While 82% of maintenance teams use a CMMS, company-wide adoption can remains low, with only 39% noting consistent use across their entire team. In the 2026 State of Maintenance Report, Limble surveyed 686 maintenance and operations professionals to find trends in CMMS use and adoption, artificial intelligence and generational attitudes to technology. 

Why do workers not use a CMMS?

Reasons why a CMMS may not be used include:

• Technicians were never properly trained on it (47%).
• It requires a desktop and is not usable in the field (35%).
• The system takes longer than the workaround (26%).
• Its use is framed as optional (22%).
• The data inside it is outdated (13%).

One big question is how to integrate AI into systems on top of operational knowledge when much of that knowledge has never been inputted into the system at all. Underlying maintenance data may not keep pace with where it needs to be to get the most out of AI.

Meanwhile 55% of surveyed maintenance teams use AI “in some form.” For 24%, that use is common across operations, while 31% are trying AI out in a few specific areas.

“People are building discipline around how to leverage AI and use AI, but they're not being disciplined in using the system to put data into it, so that the AI actually produces something that's valuable,” says Limble CEO Gary Specter in an interview with Plant Services.

Inputting data regularly and thoroughly prevents important interpersonal information from falling through the cracks. The most personal useful networks—who knows who to ask to solve a problem, and that person’s knowledge—may never be recorded.

To do so, tell workers why the CMMS or AI system benefits them. For example, it could save time.

“By leveraging a system, you don't have to run around looking for me to approve [something],” Specter says. “If you put it in the system, I can approve it on my phone. You don't have to come find me. If you're looking for certain parts, instead of spending 60% of your time looking for parts, you can go into the system and see where the parts are located. With some systems you can have the part actually brought to the machine that you're working on.”

According to the survey, 51% of technicians say waiting on approvals cuts into maintenance time. In addition, 41% say equipment records that are incomplete or out of date cut into maintenance time. Using a CMMS for approvals and logging equipment records more consistently means more time performing maintenance. However, barriers to doing so may be as simple as the CMMS not having a field for the information, the report points out.

AI adoption isn’t ideal in all situations

Environments in which data isn’t collected will not be good environments for taking advantage of AI.

“The problem with AI is dirty data in, dirty data out,” says Specter.

Specifically, if your technicians don’t use the AI-connected system, or record their data elsewhere, the AI analysis will be incomplete.

This incomplete data will not help maintenance operations by explaining why a system failed or which parts were broken. The lack of actionable results can create a disconnect between everyday work on the shop floor and business leaders who push for AI adoption. 

Make effective use of your workforce 

Worker turnover is another good reason to make sure as much information as possible translates from workers’ brains into the CMMS. According to Limble, 74% of maintenance organizations have at least one critical asset that only one person knows how to fix, and 47% of surveyed workers say they lost irretrievable knowledge when someone left the company.

Generational differences are also a factor. Younger people are less likely to trust AI, with 41% citing lack of confidence in AI as a reason not to use it.  

On the other hand, Gen Z wants to know the manufacturing companies they might join are high-tech. Younger workers may choose not to work in organizations that don’t have a CMMS or mobile capabilities, Specter says.  

He made sure to point out that automation does not mean maintenance and reliability jobs will disappear. 

“Somebody still has to maintain the automation,” he says. “So, even if you have a robot, or now something that someone used to do is automated, the thing that's automating it still has to have a maintenance record.”