Planned vs reactive maintenance: getting the balance right

Every facilities team runs on some mix of planned and reactive work, and every facilities manager has, at some point, been told they should be spending more time on the planned side. That advice isn't wrong, exactly, but it treats the split as a dial you can just turn up, when in practice reactive work has a way of claiming whatever hours exist regardless of how the week was meant to go.

Why the ratio itself is the wrong target

Industry commentary sometimes floats a target ratio — some fixed percentage of hours on planned work versus reactive — as if hitting a number solves the underlying problem. It doesn't. A building with an ageing asset base and a lean team will always run more reactive than a newer estate with a full complement of engineers, and chasing a borrowed ratio can push a team toward hitting a number rather than fixing what's actually driving the reactive load.

The real distinction: predictable reactive vs chaotic reactive

A more useful way to look at it is whether reactive work is predictable or chaotic. A team that knows, roughly, how many reactive jobs a typical week brings and has built capacity for that into the plan is in a genuinely different position to one where reactive work arrives as a constant, unplanned interruption to everything else. The first team has effectively absorbed reactive work into their planning; the second is being run by it.

Where the reactive load usually comes from

A high reactive load is very often the downstream cost of PPM tasks that got skipped or delayed rather than a fixed fact about a building. An asset that gets its planned service on schedule fails less often between visits; one that's been quietly slipping for months eventually turns into an emergency call. Reactive work isn't a separate category from planned maintenance so much as a bill that comes due when planned maintenance gets deferred.

• –Assets with genuinely high failure risk should carry more planned attention, not less, even if that feels counterintuitive when the schedule is already tight
• –A rising reactive count on a specific asset or building is a signal worth investigating, not just absorbing as normal wear
• –Response time targets for reactive work should be realistic and written down, not implicit — see our notes on setting reactive response times
• –Contractor-dependent reactive jobs need their own lead-time buffer, since a contractor's availability isn't within the facilities team's direct control

Setting expectations with whoever you report to

A facilities manager reporting upward on maintenance performance is often implicitly being asked "why isn't this all planned by now," and the honest answer is that zero reactive work isn't a realistic goal for most estates — some asset criticality and building age combinations will always generate genuine surprises. The more useful conversation is about trend: is the reactive load stable, growing, or shrinking relative to the size of the estate, and is the team able to explain why.

Getting that trend visible is mostly a reporting problem rather than an operational one — a team can be doing genuinely good work and still struggle to demonstrate it if reactive and planned jobs sit in disconnected records that were never designed to be compared side by side.

Two teams with the same ratio

Two estates both report seventy per cent planned, thirty per cent reactive. In the first, the reactive work is largely predictable — lamp changes, door adjustments, the same three rooms complaining about temperature every autumn — and it is scheduled into the week as known load. In the second, the same thirty per cent arrives as unplanned interruption: plant failures nobody saw coming, each one displacing planned work that then never gets rescheduled.

The ratio is identical and the two estates are in completely different health. It is the reason the ratio makes such a poor target: driving it down by deferring planned work looks like an improvement for about two quarters, and then the deferred work becomes the reactive load.

Common mistakes

• –Targeting a ratio without distinguishing predictable reactive work from genuine failures
• –Counting a planned task as done when it was deferred rather than completed
• –Never analysing which assets generate the repeat reactive calls
• –Reporting the ratio upward without the context that explains it
• –Cutting planned work to hit a number, which raises reactive load a year later
• –Treating a rising reactive share as a team performance problem rather than an asset condition signal