NEWRecognition — recognise hard work and the people going above and beyond
FacilityOptixFacilities software
Guide9 min read · Updated 23 September 2026

Building a PPM schedule that covers your statutory duties.

Most teams do not lack a PPM schedule. They lack one that gets finished, and one that can be shown to cover the duties they actually hold. Here is how to build both at once.

A planned preventative maintenance schedule has two jobs, and most schedules only do one. The first is to make sure work happens before something fails. The second — the one that decides whether you are defensible after an incident — is to demonstrate that the work a duty holder is legally expected to arrange was arranged, carried out, and recorded. A spreadsheet can do the first badly. Only a system with evidence attached does the second at all.

This guide is about building a schedule that does both. It assumes you already know your estate reasonably well, and it is written for the person who will have to defend the regime rather than the person who will draw it.

Start from the register, not a template

Search for a PPM schedule template and you will find dozens. Every one of them is a list of asset types with suggested intervals, and every one of them is wrong for your estate in a specific way you cannot see from the outside. A template does not know that your air handling unit runs twenty-four hours in a building that never closes, or that your second boiler has been isolated since 2019 and is still on the register.

The honest starting point is an accurate asset register — what you actually have, where it is, what condition it is in, and how critical it is. Build the schedule from that and the frequencies are decisions you have made. Build it from a template and the frequencies are decisions someone else made about a building they have never seen, which is difficult to stand behind when asked.

If the register is not in a state you would build on, do that first. It is covered separately in our guide to building an asset register that is actually useful, and it is worth the delay — everything below is only as good as the list it runs against.

Separate statutory from discretionary before you set a single date

This is the division that makes a schedule defensible, and it is the one most often skipped because it is tedious. Some maintenance exists because a manufacturer recommends it, a warranty requires it, or it is simply good practice. Other maintenance exists because a duty holder is legally expected to arrange it, and the frequency is set by regulation, an approved code of practice or a competent person's written scheme rather than by you.

Those two categories should never be allowed to blur. When a week goes badly and something has to slip, a discretionary service moving by ten days is a judgement call. A statutory examination moving past its date is a different kind of event entirely, and the person deciding on a Friday afternoon needs to be able to tell them apart at a glance.

This is the specific job SFG20's criticality classification does, which is the strongest argument for holding a subscription if you do not already — see SFG20 and maintenance standards, explained for what it does and does not give you.

  • Mark every routine as statutory or discretionary on the schedule itself, not in a separate document
  • For statutory items, record where the frequency comes from — the regulation, the standard, or the written scheme of examination
  • Never let a statutory item inherit a frequency from a template without checking it against the source
  • Record who the competent person is for routines that require one

The duties most estates carry

Which duties apply depends on your assets, your occupancy and your role — a point worth repeating because it is the one people skip. That said, the routines most commercial estates end up carrying cluster in a fairly predictable set, and knowing the shape of them helps you notice what is missing from a schedule you inherited.

  • Water hygiene — risk assessment under ACOP L8, with monitoring in line with HSG274; typically temperature checks on sentinel outlets and a flushing regime for little-used ones
  • Lifting equipment — thorough examination under LOLER 1998, commonly at six-monthly intervals for equipment lifting people and twelve-monthly otherwise, or in accordance with an examination scheme
  • Fixed electrical installation — periodic inspection and testing to BS 7671, at an interval recommended by the inspector and recorded on the report
  • Emergency lighting — routine function testing and an annual full-duration test, in line with BS 5266
  • Fire alarm systems — regular user testing and periodic servicing by a competent person, in line with BS 5839
  • Gas appliances and installations — maintained in a safe condition, with work carried out by a suitably registered engineer
  • Asbestos — the duty to manage under CAR 2012, including keeping the management plan and the condition of known materials under review

Treat that list as a prompt rather than a specification. Intervals change, guidance is revised, and several of these are set by a risk assessment rather than by a fixed number — which means the right frequency for your building may not be the common one. The longer version, organised as a calendar, is in our guide to the statutory compliance calendar for commercial buildings.

Set frequency by consequence, not habit

For everything that is not fixed by regulation, the useful question is not how often this asset type is usually serviced. It is what actually happens if this specific asset fails between now and the next planned visit. For a lot of plant the honest answer is that somebody is inconvenienced for a day. For a small number of items the answer involves a safety consequence, a service that cannot be delivered, or a building that has to close.

Recording criticality against the asset and carrying it into the schedule is what turns that thinking into something operational. Without it, criticality lives in a risk register nobody opens during a busy week, which means it changes nothing at the moment it was supposed to matter.

Build in slack, deliberately

A schedule that assumes every week runs at full capacity fails the first time someone is off sick or a reactive job eats the morning. That is not a discipline problem, it is an arithmetic one. Most discretionary tasks can tolerate being done a few days either side of their date without becoming a problem, and stating that tolerance explicitly is more honest than pretending every date is rigid and then quietly missing a third of them.

Statutory items are the exception, and that is precisely why they need to be marked. A tolerance you have thought about and written down is a plan. A tolerance you discover by missing dates is a pattern, and it will be read as one.

Decide what happens to a missed task

This is the single best predictor of whether a schedule survives contact with a real year. If a task slipping past its date just sits there — no owner, no escalation, no forced follow-up — it will keep slipping, and the schedule will drift from a plan into a record of intentions. The fix is not more discipline from the same team running the same process. It is making the overdue item impossible to ignore.

  • Overdue shown separately from upcoming, rather than blended into one long list
  • An owner that is a role as well as a person, so annual leave does not orphan a task
  • A standing review where the overdue list is the agenda rather than an appendix
  • A closing reason recorded when a task is stood down, so the gap is explained rather than silent

Capture the evidence with the work, not afterwards

A schedule proves nothing on its own. What proves the regime ran is the completion record: who attended, when, what they found, and the certificate or report that came out of it. Evidence collected at the point of work is contemporaneous and specific. Evidence assembled three weeks later from memory and an inbox is neither, and the difference is visible to anyone reading it seriously.

This is also where a defect found during a planned visit either becomes a job or becomes nothing. A visit that identifies a problem and produces a PDF has done half the work; linking the finding to the work order that resolves it is what closes the loop and keeps the trail intact from discovery to repair.

Review the schedule annually, against reality

A PPM regime set once and never revisited slowly stops describing the building. Plant is replaced, occupancy changes, a wing is closed, a new duty appears. An annual review that compares what was scheduled against what was actually completed — and asks why, for every material gap — is what keeps the schedule honest. It is also, not incidentally, the review an auditor will ask whether you carry out.

Key takeaways
  • Build the schedule from your own asset register, never from a downloaded template.
  • Mark statutory routines separately from discretionary ones, and record where each frequency comes from.
  • Set discretionary frequencies by the consequence of failure, not by habit or manufacturer defaults.
  • Build tolerance in deliberately, and treat statutory dates as the exception that has none.
  • Decide in advance what happens to a missed task — an overdue item with no owner will keep slipping.
  • Capture completion evidence at the point of work; a schedule without evidence proves nothing.
  • Review annually against what was actually completed, and explain every material gap.

Holding this together in one place

FacilityOptix holds the register, the schedule, the jobs and the evidence against the same buildings — one licence per company, priced on properties rather than users.

The FacilityOptix team

Written by people who work daily with facilities teams on planned maintenance, statutory inspection and the records that hold up under an inspection. This guide explains a duty; it is not advice from a competent person for your building.

See it running on your own buildings.

A live demo on your data, not a slide deck — most teams are live within a week.

Try for free