By Heather Lo,
Director of Product at Learnlight
We have more access to information today than at any point in human history. You can find a masterclass on negotiation, a webinar on executive communication, a podcast on giving effective feedback — all within a few quick clicks. Yet, skills gaps keep widening and organizations keep investing in training programs only to find behaviors don’t change.
The problem isn’t that people aren’t learning. It’s that we’ve confused learning with capability.
Completing Training Isn’t The Same As Developing A Skill
When companies identify a skills gap, the typical response is to provide more information, such as a new course or a curated content library. These are reasonable responses to a knowledge gap. However, a skills gap is a different problem entirely.
Why Practice Is The Real Issue
There’s a critical distinction between knowledge transfer, skill development, and skill mastery. Most training programs are excellent at knowledge transfer and inconsistent at skill development — they rarely get anywhere near skill mastery.
Here’s what that looks like in practice:
Step 1: You Learn The Theory (Knowledge Transfer)
Knowledge transfer is exactly what it sounds like. Let’s imagine we’re developing presentation skills: You learn that an effective presentation starts with a hook, should follow a clear structure, and uses transitional language to guide the audience.
Step 2: You Apply Your Knowledge In Low-Stakes Conditions (Skills Development)
Skill development happens when you put that knowledge into practice, usually within a training environment. So, you draft a presentation, applying the principles you’ve just learned. You execute it reasonably well, because you’re focused, not in a hurry, and you might even have your notes in front of you. You’re in a classroom setting, so you receive feedback from an instructor. And you learn from that feedback, in a low-risk, low-pressure environment.
It’s a meaningful step forward. The problem is that most programs stop here. So the skill, which was fragile and context-dependent to begin with, never properly develops.
Step 3: The Behavior Becomes Automatic (Skills Mastery)
Skill mastery is a different category entirely. It’s what happens when a behavior becomes automatic. For example, you don’t have to think about how to convey information from your presentation in a meaningful way, you just do it.
Getting there takes repeated practice across different contexts, including the inevitable small mistakes and micro-failures that are part of how skills actually embed. One application inside a training module doesn’t build that.
AI Makes The Missing Practice Affordable
The challenge with creating repeated practice opportunities at scale is that they’ve traditionally been expensive and logistically complicated. For skills that employees don’t use daily, finding meaningful practice opportunities is particularly difficult. An emerging leader learning to give effective feedback, for example, may not yet manage a team. They have no natural context in which to practice.
This is where AI trainers shift the dynamic. They make structured, feedback-rich practice available continuously, at a fraction of the cost of additional human-led sessions. A future manager can practice giving feedback long before they have a team to manage. They can make micro-failures in a low-stakes environment, iterate, and develop real fluency over time.
Pairing AI practice with human facilitation produces particularly strong results. By the time a learner sits with a human trainer, they’ve already accumulated hours of practice. The trainer isn’t starting from scratch. They’re refining, going deeper, and working on the nuances that only emerge through real experience.
Most Training Measurement Is Looking For The Wrong Thing
If the practice gap is real and widespread, why aren’t organizations catching it? Because most training measurements don’t look for it.
Traditional learning metrics, such as login frequency and activity completion, are useful as leading indicators. They tell you something about engagement — but they measure knowledge transfer, not skill development. Passing an assessment tells you the learner understood the content but it says nothing about whether they can execute the skill under real-world conditions. Measuring skill development requires a different approach.
How Should We Measure Skills Development?
The most useful signals come from learners themselves: competence ratings taken at the start of a program, during it, and at the end. How confident did you feel before? How about now? Can you point to moments where you’ve applied this differently?
It also requires manager involvement. Managers who are close to their teams’ day-to-day work are best placed to observe behavioral change. Is the training showing up in how people communicate, present, give feedback, or navigate difficult conversations?
The right training provider can make this significantly easier. Behavioral data is hard to gather without support, and how much is available depends on how much access an organization is willing to give. Surveying learners is straightforward. Extending that to managers, peers, or coworkers produces a far richer picture of behavioral change, but it requires a different level of organizational buy-in.
A good provider will help establish the right level of depth from the outset: agreeing on targets, identifying the most appropriate measurement tools, and supporting the execution throughout.
To Summarize: Don’t Skip The Part That Makes Training Work
Organizations that invest in training and see real behavioral change aren’t necessarily spending more. They’re just not stopping too soon.
The difference between a training program that shifts how people work and one that just produces certificates is almost always practice. Whether that means building structured repetition into the program design, using AI to create low-stakes practice opportunities between sessions, or involving managers in reinforcing new behaviors on the job, the mechanism matters.
If your organization is committed to closing capability gaps, that commitment is worth protecting. Build in the practice, measure the behavior, and don’t let completion data tell you the story is finished when it’s only just begun.
—-
About the author
Heather Lo is Director of Product at Learnlight, where she has spent over 13 years shaping how the company thinks about communication and language learning. She leads product strategy and portfolio execution across Learnlight’s learning solutions, working at the intersection of learner experience, technology, and organizational impact.
Before moving into product, Heather spent a decade in the classroom and in curriculum design. This grounding in how people learn informs everything she builds at Learnlight.
