The uncomfortable finding from the world's largest maintenance benchmark
The Society for Maintenance and Reliability Professionals has published a library of over 100 maintenance metrics. Their 2023 member survey found that the average maintenance organisation tracks 23 of them. The same survey found no statistically significant relationship between the number of metrics tracked and actual maintenance performance outcomes. There was, however, a strong positive relationship between the consistent use of six to eight core metrics and measurable performance improvement.
Maintenance cost as a percentage of replacement asset value
Maintenance spend divided by the total replacement cost of the plant's equipment provides a normalised view comparable across plants of different sizes. SMRP benchmark data: world-class manufacturers maintain this ratio below 2 percent. Industry average is 3.5 to 4 percent. Reactive-dominated environments often exceed 7 percent. The metric's limitation is that low maintenance cost is not inherently good — under-investment reduces cost today at the expense of reliability and asset life. The target is the ratio consistent with reliability requirements, not the lowest possible number.
The planned maintenance ratio — the leading indicator everything else follows
Planned maintenance as a percentage of total maintenance hours is the most actionable metric a maintenance organisation can track. SMRP benchmark: world-class is above 80 percent planned. Average is 50 to 60 percent. Below 40 percent, emergency work volume continuously displaces the planning function. Aberdeen Group's 2023 benchmark found that top-quartile performers share consistently low emergency work ratios (below 15 percent), high planned ratios (above 75 percent), and MTTR under 60 minutes for most equipment categories.
The relationship between planned maintenance ratio and productive time is explored in detail in our guide on why maintenance teams are only productive 29% of the time. The planned ratio is the lever that moves everything else.
MTTR tracked by equipment category — not just overall
MTTR at the equipment category level reveals where improvement opportunities lie. A plant where MTTR exceeds 120 minutes for hydraulic faults but stays below 45 minutes for electrical faults has a specific knowledge gap in hydraulic diagnosis — not a general maintenance performance problem. This granularity requires that work orders are assigned to equipment categories and fault types consistently, not just equipment IDs. Plants where work orders record only machine number and "repaired" as resolution cannot extract category-level insights from their data.
This is the same argument made in our guide on what makes a CMMS actually useful. The metric is only as good as the data feeding it.
Emergency work percentage — the most sensitive early warning indicator
Emergency work — unplanned work requiring immediate response that displaces planned activity — is the leading indicator of reactive culture. SMRP data shows that plants reducing emergency work below 15 percent sustain that position. The lower failure rate from a consistently executed PM programme self-reinforces the planned maintenance ratio. The reduction from 30 percent to 15 percent typically takes 18 to 24 months and is the single most impactful transition a maintenance organisation can make.
For context on what drives emergency work rates down, see our guides on MTBF and the interventions that actually move it and TPM implementation outcomes. The data points to the same conclusion: consistent PM execution, supported by good information at the moment of fault, is what breaks the reactive cycle.
YAFEX gives maintenance managers visibility into MTTR by equipment type, fault frequency, and resolution patterns from their plant's own work order history. Talk to us.
Ready to put this into practice?