A 2025 S&P Global study reveals significant gaps in power measurement among data centre operators, risking operational stability as AI workloads intensify and energy demands grow volatile.
A significant minority of data centre operators lack the visibility considered essential for scaling infrastructure to support AI, according to a late‑2025 S&P Global 451 Research brief commissioned by Janitza. The study of 208 industry professionals found that roughly one quarter do not record power consumption at their principal sites, a gap the report frames as a structural business risk as AI workloads drive more volatile and concentrated energy demand.
Industry respondents reported a wide range of power usage effectiveness (PUE) outcomes. Just over half indicated PUE values between 1.5 and 2.0, while 23% acknowledged they do not monitor PUE at all. According to the S&P Global brief, the absence of decision‑grade, real‑time energy data undermines operators’ ability to perform predictive maintenance, plan capacity, and protect equipment life‑cycles, factors that directly affect uptime and return on investment for high‑density installations.
The brief and supplementary material from Janitza highlight how modern AI workloads produce rapid power swings, industry estimates cited in the research point to fluctuations of 40–70% within milliseconds, placing new stresses on power distribution and cooling assets. “In an environment where milliseconds matter, flexibility and data expertise are the critical differentiators,” the report states. Those transient peaks and troughs, the analysis warns, increase exposure to power‑quality events that can precipitate outages or accelerate component failure.
For operators targeting rack densities that are climbing into the 40–120 kW range, the research stresses comprehensive monitoring across the full electrical chain, from utility interface to sub‑distribution and individual racks, as a competitive necessity. Janitza says granular waveform capture, harmonics analysis and residual current detection can reveal emerging faults earlier, enabling intervention before incidents force unplanned downtime. The company points to commercial examples in which modular monitoring cut commissioning time and improved availability metrics, noting that such capabilities also support compliance with normative specifications such as EN 50160 and IEEE 519.
Janitza further markets compact metering hardware and software suited to these tasks, including DIN‑rail devices that measure multiple current channels and harmonics up to the 31st order, and GridVis® analytics for continuous power‑quality assessment. While these products are presented as tools to manage risk and boost energy efficiency, the report underscores a broader point for facility managers and owners: without live, high‑resolution energy telemetry, investment in AI‑optimised space and cooling may be difficult to justify or to operate safely at scale.
For stakeholders focused on industrial decarbonisation, the findings carry dual implications. Accurate, timely energy measurement is a prerequisite both for protecting capital equipment under dynamic loads and for validating the emissions and efficiency gains claimed by green infrastructure projects. Government figures and corporate ESG programmes alike increasingly demand verifiable energy performance; the study suggests many operators are not yet positioned to supply that evidence.
The S&P Global brief concludes that as AI workloads continue to concentrate compute and power density, operators who embed robust electrical measurement and analytics into their estates will be better placed to extract value, reduce operational risk and meet evolving regulatory and sustainability expectations. Janitza, which commissioned the study and offers related measurement solutions, frames the capability gap as an urgent commercial and technical priority for data centre owners and integrators.
- https://dcnnmagazine.com/infrastructure/power-cooling/one-in-four-dc-operators-fails-to-track-energy-usage/ – Please view link – unable to able to access data
- https://www.janitza.com/en-us/news/s-and-p-global-business-impact-brief-the-ai-era-exposes-a-gap – A recent S&P Global Business Impact Brief highlights that one in four data centre operators fail to monitor power usage data. The brief underscores the challenges posed by AI workloads, which place unprecedented pressure on electrical systems. Without real-time visibility into power usage and quality, operators risk performance constraints, equipment degradation, and reduced long-term returns. The brief emphasizes the necessity of decision-grade power data for managing next-generation data centres effectively. ([janitza.com](https://www.janitza.com/en-us/news/s-and-p-global-business-impact-brief-the-ai-era-exposes-a-gap?utm_source=openai))
- https://www.janitza.com/en/news/ai-data-centers-power-quality-more-critical-than-ever – An article by Janitza discusses the critical importance of power quality in AI-driven data centres. It notes that rising energy demands and rapid load fluctuations are creating unprecedented challenges for utilities and operators. The piece emphasizes the need for live energy data to ensure high availability and prevent unscheduled shutdowns. It also highlights the necessity of monitoring power quality according to normative specifications to detect weaknesses and avoid downtime. ([janitza.com](https://www.janitza.com/en/news/ai-data-centers-power-quality-more-critical-than-ever?utm_source=openai))
- https://www.janitza.com/en-us/products/umg-806 – Janitza’s UMG 806 is a compact DIN rail energy measurement device designed for precise monitoring of energy flows and power quality. It features four current measuring channels, residual current detection, and harmonics current up to the 31st harmonic. The device aims to help operators identify deviations, recognize sources of interference, and proactively increase the efficiency and protection of their systems. ([janitza.com](https://www.janitza.com/en-us/products/umg-806?utm_source=openai))
- https://www.janitza.com/en-us/products/gridvis/industries/data-center – Janitza’s GridVis® software offers solutions for monitoring electrical high availability in data centres. It enables timely recognition of voltage events, helping to detect weaknesses and prevent unscheduled shutdowns. The software supports compliance with normative specifications such as EN 50160 and IEEE 519, providing prepared power quality reports and tools for evaluating electrical high availability. ([janitza.com](https://www.janitza.com/en-us/products/gridvis/industries/data-center?utm_source=openai))
- https://www.janitza.com/en-us/know-how/application-reports/short-reference-faster-data-center-commissioning – A case study by Janitza demonstrates how a global technology and social media leader accelerated data centre commissioning by 44% using modular power monitoring solutions. The study highlights the importance of readily available products and expert on-site support in reducing deployment times and mitigating risks of unplanned outages. The solution provided live visibility into waveform capture, harmonics analysis, and multi-circuit monitoring, enhancing the reliability and efficiency of the data centre. ([janitza.com](https://www.janitza.com/en-us/know-how/application-reports/short-reference-faster-data-center-commissioning?utm_source=openai))
- https://www.janitza.com/en-us/services/datacenters – Janitza offers comprehensive measurement technology solutions for data centres, focusing on energy management, power quality, and residual current monitoring. The services aim to provide a comprehensive overview of the energy supply, ensuring energy efficiency, safety, and high availability. The solutions are designed to meet the tough requirements of modern data centres, supporting compliance with various standards and facilitating the development of energy, residual current, and power quality monitoring systems. ([janitza.com](https://www.janitza.com/en-us/services/datacenters?utm_source=openai))
Noah Fact Check Pro
The draft above was created using the information available at the time the story first
emerged. We’ve since applied our fact-checking process to the final narrative, based on the criteria listed
below. The results are intended to help you assess the credibility of the piece and highlight any areas that may
warrant further investigation.
Freshness check
Score:
8
Notes:
The article references a late-2025 study commissioned by Janitza and published in early 2026. The study’s findings are corroborated by a press release from Janitza dated February 4, 2026. ([janitza.com](https://www.janitza.com/en-us/news/s-and-p-global-business-impact-brief-the-ai-era-exposes-a-gap?utm_source=openai)) The article was published on March 16, 2026, indicating timely reporting. However, the study’s findings have been reported by other sources, such as S&P Global’s coverage in October 2025. ([spglobal.com](https://www.spglobal.com/energy/en/news-research/latest-news/electric-power/101425-data-center-grid-power-demand-to-rise-22-in-2025-nearly-triple-by-2030?utm_source=openai)) This suggests that while the article is recent, the core information has been available for several months.
Quotes check
Score:
7
Notes:
The article includes direct quotes from the Janitza press release. However, these quotes are not independently verifiable through other sources. The lack of external verification raises concerns about the authenticity and accuracy of the statements.
Source reliability
Score:
6
Notes:
The article is published on DCNN Magazine, a niche publication focusing on data centre and network news. While it provides industry-specific insights, its limited reach and potential biases may affect the reliability of the information presented.
Plausibility check
Score:
8
Notes:
The article discusses the challenges data centre operators face in monitoring power usage, particularly with the rise of AI workloads. This aligns with broader industry concerns about energy consumption and infrastructure demands. However, the specific claim that one in four operators fails to track energy usage is not independently verified, which diminishes the overall credibility of the claim.
Overall assessment
Verdict (FAIL, OPEN, PASS): FAIL
Confidence (LOW, MEDIUM, HIGH): MEDIUM
Summary:
The article presents timely information on data centre operators’ challenges in monitoring power usage, referencing a recent study. However, the reliance on a single, unverified source (Janitza’s press release) and the lack of independent verification for key claims significantly undermine the article’s credibility. The absence of corroborating evidence and potential biases associated with the source warrant a cautious approach to publishing this content.
