People come back from the course. Does anything change in how they work, and what decides whether it does?

Where outcomes are measured objectively, training on its own rarely moves them. A Campbell Collaboration review of workplace safety interventions brought together 100 studies: 16 randomised trials, 30 controlled before-and-after studies and 54 with repeated measures. Most judged results by injury rates rather than by what people said, and about a third were rated high quality. Its summary of training is blunt: behavioural approaches, "such as general coaching and feedback or safety training, are less effective. This does not mean that safety training is not relevant, but rather it is ineffective in the absence of other efforts." What worked better was changing the work itself, and combining measures at and across levels of the organisation. Two limits matter. This is safety training, not workplace training in general. And within the review, training has only two studies of its own, both rated high quality: from them the authors conclude that there is limited evidence of no effect shortly after training, and a moderate effect at medium-term follow-up, which in the one study that found it meant Swedish telephone repair workers two years on. Read this as a caution against expecting a course by itself to change results, not as a measure of how much training helps. (Read in library)

Support from colleagues and supervisors predicted whether people applied and shared what they learned. A survey of 541 German-speaking employees who had been trained in the previous two years found that support from both supervisors and peers predicted two things after training: applying it, and sharing it with colleagues. Motivation and volition carried those effects. Both applying and sharing were linked to a two-item self-rating covering people's own performance and job satisfaction, not to any measured output. In a second study, interviews with 51 participants found that most new work contacts made after training were inside their own organisation, and that how far people applied the training predicted how much their network of contacts changed. (Read in library)

The job people went back to shaped whether the training was used. In eight mid- to large-size Hungarian companies, 311 employees were surveyed between 13 and 120 days after a soft-skills course, such as stress management, time management or leadership development. More job resources went with more work engagement, more motivation to apply the training and more opportunity to apply it, and through those with more transfer. Higher job demands went with lower engagement, and through it with fewer opportunities to use the training. Neither resources nor demands was linked to transfer directly. The model explained 68.6% of the variation in transfer — that is, the share of the differences between people that the model accounts for, not an improvement in performance. (Read in library)

In a US sample, peers, feedback and capacity mattered for wanting to apply training, and supervisors for wanting to share it. The same research team tested what predicts motivation to apply and to share training, first with 537 German-speaking employees and then with 507 US employees surveyed twice. In the US study, peer support, feedback and transfer capacity were positively associated with motivation to apply the training; supervisor support and opportunity to apply were not. Motivation to share was associated with supervisor support, peer support and feedback. Among the individual factors, growth orientation, reflecting on the training and clear goals were relevant predictors. This study measured what people intended to do, not what they then did, and the authors note that the sample was skewed towards managers. (Read in library)

At firm level, training managers in mental health goes with better results, but the study did not test cause. A survey of employers in England ran for four years, with between 1,551 and 1,904 firms a year, 7,139 responses in all. Firms that provided line-manager training in mental health reported better recruitment, retention, customer service and business performance, and less long-term sickness absence due to mental ill-health. There was no significant association with three other sickness-absence measures. The authors say they did not test the causal pathway. Response rates were 15 to 17 per cent, which they say may limit how far the findings generalise. Note what this study is and is not: it compares firms, and says nothing about whether any manager changed what they did after training. (Read in library)

Is the manager the key? The same team found different answers in two samples. In the German-speaking survey, supervisor support predicted applying the training. In the US study, it did not, while peer support, feedback and capacity did. But the two studies did not measure the same thing: the first asked what people did with their training, the second only whether they wanted to. Peer support mattered in both. A result that falls short of statistical significance is also not proof that managers make no difference, particularly in a sample where most respondents were managers themselves. What the evidence does not support is leaving transfer to the manager alone.

Opportunity to use the training mattered in one study and not in another. In the Hungarian companies, opportunity to apply the training was one of the routes from job resources to using it. In the US study, opportunity to apply was not associated with wanting to use it. Again the outcomes differ: one study measured use, the other motivation, so the two findings are not in direct conflict.

The four workplace studies are self-reported, and mostly measured once. Nobody observed what people did at work or measured their output independently. Performance was people's own rating, and the firm-level study relied on what employers reported. These are associations, consistent with training working through use, but they do not show that training caused the change. The safety review is the exception and the one furthest from a management course: it rests largely on injury records, and what it shows is that training by itself rarely moves them.

The settings are narrow. German-speaking and US employees, soft-skills courses in eight Hungarian companies, mental-health training for managers in England, and safety interventions at work. The direction is consistent; the sizes do not transfer.

Training can change how people work, but in these studies the course is only half of it. Whether people used what they learned depended on what surrounded the course: colleagues who back it, feedback, resources, and enough room in the job to try it. Heavy demands worked against it. So the more useful question after a course is not only "was it good?" but "did people get the chance and the support to use it?"

Three qualifications keep this honest. The workplace evidence rests on self-reports, mostly taken once, so it shows what goes together, not what causes what. The manager's role is unsettled: support from peers was linked to using training in one sample and to wanting to use it in the other; support from supervisors only to using it in the first. And none of this shows how much training lifts performance. No study here measured anyone's output independently, and in the one place where results were measured objectively — injuries at work — training on its own rarely moved them.

⚠️ None of these studies was conducted in the Gulf. The four workplace studies come from German-speaking countries, the United States, Hungary and England, and no Gulf country appears anywhere in the safety review. The direction is consistent enough to act on; the numbers are not local.

1. Agree what each person will use the training for, before the course. Clear goals predicted wanting to apply and to share what was learned in the US study, which did not measure whether people then did.
2. Build reflection and practice into the course. Reflecting on the training was a relevant predictor, and the team also found effects for how the training was designed and how applicable its content was.
3. Protect time to use it. Capacity to apply predicted wanting to apply it in the US study. In the Hungarian companies, the only significant route from heavy job demands to less use of training ran through lower engagement first and fewer opportunities second; demands and opportunity on their own were not linked.
4. Organise support from colleagues, not only sign-off from the manager. Peer support was linked to using training in the German-speaking study and to wanting to use it in the US one. Supervisor support was linked to using it in the first but not to wanting to use it in the second, which is not the same as showing that managers do not matter.
5. Give feedback on how people are using it in the weeks after. Feedback predicted wanting to apply and to share what was learned.
6. Ask people to share what they learned. In the German-speaking study, sharing was a second route to results alongside applying the training. The US study looked at what predicts sharing, not at what sharing leads to.
7. Judge training by use, not attendance — and don't expect a course on its own to move business numbers. Ask participants, and the people they work with, what they now do differently.

A training requirement counts training, not use.

Saudi Arabia now requires larger employers to train Saudi graduates and job seekers every year. Our brief on graduate programmes sets out the rules. A requirement like this is met by delivering the training. The studies here suggest that its value depends on what happens afterwards. Offered as reasoning, not as a finding: record whether trainees use what they learned alongside the numbers you report.

Where hierarchy is steep, colleagues may be an easier lever than the manager. Our brief on manager development argues that in higher power-distance workplaces, training imported unchanged may not fit. Peer support mattered in both samples that tested it here. That makes structured support among colleagues a lever that does not rest on the manager alone. This too is reasoning rather than a Gulf finding.

We hold no Gulf study on training transfer. That is an evidence gap, not a settled matter.

All five sources below are open-access, licence verified per article.

1. Safety interventions for the prevention of accidents at work: A systematic review — Campbell Systematic Reviews
2. How to maximize the impact of workplace training: a mixed-method analysis of social support, training transfer and knowledge sharing — European Journal of Work and Organizational Psychology
3. The positive gain spiral of job resources, work engagement, opportunity and motivation on training transfer — International Journal of Training and Development
4. When formal training leads to informal learning among colleagues: antecedents of training transfer and knowledge sharing — Frontiers in Organizational Psychology
5. The relationship between line manager training in mental health and organisational outcomes — PLOS ONE