Using Data to Improve Training Programs

𝐓𝐡𝐞 𝐒𝐞𝐜𝐫𝐞𝐭 𝐭𝐨 𝐓𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐓𝐡𝐚𝐭 𝐀𝐜𝐭𝐮𝐚𝐥𝐥𝐲 𝐖𝐨𝐫𝐤𝐬? 𝐒𝐭𝐚𝐫𝐭 𝐚𝐭 𝐭𝐡𝐞 𝐄𝐧𝐝. 🏁 I used to think my job as an L&D professional started with a syllabus. I was wrong. Recently, I was tasked with building a learning solution for our Talent Acquisition (TA) team. The goal wasn’t just to "train recruiters"—it was to solve a business problem. Instead of looking at what they needed to know (Level 2), I started with what the business needed to achieve (Kirkpatrick Level 4). The "Reverse" Approach I didn’t start with slides. I started by analyzing Voice of the Customer (VOC) survey results, focusing on various metrics from both Hiring Managers and Candidates. Working Backwards: ✅ Level 4 (Results): I defined the business KPI. ✅ Level 3 (Behavior): Based on the VOC metrics, I identified the specific actions recruiters needed to change—specifically around "Precision Intake" and "Candidate Experience Management." ✅ Level 2 & 1 (Learning & Reaction): Only then did I design the actual training content that addressed those specific behavior gaps. The Result? The training didn't feel like a chore; it felt like a solution. Because I built it based on the actual metrics revealed in the VOC surveys, the TA team saw immediate value, and the business saw a measurable shift in hiring efficiency. The Lesson: If you want your learning solutions to be more than just "check-the-box" exercises, stop asking "What should we teach?" and start asking "What does the data say I need to solve?" How do you use VOC data to shape your enablement programs? 👇 #LearningAndDevelopment #InstructionalDesign #TalentAcquisition #KirkpatrickModel #Enablement #DataDrivenLD #BusinessImpact

From chatbots that personalize microlearning to systems that predict who’s likely to disengage, artificial intelligence (AI) is changing how we train and learn. AI opens new opportunities to improve on some of the challenges with traditional training models such as scalability, personalization and real-time feedback. Core AI applications in the L&D space can be broken down into four categories: Artificial Intelligence (AI) Platforms: These tools tailor difficulty, pacing and topics in real time. An AI-enhanced platform can tailor the content to the learner based on their performance trends. Natural Language Tools: These are used to summarize content, create quizzes and provide conversational coaching. These applications can reduce time spent on administrative tasks and increase the focus on building relationships and delivering value. Predictive Analytics: This category of tools help learning leaders identify skills gaps and forecast learner success. Virtual Coaches and Chatbots: These tools reinforce knowledge through spaced repetition and feedback loops. AI-Powered Learning: A Case Study Streamline Services is a fifth-generation plumbing, electrical and HVAC company that handles up to 200 calls a day and serves thousands of customers each month. The company is using AI to not only coach employees but also identify areas where the team needs skills development or training. Streamline adopted an AI-powered virtual ride along platform to help transform everyday customer interactions — both in the field and in the call center — into powerful, data-driven learning opportunities. Traditionally, managers and trainers could only coach based on a handful of ride alongs or recorded calls each month. With AI, every service visit and customer conversation has become searchable, analyzable and coachable. AI highlights key themes including customer concerns, missed opportunities and tone shifts, allowing trainers to see real patterns instead of isolated incidents. The training team and managers use this knowledge to design training and structure coaching for individual needs. Because AI is deepening Streamline’s understanding of customer needs, the L&D team can develop targeted training that improves customer service and empathy across the company. Streamline’s experience illustrates how AI is fundamentally changing the learning process — from reactive coaching based on limited observation to proactive, personalized development powered by real data. This case study showcases how technology can elevate human performance rather than replace it. AI offers the ability to provide more learning opportunities and personalized learning across roles and industries. L&D professionals need to embrace this change and evolve alongside the technology. The future of learning isn’t artificial — it’s intelligently human. #LearningandDevelopment #AI #FutureofLearning

One of our clients—an international energy company—was undergoing a massive transformation, shifting from oil to e-mobility and sustainable fuels. The board’s mandate was clear: build a workforce ready for tomorrow’s challenges. During my first week, I visited a remote field site. Standing beside a team of engineers, I could sense their anxiety about unfamiliar technologies, stricter compliance audits, and the relentless pressure to deliver results. The old training modules? They barely scratched the surface of what these teams truly needed. We soon realized that off-the-shelf courses just weren’t enough. Understanding how people actually felt about new work processes was essential. I spent hours with field and office teams—listening, mapping out real pain points, and asking sometimes uncomfortable questions. How can we help our people make critical decisions on the ground? How do we build capability at scale, rather than just ticking compliance boxes? Once we gained that clarity, everything began to shift. Our team created an interactive learning journey—complete with role-based simulations, gamified crisis scenarios, and data-driven feedback loops. Each module put learners in the driver’s seat, dealing with real-life emergencies or optimizing EV infrastructure in realistic ways. It wasn’t all smooth sailing. Our first pilot exposed significant gaps—some learners felt overwhelmed, while others needed more hands-on support.We responded quickly by launching peer forums, field workshops, and targeted communications to bridge those divides. Within just 90 days, employees became noticeably more confident. Sites reported improved safety, efficiency, and even reduced downtime. This experience reinforced for me how real listening, strategic design, and a willingness to adapt can transform not just results, but the culture itself. I aim to make every learning initiative feel like a story worth living—for teams and for the business. #LearningAndDevelopment #EnergySector #Transformation #CriticalThinking #ProblemSolving #EVReady (Photo by <ahref="https://lnkd.in/gQWCp5Qf">Stockcake</a>)

🚨 University of Arizona Football | Performance Systems Overview (2022–2023) During the 2022–2023 season, we had the opportunity to build out a truly integrated performance system at Arizona — blending objective monitoring, individualized training, and collaborative decision-making to support player development and availability. Here’s what that looked like behind the scenes: 🔧 System Components – Developed a centralized Athlete Management System (AMS) – Integrated force plate CMJs on GD-2 and GD+1 – Mapped GPS data to every drill in practice by volume, intensity, and density – Built a stoplight readiness model (Red/Yellow/Green) based on force, asymmetry, and wellness inputs – Weekly 1080 Sprint profiling to target individual acceleration deficits and monitor trends 📊 In-Season Monitoring Strategy – Combined neuromuscular data (jump height, RSImod, asymmetry) with GPS and workload trends – Used CV% and SD thresholds to flag meaningful fatigue changes – Adjusted pre-practice prep, lifting intensities, and recovery based on G+1/G-2 trends – Created individual and positional reports shared daily with performance and coaching staff 📈 Results – Logged 31 new top speed records – Saw a 35% reduction in soft tissue injuries with minimal hamstring-related time-loss – Aligned training with the competitive calendar: Winter → Spring → Camp → Season → Postseason – Worked closely with the Performance Director to manage daily decisions around practice structure and player availability 🎯 Takeaway What made the difference wasn’t any one piece of tech or protocol — it was the ability to tie together force diagnostics, GPS load, sprint data, and on-field context into a unified decision-making system. Building that bridge between data and action is where the real impact happens. #SportsScience #AthleteMonitoring #PerformanceAnalytics #SpeedDevelopment #InjuryPrevention #CollegeFootball #ForcePlates #GPS #1080Sprint #SpellmanPerformance #ArizonaFootball

🛑 Stop Guessing. Your Data Predicted This Hamstring Tear Weeks Ago 💬 Want the dashboard used for this analysis? Comment “SPRINT” and I’ll send it to you. In modern football, training high-speed running is no longer optional — it’s essential. Yet, many training environments rely heavily on: • small-sided games • possession drills • reduced spaces 👉 The problem? These methods often fail to expose players to high-speed and sprint demands. So while we improve technical and tactical aspects, we may be underpreparing players for the most demanding moments of the game. 📌 Case study insight: I analyzed a real hamstring injury case using GPS data and a custom dashboard. The injury occurred exactly at the moment the player reached his seasonal peak speed, after weeks of insufficient weekly load — both in terms of distance covered at high speed and exposure to maximal velocity. 📊 Here’s what the data showed in the 30 days before the injury: ➡️ Match demands: • Avg Max Speed: 30.9 km/h • Top Speed: 34.4 km/h • Avg Sprint Distance: 155 m ➡️ Training exposure: • Avg Max Speed: 24.3 km/h • Top Speed: 29.3 km/h • Avg Sprint Distance: 7 m 📉 Last 15 days before injury: • Total Sprint Exposure: 33 m ⚠️ The issue wasn’t total load. It was the gap between training and match demands. The player was exposed in competition to: 👉 maximal speeds 👉 high eccentric stress …without ever being prepared for it in training. 💡 The reality: Sprint isn’t the problem. Lack of sprint exposure is. 🧠 Key takeaway: If your players reach >95% of their Vmax in matches… but never do it during the week… 👉 you’re not preparing them — you’re exposing them. 🎯 Practical implications: ✔️ Monitor High-Speed Running & Sprint as injury metrics ✔️ Ensure weekly exposure to >90–95% Vmax ✔️ Individualize thresholds (avoid absolute values) ✔️ Integrate running-based work when needed #FootballPerformance #SportsScience #SprintPerformance #ForceVelocity #StrengthAndConditioning #HSR #Highspeedrunning #Sprint #hamstring #hamstringinjury #hamstringprevention #DataAnalysis #PowerBI #SoccerScience #ReturnToPlay #PlayerProfiling #CFCInsights #dataviz #datafam #powerbi #tableu #Vizathon #sportsanalytics #sportsscience #gpsanalysis #soccer #matchanalysis #traininganalysis #soccerdrills #sportHorizon #performanceanalysis #datascience #performanceInsights #scouting #soccerscout #monitoring #trainingsoccer #hamstring #injuryhamstring #peakofspeed #sprintraining #sprintsoccer #speedtraining #PreparazioneAtletica #LiverpoolFc #NeuromuscularEfficacy #Loadmonitoring #ACC #DEC

Inspired by the latest FSI Training conference and Fabio Nakamura's lecture, I wanted to share the methodology for data analysis that I have implemented within my team. Previously, I analyzed drill intensity compared to full match data. My approach involved looking at intensity metrics over the full duration of a match and comparing them with training data. This method proved inadequate for capturing individual training performance and match intensity. In the last few days, I have focused on a more precise approach by segmenting match data into individual cuts. This allowed me to establish thresholds for each player, enhancing the accuracy of the analysis. To streamline the process before importing the data into Power BI, I divided each player's match data into segments of 3, 4, 5, 6, 8, and 10 minutes. For each segment, I calculated the average of the three best values for the variables: Total Distance, High-Speed Running (HSR) Distance, Sprint Distance, Acceleration Efforts (Zones 2+3), Acceleration Distance, Deceleration Efforts (Zones 2+3), Deceleration Distance, and Player Load. The rationale behind averaging the three best values, rather than using a single best value, is that outliers can create unrealistic thresholds. For example, during a short two-minute period, an athlete may be highly motivated, resulting in an intensity peak that does not represent sustainable performance. Averaging the three best values provides a more reliable and representative benchmark. By dividing drill values by these thresholds, I calculated the percentage of match intensity. This adjustment revealed that the previously analyzed drill intensities were often overestimated by up to 50% in some cases. An example in the images involves two players with different game thresholds. When examining absolute data, it appears that Player 1 experiences a higher mechanical load (in terms of accelerations and decelerations). However, both players exhibit similar intensity levels when compared to their match thresholds. This discrepancy becomes even more apparent when analyzing a full 90-minute match, as was done in my earlier approach. This finding underscores that absolute values alone cannot provide meaningful insights into individual player intensity during training. Each player and each drill must be carefully analyzed to draw accurate conclusions, such as determining whether a player was exerting sufficient effort. The next step in this process is automation. Ideally, the program should recognize drill duration and automatically adjust it to the match intensity thresholds. I'm looking forward to chatting about this approach to data analysis. Any ideas or suggestions on how to enhance this method further are welcome! #sportsscience #dataanalysis #strengthandconditioning #soccer

I was just speaking with the L&D Leader of a multi-billion dollar business who shared their journey to securing the business data needed to prove L&D's impact, a common struggle for many of us. They’d been on both ends of the spectrum: the Fortune 500 company where a high-ranking person refused to share business data and their current role where stakeholders are willing to hand over the data. For L&D professionals, getting access to those business metrics is half the battle. Here is the strategic approach they used to build an indispensable L&D function: 1. Focus on the business's biggest pain points (quantified with data) They targeted major, quantifiable business risks. Their first focus was fixing a massive problem: Ridiculously high turnover in one of the business units. They were also intensely interested in attrition, seeing the correlation between how they were preparing people and the number of people leaving. 2. Deliver wins before asking for the keys They built trust by showing immediate, quantifiable value first, offering to help with no questions asked. This resulted in: - Increasing the production output of new starters by focusing more on the actual work during training - Then shaving weeks off of a multi-month training program for new starters due to greater focus on performance and impact and then asking whether there was a more efficient way of achieving the same results - Which all resulted in business partners sharing more data with them because they saw such a huge impact on their day-to-day work. 3. Mirror the metrics that matter Their team now formally aligns L&D goals with business-driven outcomes. They write goals based on the same business metrics their stakeholders use when meeting with their own teams. Their future goals include things like: - Reduce x amount of time in the classroom - See x amount of proficiency on calls - Achieve x amount of billing 4. Provide proactive visibility (report out constantly) They don't wait for stakeholders to ask for updates. They report out L&D's impact quarterly, transparently and proactively, putting it in the hands of stakeholders. This strategic visibility ensures L&D is never overlooked. This transformation has shifted L&D from a service line that could be cut to a strategic partner that the business says, "We can't live without you". There’s so much to learn from and admire about this L&D leader’s approach, but in a nutshell: You must be married to the business's challenges, not just delivering learning in the hope of affecting them. We're rarely going to be invited to the conversations we want to be in and so we need to take our opportunities, deliver impact, use successes as leverage and reinforce - via our actions - that we are a crucial factor when it comes to driving performance and results.

🏋️♂️ What if you could spot signs of overtraining before they hit you? 📄🔥 New review out - Sport psychophysiology & HRV! 🔥🫀🧠 “Monitoring training adaptation: a scoping review of the relationship between self-reported subjective variables and resting vagally-mediated heart rate variability (vmHRV) in adult athletes” led by Carla Alfonso, with Valerie Helen Haydt, Mark Allen, & Luis Capdevila - Link below ⬇ 💡 Why this matters for sport & performance: Athletes and coaches are always looking for the sweet spot between pushing hard and recovering well. Our review looked at how subjective measures (fatigue, sleep, mood…) relate to resting HRV — a key marker of autonomic function. 💭 This is something you can do with HRV4Training from Marco Altini, PhD: right after measuring your HRV each morning, you fill out a short questionnaire. Both pieces of information — numbers and feelings — are essential for better training monitoring and smarter planning. ✨ What we found: 📈 Combining self-reports + HRV gives a more complete picture of training adaptation. ❗ The link is promising but not fully understood — more long-term, high-quality studies are needed. 🛠 Practical tip: don’t rely on just numbers or just feelings — use both. This works originated during Carla Alfonso research stay for her PhD at Performance Psychology - German Sport University Cologne - Deutsche Sporthochschule Köln 🇪🇸🇫🇷🇩🇪 ✍️ These findings are also making their way into the complete HRV Guide for Athletes & Coaches — the book I’m writing this summer with Marco Altini, PhD, Dr Emma Mosley, and Daniel Plews PhD. A full, practical, and science-based handbook for using HRV in sport. Coming next Spring! #HRV #AthleteMonitoring #SportScience #Performance #TrainingAdaptation #SportsPsychology

𝗥𝗲𝗺𝗼𝘁𝗲 𝗢𝗽𝗲𝗿𝗮𝘁𝗶𝗼𝗻𝘀 + 𝗥𝗲𝗮𝗹-𝗧𝗶𝗺𝗲 𝗗𝗮𝘁𝗮: 𝗧𝗵𝗲 𝗳𝘂𝘁𝘂𝗿𝗲 𝗼𝗳 𝗜𝗻𝗱𝘂𝘀𝘁𝗿𝗶𝗮𝗹 𝗧𝗿𝗮𝗶𝗻𝗶𝗻𝗴? 🏭 Virtual training is transforming how industries approach complex operations. From mining to aquaculture, immersive simulation combined with live IoT data is transforming workforce development. Companies like Minverso are proving that plant process simulation isn't just about training — it's about creating safer, smarter operations across entire industries. 🎯 𝗧𝗵𝗲 𝗯𝗿𝗲𝗮𝗸𝘁𝗵𝗿𝗼𝘂𝗴𝗵 𝗮𝗽𝗽𝗿𝗼𝗮𝗰𝗵: ➡️ Immersive plant simulation — Practice every stage of complex processes virtually ➡️ Real-time IoT integration — Live data feeds from actual equipment and sensors ➡️ Zero operational risk — Learn dangerous procedures without real-world consequences ➡️ Faster learning curves — Visual, interactive training vs. traditional methods 🌊 𝗥𝗲𝗮𝗹-𝘄𝗼𝗿𝗹𝗱 𝗶𝗺𝗽𝗮𝗰𝘁 𝗮𝗰𝗿𝗼𝘀𝘀 𝗶𝗻𝗱𝘂𝘀𝘁𝗿𝗶𝗲𝘀: ➡️ Aquaculture: Simulate fish farming operations & water quality management ➡️ Mining: Practice equipment operation, safety protocols, emergency response ➡️ Manufacturing: Train on production lines, quality control, maintenance procedures ➡️ Energy: Simulate power plant operations, grid management, safety systems 🤖 𝗧𝗵𝗲 𝗴𝗮𝗺𝗲-𝗰𝗵𝗮𝗻𝗴𝗲𝗿: 𝗟𝗶𝘃𝗲 𝗱𝗮𝘁𝗮 𝗶𝗻𝘁𝗲𝗴𝗿𝗮𝘁𝗶𝗼𝗻 When VR training connects to real-time plant data, trainees experience: ➡️ Actual equipment performance metrics ➡️ Real environmental conditions ➡️ Live system alerts and responses ➡️ Decision-making with real consequences (virtually) Why this matters: Traditional training teaches theory. VR + IoT teaches reality — without the risks, costs, or downtime of on-site practice. The future of industrial training isn't just virtual. It's virtually connected to the real world, creating workforces that are prepared for anything because they've already experienced everything.

Most LLM agents stop learning after fine-tuning. They can replay expert demos but can’t adapt when the world changes. That’s because we train them with imitation learning—they copy human actions without seeing what happens when they fail. It’s reward-free but narrow. The next logical step, reinforcement learning, lets agents explore and learn from rewards, yet in real settings (e.g. websites, APIs, operating systems) reliable rewards rarely exist or appear too late. RL becomes unstable and costly, leaving LLMs stuck between a method that can’t generalize and one that can’t start. Researchers from Meta and Ohio State propose a bridge called Early Experience. Instead of waiting for rewards, agents act, observe what happens, and turn those future states into supervision. It’s still reward-free but grounded in real consequences. They test two ways to use this data: 1. Implicit World Modeling: for every state–action pair, predict the next state. The model learns how the world reacts—what actions lead where, what failures look like. 2. Self-Reflection: sample a few alternative actions, execute them, and ask the model to explain in language why the expert’s move was better. These reflections become new training targets, teaching decision principles that transfer across tasks. Across eight benchmarks, from home simulations and science labs to APIs, travel planning, and web navigation, both methods beat imitation learning. In WebShop, success jumped from 42 % to 60 %; in long-horizon planning, gains reached 15 points. When later fine-tuned with RL, these checkpoints reached higher final performance and needed half (or even one-eighth) of the expert data. The gains held from 3B to 70B-parameter models. To use this yourself:, here is what you need to do: • Log each interaction and store a short summary of the next state—success, error, or side effect. • Run a brief next-state prediction phase before your normal fine-tune so the model learns transitions. • Add reflection data: run two-four alternative actions, collect results, and prompt the model to explain why the expert step was better. Train on those reflections plus the correct action. • Keep compute constant—replace part of imitation learning, not add more. This approach makes agent training cheaper, less dependent on scarce expert data, and more adaptive. As models learn from self-generated experience, the skill barrier for building capable agents drops dramatically. In my opinion, the new challenge is governance and ensuring they don’t learn the wrong lessons. That means filtering unsafe traces, constraining environments to safe actions, and auditing reflections before they become training data. When rewards are scarce and demonstrations costly, let the agent learn from what it already has, its own experience! That shift turns LLMs from static imitators into dynamic learners and moves us closer to systems that truly improve through interaction, safely and at scale.