Managers Are Guessing at Future Skills. There's a Better Way

The Manager Skills Gap: Why Managers Fall Behind

Managers are falling behind faster than the people they lead. According to new research from TalentLMS, 47% of respondents said some of their skills had become obsolete over the past five years. The gap is sharpest year to year: 21% of managers said their skills became outdated in the last year alone, compared with just 10% of employees.

Why? Dimitris Tsingos, CEO at Epignosis, points to practice. “As managers move away from hands-on work, they get less practice, and practice is what keeps a skill current,” he says. AI accelerates that decay: “As we rely on it more, we may lose essential skills through lack of practice.”

Managers also struggle to see ahead. Some 38% reported difficulty knowing which skills their team will need in the coming year, and 36% said they struggle to keep up with how quickly AI is changing their team’s skill needs.

The fix isn’t more hands-on work on top of leading. It’s treating learning as part of the job.

Stop Guessing: Responsive Skill Planning

A majority (70%) agree that as roles and responsibilities shift, more agile skill practice is vital for employees, but only 16% say they see that happening in their organization. That gap matters because workplace planning is often treated as a fixed forecast. Dimitris Tsingos argues that when work is changing this fast, naming the exact skills you’ll need 12 months out doesn’t remove the uncertainty—it can create a false sense of certainty.

Instead, managers need more responsive ways to identify and develop skills, including:

• Gathering input from their teams
• Regularly reassessing which skills matter most
• Adjusting learning priorities as business needs evolve

“AI makes this even more urgent,” says Tsingos. In that environment, the advantage belongs to the organizations that can close the distance between spotting a new skill need and acting on it.

The Self-Taught Employee: Making Learning Visible

More than half of respondents say they primarily develop new skills by figuring things out on their own. Tsingos sees this as a sign of resilience, not a problem: employees test an approach, adjust, try again, and learn through the work itself.

The concern is when this happens and no one knows which skills are being acquired. Without visibility, capability grows unseen and unused, and companies end up hiring externally for skills already inside the team.

The opportunity is to make self-led learning more intentional. Tsingos recommends giving employees the tools to experiment, connecting development to their actual roles, and providing feedback along the way. AI can be a powerful part of that support, not by removing people from the process, but by helping them learn, test ideas, and make better decisions in the flow of work.

Learning vs. Working: Making Learning a Priority

Balancing current work with skill training for future work has always been a challenge, and the data bears this out. TalentLMS found that work priorities keep pushing learning aside for 44% of respondents.

“When 44% of people say work keeps pushing learning aside, it means that learning is treated as the thing you do once the real work is finished,” Tsingos says. “There will always be another deadline, another customer request, another priority. If learning has to win that fight on its own, it loses every time.”

Tsingos believes we’re already seeing the cost, pointing to TalentLMS’s Learning Debt report which found that 41% of employees say their role has evolved faster than their company’s ability to train them.

“As the line between learning and doing keeps blurring, the companies that move fastest will be the ones that build capability through the work itself, not alongside it,” he says. “That is a leadership choice: give people the time, tools, and support to build the skills the business needs.”

Effective leaders lead learning by example: they keep learning, create room for experimentation, use AI thoughtfully, and redirect human time toward the decisions, relationships, and problems that need it most.