It wasn’t so long ago that manufacturers needed to be convinced that the connected worker was necessary. That, however, is no longer the case. Here’s why. For decades, manufacturers operated with minimal turnover, but as retirements mounted, their institutional knowledge began walking out the door.
At the same time, manufacturers found themselves grappling with other serious challenges. Automation was making equipment more complex to maintain, supply chains were increasingly subject to unexpected logjams and tariffs turned change management into a permanent job function.
On-site answers
In manufacturing sites around the globe, companies have started investing in what’s known as the connected worker. At its simplest, the idea is to put current work instructions, training and troubleshooting guidance in a worker’s hands while they are doing the job. The aim is to help them handle more on their own. Instead of waiting for a supervisor or tracking down a veteran technician, the worker can find the right procedure on their own and determine what to do next to keep the work moving.
At too many facilities, veterans have become the unofficial answer desk. They get pulled away from their own work – and sometimes called in on weekends – to solve problems somebody else could handle with the right instructions. A connected-worker system spreads that knowledge throughout the company, so the experts aren’t constantly interrupted and the line keeps moving.
Moving instructions from a binder to a screen only solves part of the problem, though. Say a laser operator runs into trouble during a material changeover or spots an edge quality issue. They need the answer at the machine, not a vague memory of something covered months ago in training.
After all, not every job can be learned in four weeks. Some equipment takes months to master, especially sophisticated systems such as an automated laser cutting cell. With the right guidance at hand and consistently available, workers can begin learning on the floor sooner and take on more responsibility as their skills develop.
Nucleus Research surveyed 400 manufacturing and distribution leaders and found that nearly half of the organizations it classified as leading adopters were using digital on-the-job training. More than 70 percent of those organizations said new employees could work independently at the expected quality level within four weeks.
One manufacturer in the Nucleus study reduced the time required for employees to reach full operational competency from approximately 300 working days to 125. The equipment was complex and the traditional model kept people in training until they were certified before putting them on the floor. With a connected-worker platform, they could train on the job, working through the same procedures they’d use after becoming fully qualified.
Engaging operators
“Knowledge capture” can sound ominous to someone who spent decades mastering a job. A veteran worker may reasonably ask whether management is trying to learn everything they know before replacing them with somebody cheaper. Companies need to take that concern seriously if they expect people to participate.
Don’t sit an experienced worker at a desk and ask them to recite the manual. Record them doing the job and explaining the decisions along the way. Then let the workers who use those instructions correct a bad step or recommend a better one. If their suggestions disappear into a black hole, they’ll stop making them.
Many companies foster a culture of knowledge sharing across facilities. When one facility develops a more effective process, that improvement can be adopted elsewhere. Employees are recognized for their contributions and the company benefits from faster, more consistent problem-solving.
A skills-based system addresses a related problem. Companies often organize people around job titles without knowing what else they can do. I’ve seen a company eliminate a department, then discover that those employees had precisely the cross-skills it needed somewhere else. It had to hire new people to replace capabilities already present in the workforce.
To that end, a skills-based system would’ve shown what was there. It can also help workers receive credit for capabilities the company might otherwise overlook. Top performers get frustrated when the company doesn’t know what they can do.
On the question of monitoring, the team has to trust its employees. Otherwise, why hire them? The most successful implementations use information to create independent teams, not to delegate every decision from the top or turn every digital record into potential evidence.
Limiting access
Where employee protections limit who can see individual activity or skills data, the answer begins with strict controls over access. I encountered this with the German Works Council, which imposed clear restrictions on who could view the information and from where.
Once those rules were reflected in the system, the conversation moved to implementation. Involving employees and their representatives early also helps them understand what the tool is for and how the information will be used.
Rollouts fail for a few consistent reasons. Some companies begin without a specific problem to solve. Others try to do too much at once. And some lack genuine leadership commitment – not just support in principle, but a belief that the change matters and deserves the company’s time.
A carpet manufacturer took that approach with a changeover that was causing too much downtime. It documented the procedure and tested it on one line during one shift. When the changeover time fell, the company introduced the same process on the other shifts and eventually at other locations. It reported saving $500,000 in the first year.
Raising the stakes
As AI is added to the shop floor, the connected worker becomes more critical, not less. I’ve started thinking of each AI tool as an extremely confident junior team member. It needs validated procedures and clear boundaries, just as any new hire does.
In a laser operation, for example, an AI assistant shouldn’t improvise a cutting parameter, maintenance procedure or response to a machine alarm simply because its recommendation sounds reasonable. Its answers must be grounded in validated machine documentation, approved procedures and the company’s own operating knowledge.
If a facility hasn’t captured the right guidance for its workers, it hasn’t captured it for its AI agents either. Guidance that merely sounds plausible can create a quality problem or safety hazard.
The future of the connected worker isn’t about watching people more closely. It’s about preserving what the company knows, getting new employees productive faster and giving everyone enough support to act without waiting for an expert to show up. For manufacturers facing retirements, turnover and constant operational change, that’s more than another management tool. It’s a survival strategy.
As AI continues to push the boundaries of what manufacturers are capable of, stay on top of the latest developments by visiting our training and education archive.
