from the outcome that followed.
I would record whether the opportunity arose, whether the behaviour was observed and what happened afterwards. Combining these into a single success label would discard information needed to improve the programme.
Support should be concrete and proportionate. A manager might agree an appropriate opportunity, observe a specific part of the work and discuss it afterwards. They do not necessarily need access to every simulation conversation or a new administrative dashboard.
Agree what information is shared, with whom and for what development purpose. Let the participant explain relevant context. Avoid turning a learning experiment into an undisclosed appraisal process.
Where feasible, use observations that are not entirely dependent on the participant's own impression of improvement. A self-report can be useful, but it should be labelled as such. Compare like periods and explain changes in work demands, team support or the cases encountered.
If the learner could not apply the behaviour because the approval process was unclear, another simulation alone may not fix the problem. The organisation may need to clarify authority. If the learner had an appropriate opportunity but missed the cue, a new practice case can address that difficulty.
This is where workplace evidence becomes useful for learning design. It tells us whether to adjust the scenario, provide more support or change conditions outside training. I want the programme to be able to make that distinction before claiming impact.
Read the Altaius definition of Practice-Native Learning
Previous in this series: Spacing and Replay in Practice-Native Learning