I've delivered training in just about every format you can imagine over the past 30 years. Boardrooms with whiteboards. Conference centres with sticky notes. Zoom calls with breakout rooms. And now — interactive simulations where AI plays the dev team and you make the decisions.
Here's what I've learned: the format matters far less than whether the learner is active or passive. A brilliant classroom facilitator will beat a mediocre simulation. But a well-designed simulation will beat most classroom experiences — because it forces you to do the thing, not just hear about it.
The Problem with Traditional Training
Most training follows the same pattern: an expert talks, shows slides, maybe runs an exercise, and hopes knowledge transfers. The research on this is brutal. Studies consistently show that retention from lecture-based learning drops to 10-20% within a week. That expensive two-day workshop? Most of it evaporates before people get back to their desks.
The problem isn't bad trainers. The problem is that knowing about something and knowing how to do it are fundamentally different. I can explain Little's Law to you in five minutes. But until you've watched a simulated drive-through grind to a halt because you let too many cars in, you don't really understand it. Not in your bones.
What Interactive Labs Get Right
Interactive simulation-based labs flip the model. Instead of "listen, then practice later (if ever)", you practice first and learn from what happens. Here's what makes them powerful:
- Safe failure. You can blow up a Kanban system by removing all WIP limits and watch the chaos unfold. No one gets fired. No sprint gets ruined. You just hit replay and try a different strategy.
- Immediate feedback. You change a WIP limit and instantly see what happens to cycle time and throughput. There's no waiting three sprints to see if your theory was right.
- Self-pacing. Some people need to run the Penny Game simulation six times before the insight clicks. Others get it on the first run. A lab accommodates both without holding anyone back or leaving anyone behind.
- Data-driven insight. Simulations generate real metrics. You're not reading about cycle time distributions — you're creating them with your own decisions and then analysing the results.
In our Mastering Kanban programme, for example, you run a full Kanban system across three rounds. You set the WIP limits. You decide what to pull. You watch the flow metrics respond. Then you use Monte Carlo simulation on the data you generated to forecast delivery. That experience creates understanding that a slide deck simply cannot.
What Classrooms Get Right
But here's where I refuse to be dogmatic. Classroom training — whether face-to-face or live virtual — does things that simulations can't:
- Human discussion. The best learning moments in classrooms come from peer conversations. "We tried that and it didn't work because..." is worth more than any textbook.
- Contextual coaching. A skilled facilitator hears your specific situation and gives you specific advice. Simulations are generic by nature — they can't address the politics of your organisation or the quirks of your team.
- Social accountability. People are more likely to follow through on commitments made in front of other people. There's a reason cohort-based learning works.
- Certification pathways. Many professional certifications — Scrum.org's PSM, PSPO, and others — require assessed live classes. Simulations can prepare you, but the assessment happens in a different context.
Why "Both" Is the Right Answer
The framing of "labs vs classroom" is a false dichotomy. The real question is: what does the learner need at each stage of their development?
The best learning architecture uses simulation to build intuition, and human facilitation to build application. You need both.
Here's the pattern I've seen work best:
Phase 1: Build Intuition (Interactive Labs)
Before any classroom session, learners work through simulations. They experience the concepts. They build mental models. They come to the classroom with questions born from experience, not from reading.
Phase 2: Contextualise (Live Class)
Now the facilitator doesn't need to spend two hours explaining Little's Law — everyone's already felt it in the simulation. The live session focuses on application: "How does this apply to your specific team? What's your WIP right now? Where are your bottlenecks?"
Phase 3: Reinforce (Back to the Labs)
After the class, learners return to the simulations to experiment with the ideas they discussed. They try different strategies. They replay scenarios. The lab becomes a practice ground for ongoing development.
The AI Dimension
What's changed in the last two years is AI's ability to make simulations dramatically more realistic. In our interactive labs, AI doesn't just run pre-scripted scenarios — it plays the development team, generates realistic data, introduces variability, and responds to your decisions in ways that feel genuinely unpredictable. This bridges the gap between simulation and reality in ways that weren't possible even recently.
AI also enables adaptive coaching within the simulation. If you're struggling with a concept, the system can offer hints, surface relevant data, or adjust the difficulty. It's not a replacement for a human facilitator, but it's a significant step up from static e-learning.
Choosing the Right Format
Here's a practical framework for deciding which format to use:
- Use interactive labs when the concept can be demonstrated through cause and effect, when learners need to build intuition, when you need to scale across time zones, or when budget is constrained.
- Use live classes when learners need contextual coaching, when certification is required, when peer discussion adds significant value, or when change management requires social commitment.
- Use both when you're serious about lasting behaviour change. Which should be always.
This is exactly why Genius Teams offers both interactive labs and Scrum.org live classes. The labs build the intuition. The classes build the application. Together, they create understanding that actually sticks — and that's the only kind of training worth paying for.