The burnisher quits at 9:40 on a Friday night inside a 60,000 sq ft office building. Your crew leader leans it against a wall, goes home, and on Monday the property manager walks a main corridor of dull, scuffed VCT and asks why she is paying for a monthly burnish she did not get.
That call is rarely about the machine. It is about the fact that nobody could say when the pads were last changed, whether the motor had been serviced, or which machine was supposed to cover the gap. The log is the difference between a $180 part and a $4,000 account.
Track cleaning equipment maintenance by run hours, not calendar dates. Give every machine a permanent asset ID, record its hour meter at every service event, and schedule work at the hour intervals printed in the machine's manual. One entry per machine per event, dated and signed by the person who actually did the work.
What follows is not a lecture on why maintenance matters. It is a field evaluation: nine tests you can run on the logging system you already have, ordered so the cheap tests eliminate broken systems before you waste a morning on the expensive ones.
What belongs in a cleaning equipment maintenance log entry?
Most logs fail because they record breakdowns instead of service. A breakdown log tells you what already cost you money. A service log tells you what is about to.
Seven fields per entry is enough. Anything less and you cannot reconstruct history; anything more and your crew stops filling it in.
| Field | Example entry | Why it has to be there |
|---|---|---|
| Asset ID | AS-04 (20 in. walk-behind scrubber) | "The red scrubber" is not an identity. Two machines of the same model have different histories. |
| Date and shift | 03/11, 2nd shift | Ties the entry to a schedule and a payroll record you can cross-check. |
| Hour meter reading | 412.6 | The only honest measure of wear. Everything else is a guess. |
| Service performed | Squeegee blades flipped, brush inspected, tank flushed | Distinguishes routine PM from a repair, which is how you spot a machine that is failing early. |
| Parts and consumables | 2 x squeegee blade, PN 56413181 | Part numbers turn a log into a reorder list and prove consumables are actually being changed. |
| Performed by | M. Reyes | Accountability, and it tells you who to ask when the note is unclear. |
| Status and next due | In service. Next PM at 450 hrs | A log with no forward-looking due date is a diary, not a maintenance system. |
Should maintenance intervals be based on run hours or calendar dates?
Run hours, with a calendar backstop. A scrubber that runs 45 minutes a night in a small office and one that runs four hours a night in a distribution center are not the same machine after a year, even if you bought them on the same day.
Here is the arithmetic on a portfolio. Assume five nights a week, roughly 21.7 working nights a month, and the nightly run times below. These are illustrative numbers, not benchmarks: measure your own by reading the meters two weeks apart.
| Machine | Assumed nightly run time | Run hours per month | Months to reach 100 run hours |
|---|---|---|---|
| AS-04, 20 in. walk-behind scrubber, 60,000 sq ft office | 45 min | 16.3 | 6.1 |
| BU-02, 20 in. propane burnisher, monthly corridor burnish | 3 hrs, once a month | 3.0 | 33.3 |
| SW-01, ride-on sweeper, 180,000 sq ft warehouse | 2 hrs 30 min | 54.3 | 1.8 |
| VA-11, upright vacuum, nightly office route | 2 hrs | 43.4 | 2.3 |
Read that last column again. The sweeper hits every service interval roughly eighteen times faster than the burnisher. A quarterly calendar PM schedule over-services one and starves the other.
The calendar backstop matters for anything that degrades while sitting still: flooded lead-acid batteries, rubber squeegee blades, and hoses. Those get a date-based check even at zero run hours.
9 tests to run on your equipment maintenance log this week
Run them in this order. Tests 1 through 4 take about twenty minutes at your desk and on one site walk, and they kill most broken systems outright. Do not bother with tests 7 through 9 until the first four pass.
Test 1: The 60-second retrieval test
Pick one machine at random. Ask for its complete service history, start to finish, right now.
Pass: someone produces every entry for that asset in under a minute, from a binder, a spreadsheet, or a screen. Fail: anyone has to call a technician, check a truck, or say "I know we did it." A log you cannot retrieve under pressure does not exist during a warranty dispute or a client audit, which are the only two moments it has to work.
Test 2: The asset ID match test
Walk to three machines. Read the ID off the machine and find that exact ID in the log.
Pass: three for three, and the model and serial number on the log match the data plate. Fail: the log names machines by color, size, or nickname. This is the single most common failure, and it silently merges the histories of two machines, which means both PM schedules are wrong.
Test 3: The hour-meter reconciliation test
On those same three machines, read the current hour meter and compare it to the last logged reading and the date of that entry.
Pass: the hours accumulated since the last entry are consistent with how often that machine actually runs. Fail: the meter is lower than the logged figure, identical after six weeks of nightly use, or a suspiciously round number. Round numbers and frozen readings mean the log is being filled in at a desk from memory, which makes every interval in it fiction.
Test 4: The orphan asset test
Count the powered machines physically on one site. Count the machines in the log for that site. The numbers should match.
Pass: every machine on the floor has a log, including the backup vacuum in the closet and the wet/dry vac in the janitor's cart. Fail: even one orphan. Orphans are where surprise failures live, because nobody owns them and nobody is watching their hours.
Test 5: The consumables trail test
Look for the wear parts, not the repairs: pads and brushes, squeegee blades, vacuum belts, filters, gaskets, and battery water.
Pass: the log shows a last-changed date and a part number for each wear item on each machine. Fail: the log contains only breakdowns and invoices. A log that records only failures cannot prevent them, and it hides the real cost driver, which is a crew running a worn brush that pulls extra amps and cooks a motor.
Test 6: The safety inspection test
OSHA requires that portable cord- and plug-connected equipment and flexible cord sets be visually inspected before use on any shift for external defects and evidence of possible internal damage. That covers most of your vacuums, extractors, and buffers.
Pass: your log or shift checklist has a line for cord, plug, and strain relief condition, and machines with damaged cords are tagged out rather than taped up. Fail: no line item anywhere. If you water flooded batteries on site, confirm you also meet the requirement for suitable quick-drenching or flushing facilities where the eyes or body may be exposed to corrosive materials.
Test 7: The cost-per-asset test
Ask for trailing 90-day repair and parts spend for one machine. Not total shop spend. One asset.
Pass: you get a number in a few minutes because invoices are coded to asset IDs. Fail: repair invoices are filed by vendor and date only. Without cost per asset you cannot make a repair-versus-replace decision, so you will keep pouring money into the one machine that eats your maintenance budget.
Test 8: The tag-out and substitution test
Ask what happened the last time a machine went down mid-shift, and trace it in writing.
Pass: there is an out-of-service tag on the machine, a dated work order with a status, a named substitute machine, and a note to the account about any task that slipped. Fail: a verbal handoff. Verbal handoffs are how a down burnisher becomes a client complaint two weeks later, because the schedule never learned the machine was gone.
Test 9: The successor test
Hand the log to someone who has never touched these machines. Ask them to tell you which three machines are due for service in the next 30 days and what parts to order.
Pass: they answer from the records alone. Fail: they need your best tech to interpret it. A log that only one person can read is that person's memory with extra steps, and it walks out the door when they do.
Print this: equipment log site walk, one page
- Pick one machine at random and ask for its full history. Time it. Under 60 seconds?
- Read the asset ID off the machine. Does that exact ID exist in the log?
- Does the log's model and serial number match the data plate?
- Read the hour meter. Write it down. Compare to the last logged reading and date.
- Count powered machines on site. Count machines in the log. Do they match?
- Check for the backup vacuum, the wet/dry vac, and anything in a locked closet.
- Find the last recorded change date for: pad or brush, squeegee blades, vacuum belt, filters.
- Inspect cords, plugs, and strain relief on three corded machines. Any damage untagged?
- If flooded batteries: is there a watering log, PPE on site, and drenching or flushing access?
- Ask for 90-day parts and repair spend on one named asset.
- Find the last out-of-service tag and trace it to a written work order and a substitute machine.
- Ask a non-technician to name the next three machines due. Can they?
- Write today's hour meter readings on the sheet before you leave. That is your next reconciliation baseline.
When does the log tell you to replace instead of repair?
This is the payoff for test 7. Once repair spend is coded to asset IDs, the decision stops being emotional.
A common operational rule among equipment-heavy operators: when trailing 12-month repair and parts spend on a machine exceeds roughly half the cost of replacing it, stop repairing. A second trigger is downtime frequency. Two unplanned failures in one quarter on a machine that carries a nightly route is a scheduling risk, regardless of what the parts cost.
Worked example, illustrative only. Suppose AS-04 has taken $1,150 in parts and labor over twelve months and a comparable new 20 in. scrubber quotes at $4,200. That is 27 percent, so you keep it. Add a $900 traction motor and you are at $2,050, or 49 percent, and you are now buying a used machine at nearly half of new money. Run your own numbers with your own quotes.
The reason this matters beyond the repair bill: your production rates assume the machine works. ISSA publishes cleaning times that the industry uses to build those rates, and every one of them presumes equipment operating at spec. A scrubber running on a worn brush at half its rated coverage quietly converts your priced hours into unpriced overtime.
Questions operators actually ask about equipment logs
How long should I keep cleaning equipment maintenance records?
Keep them for the life of the asset plus any window in which you might file a warranty claim or defend an insurance or injury claim. OSHA does not set a general retention period for routine equipment service logs, but your insurer, your manufacturer, and some client contracts effectively do. Three years past disposal is a practical floor, and digital records make it free to keep them longer.
Do vacuums need a maintenance log, or just floor machines?
Vacuums need one, and they are usually the biggest gap. They accumulate more run hours per month than most floor machines on a nightly office route, they are the equipment your crews handle most, and their cords are what the OSHA pre-shift visual inspection requirement is squarely about. Log belt changes, filter changes, and cord condition at minimum.
Should the cleaner or the supervisor fill out the log?
Split it. The cleaner records what happened on the shift: hour meter, pre-use inspection, anything abnormal. The supervisor or technician records the service event itself, with parts and the next due reading. If you make cleaners responsible for the technical entries, the log gets vaguer every week; if you make supervisors responsible for daily readings, they get invented.
What do I do if a machine has no hour meter?
Estimate run hours from the schedule instead of guessing at wear. Multiply the assigned nightly run time by the number of scheduled nights, log the estimate monthly, and treat the total as your interval trigger. Then add an inexpensive aftermarket vibration-activated hour meter, or make a metered model a requirement on your next purchase order.
Will a client ever ask to see equipment maintenance logs?
Yes, and it tends to happen at the worst time: during a healthcare or school RFP, after a slip-and-fall incident, or when a facilities director is building a case to rebid. Being able to produce a clean history for the machine involved, within a day, changes the tone of that conversation entirely.
Where CleanTrack360 fits
CleanTrack360 does not ship a dedicated asset registry, so be clear about what it does and does not do here. What it gives you is the surrounding structure: recurring work orders for scheduled PM so a 50-hour scrubber check lands on someone's shift instead of someone's memory, quality inspections with custom checklists, photo evidence and automatic scoring that you can point at a machine's cord, brush and squeegee condition, supply requests for the parts that inspection turns up, and location-based team messaging so a mid-shift breakdown reaches a supervisor rather than a wall.
Because inspection results and work orders export to CSV, you can keep hour meter readings and photo-documented checks in one place and run your cost-per-asset math outside the platform. Plans are Starter at $99 per month for up to 5 team members, Pro at $199 for up to 20, and Business at $249 for up to 50, priced per plan rather than per user, with a 14-day free trial and no credit card. Crews use it in the phone browser today; the native app is still in development.