Every guide, in one list

A maintenance plan built on an inaccurate list of assets is a plan for a building that does not exist.

Most planned schedules are arithmetic problems that were never done. The hours required exceed the hours available, and the difference is absorbed silently. For a separate product-side perspective on approaches to planning and managing time, the linked resource provides additional detail.

Most intervals are inherited rather than chosen. Some are legally fixed, some are manufacturer caution, and some are nobody's decision at all.

When the schedule is under pressure, something gets dropped. If the two categories are not distinguished in the system, the wrong thing gets dropped. For a wider operational and compliance reference, consult the Better Buildings O&M guide.

Heating fails in the first cold week and cooling fails in the first hot one, because both were last touched a long time before.

A shutdown is a fixed quantity of time in which a variable quantity of work has been promised. Overrunning is the default outcome.

A large share of maintenance time is spent working out what a job means. That time is charged to the job and blamed on the technician.

Priority set by the person reporting produces a queue in which everything is urgent. Priority set by rule produces one that can be planned.

Every schedule, every quotation and every capacity calculation rests on duration estimates that almost nobody checks against outcomes.

Most maintenance backlogs are a mixture of real work, duplicates, resolved jobs nobody closed, and things that will never be done. That mixture cannot be managed.

The closing note is where the maintenance history is written. Most of it reads 'completed', which is the same as writing nothing.

It is the most quoted number in maintenance and the easiest to move without improving anything.

A technician paid for eight hours delivers rather fewer on the tools. The gap is measurable, largely structural, and almost never measured.

Formal work sampling is disproportionate for most departments. Lighter methods give a good enough figure and do far less damage to trust.

On a multi-site portfolio, travel can consume more of the week than the work does, and it is usually invisible in the job records.

Waiting for a part, a key, an isolation or a room is the least visible category of lost time and often the easiest to reduce.

Time recorded at the end of the week from memory is fiction with a signature on it. The fix is capture at the point of work.

Mobile workforce products offer continuous location and activity monitoring. Very few maintenance questions require any of it.

Below a certain size, scheduling stops being an optimisation problem and becomes a search for any arrangement that satisfies the constraints.

Out-of-hours cover in a small department usually falls on whoever is least able to say no, and is frequently paid on a basis nobody wrote down.

Most disputes with maintenance contractors are traceable to a specification that described a visit rather than an outcome.

A four-hour response commitment is worthless without a definition of when the clock starts, what counts as a response, and what happens when it is missed.

Time-and-materials invoices are usually paid on trust, because verifying them after the fact is nearly impossible.

For statutory work, the wrong person doing the inspection correctly is the same as the inspection not happening.

Each contractor's schedule is efficient in isolation. The collisions between them are the client's problem and nobody else's.

A spreadsheet runs a maintenance schedule adequately for longer than vendors suggest. The thresholds that force a change are specific.

Maintenance software is demonstrated on dashboards and lived on the four screens a technician sees. Those are the ones to evaluate.

The environments maintenance happens in are precisely the ones mobile software is worst at, and vendors demonstrate on office wifi.

Most well-known time trackers were built for agencies billing clients at a desk. That origin explains where they fit maintenance and where they do not.

Trials are run in the quietest month, by the most capable person, on twenty tidy sample assets. That proves nothing about February.