What Each Method Produces: The Kirkpatrick Framework
Donald Kirkpatrick's four-level training evaluation framework (updated in Kirkpatrick and Kirkpatrick 2016) provides the most widely used structure for comparing training methods by outcome:
- Level 1: Reaction. Did participants find the training engaging and relevant?
- Level 2: Learning. Did participants acquire the intended knowledge, skills or attitudes?
- Level 3: Behaviour. Did participants apply what they learned in their actual work?
- Level 4: Results. Did the application produce the intended organisational outcomes?
The critical insight for comparing e-learning and immersive training is that most training investment is evaluated at Level 1 and 2, but most training value is delivered at Level 3. An employee who completes a compliance e-learning module and passes the quiz has demonstrated Level 2 learning. Whether they behave differently in the actual compliance-relevant moment (Level 3) is what determines organisational outcome.
| Kirkpatrick Level | E-learning | Immersive training |
|---|---|---|
| L1: Reaction | Varies by content quality | Typically high, novelty + engagement |
| L2: Learning | Effective for knowledge transfer | Effective for capability development |
| L3: Behaviour | Weak, rarely tested | Strong, directly targeted |
| L4: Results | Rarely evidenced | Available with structured measurement |
The Research Base for Active vs Passive Learning
Two research streams provide the theoretical foundation for why immersive training produces stronger Level 3 outcomes than e-learning.
Roediger and Karpicke (2006), the testing effect. A landmark paper in Psychological Science established that active retrieval of information from memory produces approximately 50% better long-term retention than passive re-reading or re-watching. When people are required to actively use information (through retrieval, application, or practice) rather than passively receive it, they retain it far more effectively. Immersive training operationalises active retrieval; e-learning typically operates through passive consumption.
Ericsson (2016), deliberate practice. Anders Ericsson's research, synthesised in Peak, establishes that capability development requires structured practice at the edge of current capability, with immediate feedback. E-learning can provide structured practice through quizzes and interactive elements but typically lacks the realism and immediate feedback that produces the deliberate practice conditions Ericsson identifies. Immersive training with professional actors provides more authentic practice conditions.
Sidestream's own academic research, building on the UCL, Cambridge and Bocconi research base, found that immersive role-play was approximately 20% more effective than passive modalities (slide-show, video e-learning) at teaching communication skills. The secondary finding around Dunning-Kruger patterns (participants self-reporting high confidence in skills they had not in fact mastered) pointed to a specific advantage of immersive method: it surfaces the gap between self-perceived capability and actual performance.
When to Use E-Learning
E-learning is the right choice when:
- The learning objective is knowledge transfer. Policy content, compliance facts, product information, process knowledge: these are well-suited to e-learning. The learner needs to acquire information, not to rehearse behaviour.
- Scale matters and behavioural precision does not. E-learning scales cost-effectively to thousands of learners. Where the objective is broad awareness across a large population, e-learning is more cost-effective than immersive training.
- On-demand access is required. E-learning is available at any time, on any device, in any location. For organisations with distributed workforces or mandatory refresher requirements, e-learning's accessibility is a genuine advantage.
- The behaviour is low-stakes and self-evident. Some behaviours (following a documented process, completing a form correctly, navigating a system) are well-learned through e-learning because the behaviour is clearly specified and the consequences of error are low.
When to Use Immersive Training
Immersive training is the right choice when:
- The objective is observable behaviour change. Leadership behaviour, speak-up culture, harassment-intervention behaviour, coaching skills, sales-conversation quality: these require rehearsal of specific observable moments, not knowledge transfer.
- The behaviour is consequential and high-stakes. Patient safety, financial conduct, harassment prevention, crisis command: the cost of sub-optimal behaviour is material. The higher the stakes, the more the investment in behaviour change is justified.
- The behaviour is specific and rehearsable. Immersive training works best when the specific behavioural moment can be scripted. "Being more inclusive" is too vague. "The meeting moment where a junior colleague's contribution is cut off" is specific and rehearsable.
- Engagement with sceptical populations is required. Senior and sophisticated cohorts that have grown sceptical of conventional training engage with immersive method more than with e-learning, because the novelty and realism break past the defensive posture conventional training encounters.
Why E-Learning Fails for Compliance Behaviour Change (Post-October 2024)
The October 2024 Worker Protection Act all-reasonable-steps duty has made this distinction operationally significant. Pre-2024, compliance procurement often defaulted to e-learning because it was defensible on a "training has been delivered" basis. Post-2024, tribunal practice reads behavioural evidence as the test of compliance, not completion records.
An employer who can demonstrate that their harassment prevention training produced observable bystander-intervention behaviour (Kirkpatrick Level 3) is in a stronger compliance position than one who can only demonstrate completion records from an e-learning module. This has driven a measurable shift in harassment prevention procurement from e-learning to immersive training for serious-purpose buyers.
The Combined Architecture: E-Learning Plus Immersive
The strongest training designs combine both:
- E-learning for foundations: what the policy says, what constitutes harassment, what the reporting process is, what the legal framework requires.
- Immersive training for behaviour: the bystander-intervention moment, the disclosure-response conversation, the manager-handling of conduct concerns.
This architecture uses each method for what it does best. E-learning produces the knowledge base cost-effectively at scale. Immersive training develops the behavioural capability for the specific high-stakes moments. The combination produces both knowledge and capability, which is what the post-2024 regulatory standard increasingly expects.
Practical Scenarios: Which to Use When
These worked examples illustrate the decision framework in practice.
Scenario 1: Annual compliance refresher for 2,000 employees. Use e-learning. The objective is to ensure all employees are aware of policies and legal requirements. Scale matters. Behavioural precision at individual level is not the primary objective. E-learning is the efficient and cost-effective choice.
Scenario 2: Post-October-2024 Worker Protection Act compliance programme for 150 managers. Use e-learning for foundations (what the duty is, what the policy says) plus immersive training for behaviour (bystander-intervention rehearsal, disclosure-response rehearsal, manager-handling of conduct concerns). The compliance standard requires behavioural evidence, not only awareness records. The combined architecture is the defensible choice.
Scenario 3: Leadership development for 20 senior executives. Use immersive training. The behaviours required (structured peer challenge, psychological safety creation, change-leadership conversations) are specific and high-stakes. The cohort is sophisticated and sceptical of conventional content delivery. The small cohort size makes the per-learner cost of immersive training comparable to or lower than bespoke e-learning production. Immersive is the right choice.
Scenario 4: New-employee induction for 500 joiners per year. Use e-learning for knowledge content (values, processes, compliance, product). Add structured in-person elements for team-integration and culture-specific behavioural norms. The full immersive approach is not appropriate at this scale and for this population mix; e-learning is efficient for the knowledge layer, and structured in-person work handles the cultural-integration layer that e-learning cannot.
Scenario 5: Sales-skills development for 30 account managers. Use immersive training. The specific sales-conversation behaviours (discovery questioning, objection handling, pitch structure) are rehearsable through professional-actor scenarios. Sales-behaviour outcomes (win rates, pipeline quality) are directly measurable at Kirkpatrick Level 4. The investment is justified by the commercial outcome it produces.
The Cost Comparison Done Properly
Comparing the cost of e-learning and immersive training at face value misleads procurement decisions. The correct comparison is cost-per-behavioural-outcome, not cost-per-learner.
E-learning cost model: Typically priced per learner, with a low per-head licence fee for off-the-shelf content and a higher one-off cost for bespoke production. At scale, spread across thousands of learners, the per-learner cost looks low. But if the training produces Level 1 and 2 outcomes and not Level 3, the cost-per-behavioural-outcome is infinite, because no behavioural outcomes were produced.
Immersive training cost model: Typically priced per cohort and scoped per engagement for groups of 12 to 25 participants. With a smaller cohort, the per-learner cost is materially higher than e-learning. But if the training produces observable behaviour change (Level 3) that the e-learning does not, the cost-per-behavioural-outcome is finite and often favourable relative to the cost of the behaviour the training was designed to prevent.
The regulatory case for immersive: Under the Worker Protection Act 2024 all-reasonable-steps duty, an employer whose harassment-prevention training produced observable bystander-intervention behaviour is in a stronger compliance position than one whose training produced completion records. The tribunal exposure for breach is materially higher than the immersive training cost. The relevant comparison is not cost-vs-cost but cost-vs-risk.
Related Sidestream Guides
- Immersive Simulation Training London
- Immersive Events London
- Harassment Prevention Training London, where the e-learning vs immersive distinction is most consequential post-2024
- What is the Kirkpatrick Model?
- Behaviour Change Training: The Complete UK Guide
- Glossary: 100 Behaviour Change Terms
Frequently Asked Questions
What is the difference between immersive training and e-learning?
E-learning delivers content through digital platforms: video, quizzes, interactive modules. It produces Kirkpatrick Level 1 (reaction) and Level 2 (learning) outcomes at scale and at low marginal cost. Immersive training uses scripted scenarios performed by professional actors to rehearse specific behaviours and decisions. It produces Kirkpatrick Level 3 (behaviour change) outcomes. The two are structurally different products that serve different purposes.
Is immersive training more effective than e-learning?
For behaviour change at Kirkpatrick Level 3, immersive training is more effective than e-learning. For knowledge transfer at Kirkpatrick Level 2, e-learning is often sufficient and more cost-effective. Sidestream's own academic research found immersive role-play approximately 20% more effective than passive modalities at teaching communication skills. Roediger and Karpicke (2006) found that active retrieval produces approximately 50% better long-term retention than passive re-reading.
When should you use e-learning instead of immersive training?
Use e-learning when: the learning objective is knowledge transfer (compliance facts, process knowledge, product information); scale matters and behavioural precision does not; cost-per-learner must be minimised; content needs to be accessible on demand across time zones. Use immersive training when: the objective is observable behaviour change; the behaviour is consequential (safety, conduct, leadership, patient care); and the behaviours are specific, rehearsable and high-stakes.
What is the testing effect?
The testing effect (Roediger and Karpicke, 2006, Psychological Science) is the finding that active retrieval of information from memory produces approximately 50% better long-term retention than passive re-reading or re-watching. E-learning often operates through passive content consumption; immersive training operates through active engagement. The testing effect explains part of the retention advantage of active learning methods.
Can e-learning and immersive training work together?
Yes. E-learning for foundational knowledge transfer (what harassment is, what the reporting process is, what the policy says) combined with immersive training for behavioural rehearsal (what to do when witnessing harassment, how to respond to a disclosure) produces stronger outcomes than either alone. This is the standard Sidestream recommendation for compliance training: e-learning for awareness and policy, immersive for the specific behavioural moments.
How much does immersive training cost compared to e-learning?
E-learning is typically priced per learner, with a low per-head licence fee for off-the-shelf content and a higher one-off cost for bespoke production. Immersive training is typically priced per cohort and scoped per engagement for groups of 12 to 25 participants. The cost-per-learner comparisons are misleading without adjusting for outcome: e-learning produces awareness; immersive training produces behaviour change. The relevant comparison is cost-per-behavioural-outcome.
Immersive Training vs E-Learning: The Takeaways
Immersive training and e-learning solve different problems, so the honest answer is rarely one or the other. E-learning delivers knowledge at scale and produces strong Kirkpatrick Level 1 and 2 outcomes; immersive training rehearses behaviour and produces Level 3 change. The cost question is best read as cost-per-behavioural-outcome, not cost-per-learner.
- Use e-learning for knowledge transfer, scale and on-demand access; use immersive training for observable, high-stakes behaviour change.
- The research base, Roediger and Karpicke on the testing effect plus Sidestream's own UCL, Cambridge and Bocconi work, explains the retention and behaviour advantage.
- Since the October 2024 Worker Protection Act duty, tribunals read behavioural evidence, not completion records.
- The strongest designs combine both: e-learning to inform, immersive training to embed.