Preventive Maintenance Log
The record of preventive maintenance that has actually happened. Every visit against the date it was due, so the sheet works out how many days early or late each one ran, which is the difference between a program that exists and one that is only written down.
- Days early or late, calculated from due against done
- Completion and on-time compliance reported separately
- A tolerance agreed before logging starts, not argued afterwards
- Missed PMs carrying a name, a reason and a rescheduled date
Preventive Maintenance Log
What happened, and how late
| # | Log row | Days | Result | Corrective WO |
|---|---|---|---|---|
| 1 | AHU-03, quarterly filter and belt service | |||
| 2 | PUMP-02, annual bearing inspection | |||
| 3 | RTU-01, semi-annual coil clean |
The document you will get. Download for the full, editable file.
Who this preventive maintenance log is for
A log is filled in by the people doing the work and read by people deciding whether the program is real.
The technician or supervisor keeping it
You are entering two dates and a result. The tolerance decides whether your three-day slip reads as on time or late, which is exactly why it is agreed up front rather than after the fact.
The planner reviewing intervals
Section 6 is yours. PMs that never find anything are candidates for a longer interval; PMs that always find something are candidates for a shorter one or a redesign.
Whoever answers for compliance
You need on-time compliance rather than completion, because the first is the only one that says the asset was protected at the interval it was supposed to be.
An auditor or incoming provider
You are reading history. Missed and skipped PMs, who authorized each and whether the risk was accepted tell you what kind of program you are inheriting.
Which section matters the most in your sector
Everyone logs completion. What differs is how much a late visit actually costs, and that is what decides whether on-time compliance is a nicety or the number that matters. If you run one of these, the sector page goes further than the template does.
- Data centers
On-time compliance. Where redundancy is the safety margin, a PM run three weeks late has quietly removed part of it for three weeks.
Data center maintenance software - Healthcare
Missed and skipped register. A skipped PM in a clinical area needs a named authorizer and an accepted risk rather than a blank cell.
Healthcare maintenance software - FM service providers
The two percentages together. Completion is what a client contract usually measures, and on-time is what actually protects you when something fails.
FM service provider software - Commercial real estate
Monthly summary. Compliance dips in the same months every year across a portfolio, and that is a resourcing pattern rather than a people problem.
Portfolio maintenance software - Retail and malls
Tolerance. Multi-site rounds slip by days routinely, so the agreed tolerance decides whether your compliance figure is meaningful or theatrical.
Retail maintenance software - Education
Months where compliance drops. Term-driven access makes the dips predictable, which makes them plannable rather than inevitable.
Campus maintenance software
What a preventive maintenance log should contain
A preventive maintenance log is the backward-looking record of PM visits that have actually taken place. Each row pairs the date a task was due with the date it was done, works out how many days early or late it ran, and records the technician, hours, result, deficiencies found and any corrective work order raised.
A. Fields specific to a preventive maintenance log
| Field | What goes in it | Why it earns its place |
|---|---|---|
| The agreed tolerance | How many days either side of due before it counts as late | Agreed before logging begins, because settling it afterwards means the compliance figure reflects whoever last argued about it rather than a standard. |
| What happens to a missed PM | Carried forward, canceled with the next occurrence standing, or escalated | Three defensible answers and one indefensible situation, which is having no rule and deciding case by case under pressure. |
| Due date and done date | Both real dates, entered separately | The pair is the whole instrument. A log that records only the date something was done cannot tell you anything about whether the interval was respected. |
| Days early or late | Calculated, positive for late and negative for early | The one column nothing else in the set produces. It turns a yes or no about completion into a measure of how well the schedule is actually being held. |
| Result | On time, late, missed or skipped, spelled exactly | Four states rather than two. Skipped with authorization and missed entirely are very different events, and averaging them hides the one that matters. |
| Technician and hours | Who did it and how long it took | Hours accumulate into the labor figure that makes a resourcing argument, and the name is what makes a pattern of late visits addressable. |
| Deficiencies found | How many, per visit | Feeds the find rate, which is the diagnostic most programs never calculate and the one that tells you whether an interval is right. |
| Corrective work order | The number raised from what the PM found | Without it a deficiency found on a PM is a note. With it the PM has generated work somebody owns, which is the point of inspecting at all. |
| Completion percent | Of the PMs logged, how many were done at all | The friendlier number and the less informative one. It says the work eventually happened, with nothing about when. |
| On-time compliance percent | Of those due, how many landed inside tolerance | The number that says whether the asset was protected at the interval intended. Reported alongside completion rather than instead of it. |
| Missed and skipped register | Why, who authorized it, when it is rescheduled, risk accepted | The authorizing name is the field that changes behavior. A missed PM with nobody against it will be missed again the same way. |
| What the log is telling us | Repeatedly late assets, dips by month, and intervals to change | The section that makes the log worth keeping beyond proving the work happened. PMs that never find anything and PMs that always do are both interval problems. |
Completion and on-time compliance answer different questions, and reporting only the first is the most common way a maintenance program flatters itself. Completion asks whether the work eventually happened. On-time compliance asks whether it happened at the interval that was chosen to protect the asset. A program running at ninety eight percent completion and sixty percent on-time is not a program in good health that is slightly behind; it is a program where four in ten intervals were not actually honored, and the intervals were the entire basis on which somebody decided the asset was safe to run. The distinction only exists if the due date and the done date are both recorded, which is why they are two columns rather than one, and it only means anything if the tolerance was agreed in advance. Settle that first, in writing, at whatever number suits your operation. A generous tolerance honestly applied is far more useful than a strict one everybody quietly negotiates around.
B. What it looks like filled in
Five rows from one month's log, with the month's totals beneath. Completion for the month reads well; the days column is where the actual story is.
| Asset | Due | Done | Days | Result |
|---|---|---|---|---|
| AHU-03 | 01 Dec | 03 Dec | +2 | On time |
| RTU-01 | 05 Dec | 04 Dec | -1 | On time |
| CH-01 | 08 Dec | 26 Dec | +18 | Late |
| PUMP-02 | 12 Dec | n/a | blank | Missed |
| GEN-01 | 15 Dec | 17 Dec | +2 | On time |
| Month | 42 due | 39 done | avg +4.2 | 93% done, 79% on time |
Ninety five percent completion and seventy nine percent on time are the same month described two ways, and only one of them is useful. With a seven-day tolerance agreed in advance, the two-day slips read as on time and nobody needs to argue about them, which is exactly what agreeing it beforehand buys you. The chiller at plus eighteen is the row that matters: it was done, so it counts toward completion, and it was eighteen days past an interval somebody chose deliberately, so for those eighteen days the asset was running outside the protection the program promised. The pump row shows why the days column is left blank rather than filled with a zero when a PM has not happened. A zero would read as on time and quietly inflate both percentages; blank keeps it visible and pushes it into the missed register, where it needs a name against it and a rescheduled date before it disappears from view.
Excel is the one to use, since the days early or late, both compliance percentages and the monthly summary all calculate from the dates you enter. Word and PDF for the version that goes into an audit pack or a client report. Free, and yours to rebrand.
How do you keep a preventive maintenance log?
Settle the rules first, then log two dates per visit and let the arithmetic do the rest. The review at the end is what makes it worth keeping. Six steps.
Agree the tolerance and the missed-PM rule first
How many days either side of the due date still counts as on time, and what happens when a PM is missed: carried forward, canceled with the next occurrence standing, or escalated. Settling both before logging starts is what stops the compliance figure becoming a matter of opinion.
Name where PMs are raised from
A schedule, a CMMS or a contract program. The log measures adherence against whatever that source says was due, so if the source is unclear the adherence figure is measuring nothing in particular.
Log two dates for every visit
The date it was due and the date it was actually done, both real. The days column works out early or late on its own. Leave done blank until it happens, so days stays empty rather than showing a wrong number.
Record the result, and use all four states
On time, late, missed or skipped, spelled exactly that way so the summary counts them. Skipped with authorization and missed entirely are genuinely different events and collapsing them hides the one worth acting on.
Push every missed or skipped PM into the register
With why it did not happen, who authorized it, when it is rescheduled to and whether the risk was accepted. A missed PM with nobody named against it will be missed again in the same way.
Review what the log is telling you, and change intervals
Which assets are repeatedly late, which months compliance drops in, which PMs never find anything and which always do. The first pair is a resourcing pattern; the second is an interval that is wrong in one direction or the other.
A log versus the schedule it measures
These two documents look similar and point in opposite directions in time. They share no fields at all, which is unusually clean and worth knowing before you pick one.
| Aspect | This log | The preventive maintenance schedule |
|---|---|---|
| Direction | Backward: what happened | Forward: what is coming |
| A row is | One visit that took place | One task placed in the months it falls due |
| The key column | Days early or late | Hours in each month |
| The question it answers | Was the interval actually honored | Can the year be delivered at all |
| What it proves | The program is real, with evidence | Nothing yet, it has not happened |
| Used by | Auditors, clients, whoever changes intervals | The planner building the year |
Neither document can do the other's job, and the fields do not overlap anywhere. The schedule holds estimated hours against months, which is capacity planning; the log holds actual dates against due dates, which is adherence measurement. That means a team can have an immaculate schedule and no idea whether it was followed, or a detailed log and no ability to say whether next year is achievable. Both situations are common, and each is invisible from inside the other document. Worth noting the log also sits apart from a preventive maintenance work order, which records one visit in full: what was found, what was done, readings taken, parts used. The work order is the depth on a single job; the log is the pattern across hundreds of them. If you want to know whether a specific PM was carried out properly, read the work order. If you want to know whether your intervals are being honored, only the log will tell you.
When the template starts to feel limiting
As a record of what happened, a spreadsheet log works. What it cannot easily do is any of the four things you keep it for.
The log and the schedule are separate files
Due dates are typed in from somewhere else, which means they can be typed in wrong or typed in after the fact, and nothing reconciles the two.
Patterns need somebody to look for them
Repeatedly late assets and months where compliance dips are both visible in the data and invisible on the page until a person sorts, filters and notices.
It is only current when somebody updates it
A log is a task in itself, and it is the task that slips first in exactly the busy months whose compliance you most wanted to measure.
Interval changes do not flow anywhere
The review concludes an interval should move from quarterly to semi-annual. That conclusion then has to be carried by hand into the schedule, the task library and the CMMS.
What running this in Facilio looks like
The template is the paper version of this record. The columns are the same; the difference is that adherence is computed from work the system already holds rather than from dates somebody retyped.
Audit Report Intelligence
Adherence is a read, not a compilation
Completion and on-time compliance come from the jobs themselves, so both figures exist continuously rather than being assembled at month end from a sheet somebody had time to update.
Ops Performance Intelligence
Lateness surfaces while it is still fixable
A PM drifting past its due date is visible in the week it happens rather than in a retrospective review, which is the difference between preventing a late visit and recording one.
Work Completion Validator
A logged PM is a PM that was actually done
Completion is held against evidence, so the numbers the log reports cannot be inflated by jobs closed without the work behind them.
Contractor Work Tracker
Contracted adherence is measured the same way
Where PMs are delivered under contract, on-time compliance sits against the provider, so the conversation at review is about the same figure both parties can see.
Hallucination-free by design. Atom AI answers from the records in your tenant rather than generating plausible text, so an empty field reads as empty rather than filled in for you.
Frequently asked questions
What is a preventive maintenance log?
A preventive maintenance log is the backward-looking record of PM visits that have actually happened. Each row pairs the date a task was due with the date it was done, works out how many days early or late it ran, and records the asset, the task, the technician, hours, the result, deficiencies found and any corrective work order raised.
It is deliberately not forward-looking. What is coming lives in a preventive maintenance schedule; this document exists to prove what was done and to measure how closely the intervals were actually held.
What should a preventive maintenance log include?
The settings first: where PMs are raised from, the tolerance allowed on a due date and what happens to a missed PM. Then one row per visit carrying due date, done date, days early or late, asset ID, task, frequency, technician, hours, result, deficiencies and the corrective work order number.
Then the reporting: a results summary, a compliance block giving completion and on-time percentages separately, a register of missed and skipped PMs with an authorizing name, and a review section covering repeatedly late assets and intervals worth changing.
What is the difference between a PM log and a PM schedule?
Direction. The schedule is forward-looking and holds estimated hours against the months tasks fall due, so it answers whether the year can be delivered. The log is backward-looking and holds actual dates against due dates, so it answers whether the intervals were honored.
They share no fields, which is unusually clean. You can have an excellent schedule and no idea whether it was followed, or a detailed log and no way to say whether next year is achievable. Most teams need both.
How do you calculate PM compliance?
Two ways, and the template reports both because they answer different questions. Completion is the share of due PMs that were done at all. On-time compliance is the share of those due that were done inside the agreed tolerance, measured against the same denominator so the two figures are comparable.
Reporting only completion is the common mistake. A program at ninety eight percent completion and sixty percent on-time is not slightly behind, it is a program where four in ten intervals were not actually honored, and the interval was the reason anybody believed the asset was protected.
What tolerance should we set on a due date?
There is no universal figure, and the template deliberately leaves it to you. Something like plus or minus seven days suits many building maintenance operations; process plant with tighter intervals often sets less.
What matters far more than the number is that it is agreed before logging starts and written down. Without it, every borderline visit becomes an argument and the compliance figure ends up meaning whatever the last person to touch the sheet decided it meant.
What do you do about a missed PM?
Decide the rule in advance: carried forward, canceled with the next occurrence standing, or escalated. All three are defensible; having no rule and deciding case by case under pressure is not.
Then push it into the missed and skipped register with why it did not happen, who authorized it, when it is rescheduled to and whether the risk was accepted. The authorizing name is the field that changes behavior, because a missed PM with nobody against it will be missed again in exactly the same way.
Can a maintenance log tell you your intervals are wrong?
Yes, and that is the part most teams never use. A PM that has found nothing for two years is either running too frequently or is not really being done, and both are worth knowing. A PM that finds something every single visit is a candidate for a shorter interval or for a redesign of the asset.
Neither shows up in a compliance percentage, which is why the template has a separate review section for it. Proving the work happened is the minimum a log does; changing intervals on the evidence is what makes it worth keeping.
Can I edit and rebrand this template?
Yes. It is free to use, edit, rename and put your own logo on, internally or for clients. No attribution required.
The Excel version is the one that earns its keep, since days early or late, both compliance percentages and the labor hours total all calculate from the dates you enter, and the Monthly Summary tab shows where in the year compliance slips. Word and PDF are the versions for an audit pack or a client report.
In one paragraph
A preventive maintenance log records what actually happened and how closely it matched what was due, which is the one thing no forward-looking document can tell you. Settle two rules before you log anything: the tolerance either side of a due date that still counts as on time, and what happens to a PM that is missed. Then record two real dates per visit rather than one, because the pair is what produces the days early or late column that nothing else in the set computes. Use all four result states, since a PM skipped with authorization and one missed entirely are different events. Report completion and on-time compliance separately and resist the temptation to lead with the friendlier number, because only on-time says the asset was protected at the interval somebody chose. Give every missed PM an authorizing name and a rescheduled date. Then use the review to move intervals, since a PM that never finds anything and one that always does are both telling you the interval is wrong.
The template is the floor, not the ceiling
Take the log; it will prove the program is real and give you an on-time figure worth reporting. When keeping it current becomes the task that slips in exactly the months you most wanted to measure, a connected CMMS takes it over: adherence computed from the work itself, lateness visible in the week it happens rather than at the next review, repeatedly late assets surfacing on their own, and an interval change flowing back into the schedule instead of being carried by hand.
