What Is Planned vs Reactive Maintenance? Definitions, Ratios and What Good Looks Like
The planned vs reactive maintenance ratio, defined precisely: the three denominators, what the 80/20 benchmark is worth, and how a good number hides bad work.
What Is Planned vs Reactive Maintenance? Definitions, Ratios and What Good Looks Like
The planned vs reactive maintenance ratio expresses the share of maintenance effort planned before execution against the share raised in response to failure. It can be calculated on work order count, labour hours or cost, and the three denominators rarely agree. SMRP defines the labour-hours version.
The definitions, and the one people get wrong
BS EN 13306:2017 defines maintenance at clause 2.1 as the "combination of all technical, administrative and managerial actions during the life cycle of an item intended to retain it in, or restore it to, a state in which it can perform the required function". Its type definitions sit in clause 7, behind the paywall, so the wording below is descriptive.
Preventive work happens before failure. Predetermined work fires on elapsed time or a usage counter (the PPM schedule). Condition-based work fires on a measured parameter: vibration, thermographic delta, filter differential pressure, oil analysis. Predictive work extrapolates that measurement to a remaining useful life rather than crossing a threshold. Corrective work restores function after a fault, splitting into immediate and deferred. Emergency work is the immediate subset that also breaks the schedule. Deferred maintenance, in the estates sense, is work consciously postponed and now costed in the backlog.
The error that ruins most ratios is treating "planned" and "preventive" as synonyms. Planning is a state of readiness: scope written, job plan attached, labour estimated, parts reserved, permits identified, access arranged. A corrective repair raised off a PPM route on Sunday, planned Monday, executed Wednesday is planned corrective work. A PPM task issued on schedule but attended by a technician with no filter in the van is preventive and unplanned; under the SMRP definition it is reactive, because it interrupted the schedule.
How the ratio is actually computed
The formula is trivial. The denominator is not.
| Denominator | Computation | What it flatters | Honestly good for |
|---|---|---|---|
| Work order count | Planned WOs closed / all WOs closed | Sites with many short tasks; a meter read counts as a chiller overhaul | Schedule discipline, not effort |
| Labour hours | Planned labour hours / total labour hours | Sites booking generously to PPM, thinly to callouts | Crew sizing. The SMRP basis |
| Cost | Planned cost / total maintenance cost | Sites whose reactive work is labour-heavy, material-light | Budget and service-charge defence |
SMRP defines the hours-based view. Metric 5.4.1 (Reactive Work) is "maintenance work that interrupts the weekly schedule, calculated as a percentage of the total maintenance labor hours". Metric 5.4.2 (Proactive Work) is (PM + PdM + corrective arising from PM and PdM) over total labour hours; 5.3.1 covers planned work on the same base. Formulas and best-in-class targets sit in the SMRP Best Practices compendium, members-only. BS EN 15341 is the parallel European set.
Run all three on the same month before choosing. A portfolio commonly reads 85 percent by count, 68 percent by hours, 55 percent by cost. All correct. Report one, and the same one every month.
Who should measure it, and at what level
Portfolio-level ratios belong on a board slide and nowhere else. Compute per asset class and per building: lifts and BMS-connected plant carry a different planned share to fabric and fit-out, and multi-tenant retail carries structurally more reactive work than a single-occupier office. Service providers on output-based contracts should split landlord scope from tenant scope, or churn moves a KPI the team does not control.
The 80/20 benchmark, and who published it
The near-universal claim is that world-class maintenance runs at roughly 80 percent planned and 20 percent reactive. No primary published study behind it could be found. It circulates in vendor knowledge bases and conference decks, unattributed and often circular. A published attempt to trace the related "80 percent of downtime from 20 percent of assets" claim concluded it migrated from Pareto and Juran's quality work "in trade content and sales material, without a dataset attached", rating confidence low.
Treat 80/20 as a convention, not a standard. The defensible targets have a named publisher and a stated formula: SMRP and EN 15341. If a contract sets 80 percent planned as a KPI, get the denominator written into the contract, or the number is unarbitrable.
What a high planned ratio hides
Anything above 90 percent should prompt an audit, not a celebration.
- Padded PPM schedules. Low-value visual inspections raise planned count and hours without touching failure rates. Check PM yield: SMRP metric 5.4.12 notes top performers generate about one hour of corrective work per hour of PM, and about 2.5 per hour of PdM. A programme that finds nothing is decorative.
- Retrospective work orders. A 22:00 callout logged next morning as planned. Compare creation timestamp against reported fault time; a job created after work started is reactive whatever its type code.
- Closure without execution. PPM closed in bulk at month end; look for clustering in the final 48 hours.
- Splitting. One failure raised as several work orders, inflating counts.
Reading it alongside backlog and reactive cost
The ratio is a rate, not a level. Read it with backlog in crew-weeks (outstanding estimated hours over weekly capacity, split into planned and ready backlog, SMRP metrics 5.4.8 and 5.4.9) and reactive cost as a share of total maintenance cost. A rising planned ratio with rising backlog and flat reactive cost means work is being pushed, not prevented.
Related terms
- PPM (planned preventive maintenance). The time or usage-triggered schedule; EN 13306's predetermined subset.
- PPM compliance. Share of scheduled tasks completed inside their window. High compliance with a low planned ratio means the schedule is too small.
- MTBF and MTTR. Operating time per failure, and mean active repair time per failure. MTTR moves with parts and job plans, not with more PPM.
- Backlog. Outstanding approved work in crew-weeks rather than job count.
- Condition-based maintenance. Preventive work triggered by a measured parameter, not the calendar.
- Wrench time. Share of paid time spent working at the asset, SMRP metric 5.6.1. It caps how much planned work a fixed crew can absorb.
FAQ
Is 100 percent planned maintenance the goal? No. Corridor luminaires, door closers and low-cost sanitaryware are cheaper to run to failure, and run-to-failure still produces reactive work orders. The target is that reactive work results from a decision rather than a surprise.
Should tenant-reported faults count as reactive? Yes, where they are maintenance on landlord-scope assets. Exclude churn, fit-out support and chargeable tenant requests from both numerator and denominator, or the ratio moves for reasons unrelated to asset condition.
How long before a new PPM programme moves the ratio? It usually gets worse first, because new inspections surface defects. If they do not, the inspections are finding nothing. Judge on backlog trend and failure rate, not the ratio in month two. Comparisons across buildings only hold within asset class.
Where a system helps
A CAFM or CMMS earns the ratio by making the planned state auditable: creation timestamp against reported fault time, job plan attached before issue, parts reserved, and a closure record that cannot be quietly backdated. That is the difference between a number you can defend to an owner and one you can only assert. FaciOS is built around that lineage.
Related reading: Muscat Municipality Requirements for Commercial Buildings: Operating Compliance (KB-010).
Sources
- BS EN 13306:2017 Maintenance. Maintenance terminology, preview pages (clause 2.1 quoted; clause 7 type definitions are not in the public preview)
- BS EN 15341:2019+A1:2022 Maintenance Key Performance Indicators, preview pages
- SMRP Best Practices, Metrics & Guidelines (70+ metrics with formula, sample calculation and best-in-class target; members-only)
- SMRP Best Practices Metrics Workshop, 2018 (metrics 5.3.1, 5.4.1, 5.4.2, 5.4.8, 5.4.9, 5.4.12, 5.6.1)
- Does 80% of Downtime Come From 20% of Assets? Tracing the Claim
[NEEDS SOURCE: original publisher of the "80 percent planned / 20 percent reactive" world-class benchmark. No primary study located; retain the caveat until one is found.]
[NEEDS SOURCE: exact best-in-class target values stated against SMRP metrics 5.3.1, 5.4.1 and 5.4.2 in the current edition of the SMRP Best Practices compendium. Verify from the members-only download before quoting any figure.]