On-the-Job Training Practices

Prototyping is proven to have the potential to transform the speed, quality & impact of instructional design: can AI finally make prototyping a standard part of our process? For years, studies have shown that rapid prototyping in instructional design: 📊 Significantly shortens development cycles (Gerber & Carroll, 2012) 📊 Improves instructional quality (Daugherty et al., 2007) 📊 Enhances the quality of stakeholder collaboration (Nixon & Lee, 2001) Despite 20+ years of evidence & tools like Balsamiq and Figma, instructional design has remained stuck in waterfall workflows with little if any testing & iteration. The question I've been exploring this week is, will AI prototyping tools change this? In this week's blog post I share what I learned prototyping a recent training design using AI. TLDR: → AI tools like Claude, Vercel & Loveable are finally making rapid prototyping in instructional design practical, fast, and accessible—transforming abstract learning concepts into testable, shareable experiences in minutes → While AI isn’t a silver bullet (it struggles with complex visuals and multi-page journeys), it does a good job of generating realistic, evidence-based scenarios, assessments, and case studies—*provided* the designer brings strong instructional expertise and prompt precision → The future of L&D lies in combining deep pedagogical expertise with AI fluency. Check out my full guide to AI prototyping for L&D, complete with prompts you can try for yourself, using the link in comments. Happy innovating! Phil 👋

Interesting paper to stick your teeth into if you're an L&D, concerned with learning transfer. 💡 The authors reviewed 71 studies to build the so-called COMPASS model, which combines two well-established models: The COM-B model (Capability, Opportunity, Motivation = Behaviour) And Baldwin & Ford's training transfer framework. In a nutshell: The COMPASS model focuses on three key components that influence soft skills transfer: 1️⃣ Trainee characteristics (e.g. prior experience, motivation, and self-efficacy) 2️⃣ Training features (e.g. content relevance, design, delivery, and support) 3️⃣ Work environment (e.g. manager support, team norms, and org culture) The research identified 69 factors influencing behaviour transfer. 🟢 The ones with favourable evidence of impact: On-the-job training Relevance of training Time-spaced training Micro-learning Pre-training materials Training assessment Trainer effectiveness/credibility Multiple instructional methods Use of technology Workshops Goal-setting Mentoring/coaching/supervision 🔵 The ones with emerging evidence of impact: Community of practice Personalization Variability and increasing complexity Facilitation or assistance Feedback Group assignment Observation of others Reflection Role play Lots to chew on, and Sejaal Tilwani made a little overview, including some practice recommendations, in the latest Learning Brief Newsletter: https://lnkd.in/eMrniWs6

It was 3:42 AM. A junior data engineer had just deployed an update to a core ETL pipeline. Minutes later, critical data jobs started failing, no reports were loading, dashboards were blank, execs across 5 time zones started pinging Slack. Panicked, he called the on-call senior. His first words were exactly, “I think I broke something in the latest DAG. Data isn’t moving. I’m really sorry, it’s on me. How should we tackle this?” The senior didn’t say: → “Did you not test locally?” → “Why didn’t you follow the migration checklist?” → “This is why we don’t deploy at night.” Instead, he said: “Breathe. I’m here.” Within 15 minutes, they had looped in another data engineer and the analytics lead. They spun up a staging pipeline, checked data snapshots, and started tracing the failure. By morning, the pipeline was fixed. No critical data was lost. The dashboards were live before the CEO even noticed. Was it a serious incident? Absolutely. Could it have been avoided? Most likely. But it was also a huge learning moment. No one sent blame emails. No one asked for “accountability reports.” In the post-mortem, here’s what was said: When production’s on fire, it’s not about who to blame. It’s about who steps up, who calls for help, and who’s willing to fix it together. He flagged the issue early, owned up, and asked for backup. That’s what you want in a teammate. Mistakes will happen. Even in big data. Even with your most careful plans. But when someone tells you the pipeline’s on fire at 3 AM, → Don’t make them feel smaller. → Hand them a playbook, not a punishment. Lessons come after. First, you help put out the fire. That’s real data engineering leadership. That’s what I’ve learned leading teams in the trenches as a Senior Engineer

I wanted to become an expert in ‘Behaviour Change’ But I struggled till an incident showed me what I was missing The story & the 2 Behavior Change principles every leader must know… It’s May and the Indian summer is at full strength. I am taking a session for school teachers but they don’t want to be here. They don’t want to attend another training that will be boring, disrespectful & useless. But our research had revealed these issues so I have some tricks up my sleeve. Our content is bite-sized, actionable & solves their top problems. Our trainings are full of games & movie themes. Most importantly we treat teachers with genuine respect. It works & the trainings go incredibly well. Teachers tell their friends & next day the attendance doubles. At the end, I ask: “Will you apply all this ?” “Yes” they answer. Then I follow up: “Do you think they will work ?” ”No” they reply. I am taken aback: “So why will you apply them ?” ”Because you were kind to us & you really believe they will work. We will try them for you.” This was unexpected. I wanted to convince teachers with our rigorous research & testing. But the trust was established by how we had treated them & our passionate belief in the content. Then I remembered Everett Rogers. Most know his innovation model: Innovators → Early adopters → Early majority… Few know his core insight behind the model: “Change is a social process. People change because of other people.” This explained what was happening here. Then teachers went back to their classes, tried our stuff & blew up our messages: ”OMG!! This works. Why were we not taught this before ?” ”This child who had never answered, answered today. I now believe every child can learn.” We had assumed these mindset shifts would take a long time. But many teachers were making them a lot faster. Guskey’s model provided the explanation: People do mindset trainings for teachers but that rarely works. What works better is to make it easy for teachers to try new behaviours. Once they experience success, they change their own mindsets. So ❌ Mindset Change → Action Change → Success ✅ Action Change → Success → Mindset Change ————————— Generalising these 2 principles for any leader trying to change behavior: 1️⃣ MESSENGERS MATTER → Make sure they are kind, fun, inspiring & respectful → It all starts with people liking & trusting them 2️⃣ FOCUS ON MAKING PEOPLE EXPERIENCE SUCCESS QUICKLY → People can disagree with you. They can’t disagree with their own success → Focus on them experiencing success quickly with the new behavior & adoption will skyrocket. We overhauled our model based on these 2 counter-intuitive principles. This helped us scale our model to 1000s of teachers & become one of the top training organizations in India. I now use these principles with 100s of entrepreneurs & leaders I support. They work just as well. #Change #Leadership #Management Any behaviour change insight you have stumbled onto ?

“We used to in every class have a Discord. It used to be like a lot of people just asking questions about maybe like, a lab or a homework... I guess everyone’s just Chat-GPT now. Like the new classes that I have now, we still have the Discord, but nobody really talks because most or all the questions are answered by ChatGPT.” —P16, undergraduate computing student If you’ve moderated a class Discord, you’ve probably felt this shift: a once-busy channel that used to hum with “anyone stuck on Q3?” goes quiet. Not because students stopped needing help, but because they started getting it elsewhere. A new study by Hou et al puts language to what many of us have sensed. Based on 17 interviews across seven R1 universities, students described a social rerouting of help-seeking: 13 of 17 said peer requests are now mediated by GenAI (often “ask GPT”), and students noticed community spaces like Discord slowing down. However, when AI becomes the first responder, the “hidden curriculum” stops circulating. Fewer quick questions means fewer micro-mentorships, fewer perspective-shifts, less socially shared regulation — all the good stuff that builds belonging and lifts performance over time. Students save minutes, but communities lose momentum. So what can educators do about this? - Design “peer-first, AI-fast” protocols. Peer interactions build camaraderie and a sense of belonging. Educators need to design experiences that build more peer interactions and support inside classrooms, to compensate for GenAI caused declines.   - Protect mentorship routes. Research also showed that younger students are reaching out less often to senior mentors, missing out on invisible learning that comes from understanding unwritten rules and cultural norms. Educators might need to formalize “office-hours relays” (senior → junior → cohort) so guidance doesn’t vanish.  - Create informal interaction opportunities. Informal opportunities help students build relationships beyond their immediate circle, and provide entry points into additional learning communities. Have you seen AI change the quality of collaboration in your learning or work spaces? How can we preserve the “hidden curriculum” when AI takes over first-line help?  #GenAI #Education #PeerInteraction #HiddenCurriculum

I can’t stop thinking about this. If you invest in your people from day 1, they’ll invest their talents in your company tenfold. It sounds obvious, but I’ve seen firsthand how often this gets missed. I joined companies and startups with zero training: - no documentation - unclear processes - no real onboarding I was expected to figure it out as I went, and honestly, it was brutal 😭 So here’s what *actually* sets people up for success: —— 1️⃣ What does a new hire need to know but feels awkward asking? Think back to your first 30 days. ↳ How do things actually work here? ↳ Where do I go for answers? ↳ What mistakes should I avoid early on? If the answers live only in someone’s head, that’s the gap. ✅ Document anything you explain more than once. —— 2️⃣ Where are people guessing instead of being guided? When training doesn’t exist, people improvise. ↳ Clicking the wrong thing ↳ Following outdated steps ↳ Copying work that isn’t quite right That’s how errors and rework happen. Tools like Tango make this easy by turning workflows into step-by-step guides. ✅ Record one common task this week and turn it into a reusable guide. —— 3️⃣ What tribal knowledge needs to be documented? You know it’s a systems problem when there are: ↳ Constant pings ↳ Repeating the same answers ↳ Little time for deep work ✅ Have your strongest team member document one core process they own. —— 4️⃣ Are you onboarding people or overwhelming them? More information doesn’t mean better onboarding. People need: ↳ Clear priorities ↳ Time to practice ↳ Space to build confidence ✅ Use a simple 30-60-90 day framework for all new hires —— 5️⃣ Are expectations clear or just assumed? When expectations are vague: ↳ People second-guess themselves ↳ Feedback comes too late ↳ Performance feels personal instead of fixable ✅ Check in early and often and schedule 20-minute check-ins with your manager or onboarding buddy in the first 8 weeks. —— When you give people the right tools, training, and support, you get: → Faster onboarding → More consistent processes → Fewer mistakes and support tickets → Happier, more confident employees 💙 You can’t expect people to thrive without setting them up properly. Set people up to win and they will 🫶 Do you agree? #TangoPartner

Stop blaming “bad hires.” Your expectations are killing your business. I’ve seen it happen too many times: Months spent recruiting the best people. Then they join and they're forgotten about. Within weeks: The new hire checks out & eventually quits. How To Set New Hires Up For Success: 1️⃣ Provide A Structured Onboarding Plan ↳ A clear roadmap for the first 30, 60, and 90 days. ↳ Outline key responsibilities, goals & success metrics. 2️⃣ Assign A Mentor Or Buddy ↳ Someone they can turn to for guidance & quick questions. ↳ Helps them feel supported and connected. 3️⃣ Create A Training & Development Plan ↳ Teach the tools, processes & skills they need to thrive. ↳ Learning shouldn’t stop after onboarding. 4️⃣ Set Clear Job Expectations ↳ Define what success looks like in their role. ↳ Align expectations early to avoid confusion later. 5️⃣ Celebrate Wins & Show Appreciation ↳ A simple “great job” goes a long way. ↳ Recognise contributions early to build confidence. 6️⃣ Encourage A Healthy Work-Life Balance ↳ Don't overload them from day one. ↳ Set realistic expectations & support their well-being. 7️⃣ Be Patient & Give Time To Adapt ↳ Even top performers need time to settle in. ↳ Support their learning curve instead of expecting instant results. Invest in your onboarding & training. Just as much as your recruitment. Hire well. Onboard better. Great onboarding isn’t a “nice to have” - it’s a necessity. Do you agree? Comment below ⬇️ ♻️ Repost to share with your network. 👋🏼 Follow Dan Mian for more insights.

🚀 Excited to share my latest Fortune column on truly groundbreaking academic work from my co-authors Professor Karim Lakhani and Fabrizio Dell'Acqua at Digital Data Design Institute at Harvard (D^3), where I serve as an executive fellow. This remarkable field experiment with 776 Procter & Gamble professionals fundamentally challenges what we thought we knew about teamwork. The research reveals the emergence of the "cybernetic teammate"—AI that doesn't just assist but actively participates in collaboration. Three breakthrough findings: 1. AI Can Replicate Team Benefits Individuals working with AI achieved nearly 40% performance gains—matching traditional two-person teams. AI is providing the same collaborative benefits we've long attributed to human teamwork. 2. Cross-Functional AI Teams Generate Breakthrough Innovation AI-augmented cross-functional teams were 3x more likely to produce top 10% solutions. This isn't marginal improvement—it's a multiplicative effect that neither human-only teams nor AI-enabled individuals could achieve alone. 3. AI Breaks Down Silos (For Real This Time) R&D specialists with AI proposed commercially viable solutions. Commercial professionals developed technically sound approaches. AI acted as a bridge, enabling each team member to think holistically across functions—achieving the "silo breaking" that leaders have struggled to accomplish through org chart reshuffles. Bonus finding: AI collaboration increased positive emotions by 64% in teams. This isn't cold, mechanical work—it's energizing and engaging. At Seven2, we're translating this research into practice with our portfolio companies, building these AI-augmented cross-functional teams to drive innovation and competitive advantage. This is the future of collaborative work—not AI replacing humans, but human-AI ensembles that combine the best of both worlds. Read the full analysis: https://lnkd.in/ef3f3pED #AI #Innovation #HBS #D3Institute #FutureOfWork #PrivateEquity #TeamDynamics

AI isn't just a tool; it's becoming a teammate. A major field experiment with 776 professionals at Procter & Gamble, led by researchers from Harvard, Wharton, and Warwick, revealed something remarkable: Generative AI can replicate and even outperform human teamwork. Read the recently published paper here: In a real-world new product development challenge, professionals were assigned to one of four conditions: 1. Control Individuals without AI 2. Human Team R&D + Commercial without AI (+0.24 SD) 3. Individual + AI Working alone with GPT-4 (+0.37 SD) 4. AI-Augmented Team Human team + GPT-4 (+0.39 SD) Key findings: ⭐ Individuals with AI matched the output quality of traditional teams, with 16% less time spent. ⭐ AI helped non-experts perform like seasoned product developers. ⭐ It flattened functional silos: R&D and Commercial employees produced more balanced, cross-functional solutions. ⭐ It made work feel better: AI users reported higher excitement and energy and lower anxiety, even more so than many working in human-only teams. What does this mean for organizations? 💡 Rethink team structures. One AI-empowered individual can do the work of two and do it faster. 💡 Democratize expertise. AI is a boundary-spanning engine that reduces reliance on deep specialization. 💡 Invest in AI fluency. Prompting and AI collaboration skills are the new competitive edge. 💡 Double down on innovation. AI + team = highest chance of top-tier breakthrough ideas. This is not just productivity software. This is a redefinition of how work happens. AI is no longer the intern or the assistant. It’s showing up as a cybernetic teammate, enhancing performance, dissolving silos, and lifting morale. The future of work isn’t human vs. AI. The next step is human + AI + new ways of collaborating. Are you ready?

Gagné’s Nine Events of Instruction was revolutionary in its time. But that time was nearly 80 years ago. It was built for military training—linear, rigid, objective-driven. It assumes the designer controls everything, the learner starts from zero, and outcomes are best achieved by following a prescribed sequence. That’s not how learning works anymore. Modern learners are rarely blank slates. They come with prior knowledge, personal context, and the ability to access what they need on demand. They’re not sitting passively, waiting for content to be “presented.” They’re navigating ambiguity, asking questions, collaborating, and applying knowledge in complex, unpredictable environments. That’s why I’ve moved away from traditional instructional design models like Gagné—and toward frameworks that reflect how people actually learn. I draw from Learning Experience Design (LXD), which blends learning science, user experience, and accessibility to create more engaging and emotionally resonant learning. I also pull from the 5E model, which prioritizes inquiry and exploration, and Universal Design for Learning (UDL), which builds flexibility and inclusivity into every part of the design. Models like Design Thinking and Agile Learning Design keep me grounded in iteration, learner feedback, and real-world relevance. And Bob Mosher’s Moment of Need Model reminds me that not all learning happens during training—it often happens in the workflow, under pressure, when support is needed most. I don’t follow any of these models religiously. I use what fits. Because the moment we box ourselves into one system, we stop designing for people and start designing for process. Gagné made sense in a world of chalkboards and overhead projectors. Today, we’re designing for mobile, social, immersive, and AI-powered experiences. That requires more flexibility, more empathy, and a willingness to break the mold when it no longer fits. Models are helpful. Dogma is not.