Wilo has found that 74% of UK data centre operators have yet to adopt fully predictive maintenance strategies, according to a survey of 300 professionals in the sector.
The report points to continued reliance on reactive and preventative maintenance as operators face growing pressure to improve resilience, efficiency and sustainability. Respondents said their facilities lost an average of six hours to unplanned downtime over the past year.
Unexpected asset breakdowns were identified as a leading operational problem by 45% of those surveyed. A further 44% cited a lack of actionable monitoring insights, while 41% said they do not stock critical spare parts.
The data also suggests many operators have limited visibility into the performance of key cooling assets. While 91% of respondents said pumps are essential to cooling performance and facility resilience, only 31% use condition-based monitoring, and another 31% rely on periodic checks.
Two-thirds of respondents said they do not continuously measure pump energy efficiency. This can hinder efforts to improve power usage effectiveness and water usage effectiveness, two key metrics for facilities seeking to reduce energy use and water consumption.
Adoption barriers
The survey identified several barriers to wider use of predictive maintenance. Skills shortages were cited by 55% of respondents, data integration challenges by 54%, and uncertainty around return on investment by 46%.
These pressures come as data centres face rising expectations around uptime and operational resilience. The report argues that maintenance practices have not kept pace with the sector's growing importance to the wider economy.
Alice Oakes, service and support manager at Wilo, said the demands on facilities now require a different approach to asset management.
"Behind the performance of a data centre sits a complex ecosystem of assets that must operate seamlessly to maintain uptime. With the sector's recent recognition as critical national infrastructure, many facilities could benefit from upgrading their maintenance approach to meet the demands of a more challenging era."
"Predictive maintenance is a fundamental requirement for resilience, efficiency and sustainability in modern data centres. It is therefore crucial to have solutions in place that help operators overcome barriers to adoption and ensure their facilities are equipped for the future. For those seeking to avoid unplanned downtime, reactive maintenance for critical assets is simply obsolete, and preventative approaches are close behind," Oakes said.
The report describes condition-based monitoring as the use of real-time data from assets such as pumps, motors and cooling systems to identify issues before failure. Fully predictive maintenance builds on that by linking monitoring data to alerts, maintenance scheduling and spare parts planning.
This matters because downtime in data centres can have direct operational and financial consequences. Even relatively short outages can disrupt customer services, increase repair costs and place greater strain on already stretched engineering teams.
Shift in approach
Despite low current adoption, the survey indicates that many operators expect to change course soon. Among respondents without a predictive maintenance strategy, 87% said they intend to adopt one within the next six months.
This points to a potential shift in how operators manage physical infrastructure, particularly as facilities scale to meet growing demand from cloud computing and artificial intelligence workloads. More intensive computing environments can place additional pressure on cooling systems and supporting assets, making failures harder to absorb.
For suppliers and operators alike, the findings underline a gap between awareness and implementation. Most respondents recognise the importance of key equipment such as pumps, but fewer have systems in place to continuously track efficiency or predict faults before they disrupt operations.
Oakes said operators need support to move from older maintenance models to more data-led approaches.
"Operators are encouraged to begin with a free energy audit from a trusted partner like Wilo to assess the potential return on investment and sustainability gains from long-term energy savings and carbon reduction. We also advise on the best time to upgrade pumps and how this can positively affect uptime."
"Beyond that, we can support predictive maintenance strategies and the implementation of condition-based monitoring systems, while also helping facilities navigate the industry's well-documented skills shortage through targeted training, data integration and supplier guidance," Oakes said.