Preventive Maintenance in Hotels and Hospitality

Most preventive maintenance guidance is written for factories and office towers. Neither model transfers cleanly to a hotel. In a factory, the asset that matters is the production line and the building is a shed around it. In an office tower, the tenant leaves at seven in the evening and the plant room is yours until morning. A hotel has neither luxury. The asset that matters most is the guest room itself, there are between three hundred and a thousand of them, each one is a small revenue-generating unit with its own HVAC, plumbing, electrical, joinery and soft furnishings, and the building never empties. Building a hospitality maintenance programme means accepting those constraints up front and designing around them rather than pretending they are not there.

The message up front: in hospitality, preventive maintenance is not scheduled against the calendar, it is scheduled against the occupancy forecast. The engineering department that wins is the one that has negotiated a standing room allocation with revenue management and front office, uses it relentlessly, and treats the guest room PM cycle as a production line rather than a series of one-off interventions. Everything else in this guide is detail on top of that one principle.

1. What actually makes hospitality PM different

Before the checklists, it is worth being precise about the four constraints that make a hotel unlike any other building you will maintain. Every design decision in a hospitality maintenance programme traces back to one of them.

• The room is the product, and it is perishable. An unsold room-night is gone forever. Taking a room out of service for a four-hour PM has a direct, calculable revenue cost on a high-occupancy night and almost none on a low-occupancy Tuesday in the shoulder season. No other building type prices its own maintenance access this explicitly, and it is the single biggest reason hotel PM slips.
• Failures are reputational, not just operational. In a warehouse, a noisy fan-coil unit is a work order. In a hotel it is a one-star review that sits on a booking platform for years and influences pricing power long after the bearing has been replaced. The cost of a guest-facing failure is not the repair cost, it is the distribution of the complaint.
• There is no shutdown window. Twenty-four hours a day, seven days a week, with arrivals at three in the morning and breakfast service starting at six. Noisy, dusty or disruptive work has an extremely narrow permitted envelope, and in many properties the only genuinely quiet hours in guest areas are roughly ten in the morning to three in the afternoon, which is also exactly when housekeeping wants the floor.
• Brand standards are audited. If the property flies an international flag, the operator's brand standards manual specifies refurbishment cycles, condition thresholds and in many cases the PM frequencies themselves, and an inspector will arrive unannounced to check. This turns parts of the PM programme from a good idea into a contractual obligation with a franchise agreement behind it.

Put those four together and you get the defining characteristic of hospitality maintenance: the engineering department does not control its own access to the majority of its asset base. Housekeeping, front office and revenue management do. An engineering manager who has not built working relationships in those three departments will have a technically excellent PM programme with a thirty percent completion rate.

2. The guest room PM cycle: treat it as a production line

The pattern that works, and the one I would recommend to any property above about a hundred and fifty keys, is to stop thinking of guest room PM as a series of individual work orders and start thinking of it as a continuous cycle with a fixed throughput rate.

The arithmetic is simple and it is the most useful calculation in hotel engineering. Take your room count, divide by the number of times per year you intend to touch each room, and divide again by the number of working days you actually have. A four hundred key property on a quarterly cycle needs sixteen hundred room PMs a year. Across roughly two hundred and forty working days, that is between six and seven rooms every single working day, without exception, forever. If your standing allocation from front office is four rooms a day, the cycle is mathematically impossible and no amount of effort will save it. You either negotiate the allocation up, reduce the cycle frequency, or accept that the programme will not complete. Most properties I have looked at have never done this calculation, which is why their PM completion rate is a mystery to them.

Once you have the number, the operating model follows:

• A standing block, not an ad hoc request. A fixed number of rooms are blocked in the property management system every day as engineering rooms, released back the same evening. This is agreed once with front office and revenue management, written down, and reviewed quarterly against occupancy, rather than begged for each morning.
• Sequential by floor and stack. Work one floor or one vertical stack at a time. It reduces the technician's travel, it concentrates the noise on one part of the building, and it lets housekeeping plan around a known area rather than a scatter of rooms across nine floors.
• A fixed-duration, fixed-content visit. The room PM is a defined job with a defined time budget, typically ninety minutes to two and a half hours depending on room type and scope. Anything found that exceeds the visit is raised as a separate corrective work order, not absorbed into the PM. If you let the PM absorb corrective work, the cycle time becomes unpredictable and the allocation stops being enough.
• One technician, multi-skilled. The guest room PM is deliberately designed to be completed by a single competent multi-skilled technician covering basic mechanical, electrical, plumbing and joinery tasks. Splitting it across trades triples the coordination cost and the number of room entries.

3. The guest room PM checklist

This is the core deliverable. Below is a hotel preventive maintenance checklist for a standard guest room, structured the way I would build it in a CMMS: grouped by trade so the technician works logically through the room, with the pass criteria stated rather than left to judgement. Adapt the scope to your room type, but keep the structure.

Two notes on using this. First, resist the temptation to make the checklist longer. A checklist a technician cannot complete inside the allocated visit is a checklist that gets partially ticked, and partial ticking corrupts every completion metric you have. Second, build it in the CMMS as a structured task list with pass or fail responses and required readings, not as a free-text instruction block, so that the results are queryable. The general principles for this are in my PM checklists, templates and examples pillar, and the wider strategy framing sits in the complete guide to preventive maintenance.

4. FF&E condition and the soft refurbishment cycle

The room PM keeps the room working. It does not keep the room sellable, and those are different problems. Furniture, fixtures and equipment degrade on a slower cycle that no monthly inspection will catch, and hospitality is unusual in that the commercial refurbishment cycle is largely fixed by the brand and the market rather than by the engineering condition assessment.

The typical hierarchy in a branded full-service property looks roughly like this, though the intervals vary by brand, market and asset positioning:

• Touch-up and repair, continuous. Chips, scuffs, minor upholstery damage, carpet spot repairs. Handled in the room PM cycle or by a dedicated small-works team, funded from operating expenditure.
• Soft refurbishment, roughly every six to eight years. Soft furnishings, carpets, curtains, bed linen and mattresses, upholstery, decoration, sometimes bathroom re-grouting and fittings. The rooms go out of service in blocks, often a floor at a time.
• Full renovation, roughly every twelve to fifteen years. Case goods, bathrooms, wiring, layout and technology infrastructure. This is a capital project, not a maintenance activity, and the engineering team's role is largely as technical client and condition-data provider.
• Mattress and case goods replacement on their own cycle. Mattresses are frequently on a shorter and more tightly audited cycle than the soft refurbishment, precisely because they carry so much of the guest satisfaction score.

What the engineering department should be doing between those events is running a condition score on each room, updated at every PM visit. A simple one to five score per element, recorded against the room asset in the CMMS, becomes the evidence base for the refurbishment business case and for arguing that a particular floor or room type needs to be brought forward. Without it, refurbishment scheduling is decided by whoever argues most confidently in the capital meeting. With three years of condition trend data, the engineering director walks into that meeting with an argument that survives scrutiny.

5. Scheduling against occupancy, and the low season

The scheduling principle in hospitality is that PM demand is roughly constant across the year while PM access varies enormously. The engineering department that schedules a flat number of room PMs every month will fail in peak season and leave capacity unused in low season. The programme has to breathe with the occupancy curve.

What I would recommend in practice:

• Build the annual plan from the forecast, not from January. Take the twelve-month occupancy forecast from revenue management, identify the three or four lowest-occupancy months, and load the deep, disruptive and room-hungry work there. In the Gulf that usually means the deep summer. In European resort markets it is the closed or near-closed shoulder.
• Protect a minimum daily allocation even at peak. The temptation at ninety-five percent occupancy is to suspend room PM entirely. That is how a property ends up with a two-year backlog. A reduced but non-zero allocation, even two rooms a day, keeps the cycle alive and keeps the habit alive, which matters more.
• Use the out-of-order rooms you already have. Every property carries a few rooms out of order for damage or long-lead repairs. Batch the PM into that period so the room is returned to sale fully serviced rather than just repaired.
• Sequence the plant work against the load curve, not the occupancy curve. Chiller major maintenance belongs in the cool months regardless of occupancy, because that is when you have thermal redundancy. Boiler and heating plant work belongs in the warm months. These two curves often point in opposite directions and the plant one usually wins.
• Treat the low season as a project, with a programme. The annual shutdown window is not a longer version of routine maintenance, it is a compressed project with a critical path, contractor mobilisation, materials on site in advance and a room-release schedule. Plan it like one, three to four months ahead.

The scheduling mechanics themselves, the levelling, the frequency choices and the way you stop everything falling due in the same week, are covered in more depth in how to build a preventive maintenance schedule.

6. Out of order versus out of service, and why the distinction matters

This is small, it is unglamorous, and getting it wrong costs a property real money every month. Hotels distinguish between two states for an unsellable room, and the distinction is commercial rather than technical.

• Out of order means the room is removed from the sellable inventory entirely. It does not count in the available room count, which means it does not drag down the occupancy percentage or the revenue per available room. Used for genuine long-duration unavailability: flood damage, refurbishment, a failed bathroom, anything measured in days or weeks.
• Out of service means the room is temporarily unsellable but stays in the available inventory, so it does count against occupancy and revenue per available room. Used for short-duration blocks: a deep clean, a same-day repair, a room PM visit.

The consequence for engineering is direct. A routine room PM is an out of service block, and every one of them very slightly depresses the property's headline performance metrics. That is precisely why front office and revenue management resist the allocation, and why the argument for it has to be made in commercial language rather than technical language. The counter-argument that works is not "we need to do maintenance", it is that unplanned guest-facing failures generate room moves, rate concessions and compensation that cost more than the planned out-of-service time, and a maintained room sustains its rate. Whether you can actually evidence that depends on whether you have been logging guest-reported faults with enough discipline to show the trend.

The related discipline is to be honest about which state you are using. Engineering teams under pressure on completion rates sometimes book a PM room as out of order because it hides better in the reporting. It also quietly inflates the occupancy percentage, and when the finance team eventually finds it, the engineering department loses the trust it needs to negotiate the allocation at all. Use the correct code, take the visibility, and make the commercial argument in the open.

7. Back of house: plant, kitchen, laundry and pool

Guest rooms take most of the attention, but the back-of-house plant is where the catastrophic failures live. A failed guest room thermostat is a complaint. A failed chiller in August, in a Gulf property, is an evacuation-level commercial event. The plant PM programme should be built on criticality, and the hospitality criticality ranking is genuinely different from an industrial one because it weights guest impact and reputational exposure alongside safety and cost.

The table below is a working frequency baseline for a full-service hotel. It is a starting point to be adjusted against manufacturer requirements, local statutory regimes and the actual duty of your equipment, not a substitute for either.

For the frequencies themselves, rather than inventing intervals, I would anchor the programme to a recognised standard where one applies. SFG20 is the common reference for building services maintenance schedules in the UK and increasingly used in the Gulf, and ASHRAE publishes the authoritative guidance on HVAC and water system practice. For life safety, the NFPA codes and your local civil defence regime govern, and those frequencies are not yours to adjust.

Ranking the plant by consequence rather than by cost is the other half of this. The method I would use is set out in the asset criticality classification pillar, with the hospitality adjustment that guest-facing impact and brand exposure need to be explicit scoring dimensions rather than folded into a generic business-impact score.

8. Fire, life safety and the statutory floor

Everything in this guide is negotiable against occupancy except this section. A hotel is a sleeping-risk building full of people who do not know the layout, many of whom are unfamiliar with the local emergency procedures and some of whom do not speak the language of the announcements. The statutory maintenance regime exists because of exactly that, and it is the one part of the programme where deferral is not a commercial trade-off, it is a regulatory breach and in many jurisdictions a personal liability for the responsible person.

Practically, that means three things for the engineering department:

• Ring-fence the statutory PMs in the CMMS. Use a distinct work order type or priority so statutory work can never be silently absorbed into a general backlog and is reportable on its own. The work order types pillar covers how to structure this cleanly.
• Keep the evidence, not just the completion flag. Test certificates, readings, contractor reports and photographs attached to the work order. An inspector or an insurer will ask for the record, and a green tick in a system with no supporting document is worth very little at that moment.
• Plan the disruption openly. Alarm testing, sounder testing, sprinkler pump runs and full-duration emergency lighting tests all disturb guests. Fix the day and time, tell front office in advance so they can pre-warn guests and brief the night team, and never quietly move a statutory test because the hotel is busy.

9. The engineering department's operating rhythm

A hotel engineering department that runs well has a visible rhythm, and the same one appears in most good properties I have seen. It is worth writing down explicitly, because departments that drift into pure reactive work almost always lost the rhythm first and the completion rate second.

• Daily. Morning handover with the night shift. Review of the overnight guest-reported faults and the room moves they caused. Plant round with readings logged. Confirmation of today's room allocation with front office. Allocation of the day's PMs and correctives. A short attendance at the morning operations meeting, where engineering is in the room rather than reported about.
• Weekly. Statutory weekly tests completed and recorded. PM completion rate reviewed against the cycle target with the reason for any shortfall named. A walk of one guest floor and one back-of-house area by the chief engineer. Contractor visit schedule for the coming week confirmed. Stores and critical spares check.
• Monthly. Full KPI review: PM completion, reactive versus planned ratio, guest-reported fault volume by category and by floor, repeat faults, energy consumption against the same month last year, open work order ageing. Condition score summary for rooms touched that month. Energy and water performance review with the sustainability or finance lead.
• Quarterly. Allocation renegotiation with revenue management against the updated forecast. Contractor performance review. PM content review: which tasks are consistently finding nothing and should have their interval extended, which assets are failing between PMs and need a shorter interval. Brand standard self-audit.
• Annually. The low-season programme, planned as a project. Capital and refurbishment submissions supported by condition data. Statutory regime review against any regulatory change. A full PM programme review rather than a rollover of last year's schedule.

The quarterly PM content review is the one most commonly skipped and the one that most protects the programme. A hotel PM schedule that is never pruned grows steadily until the team cannot complete it, at which point the technicians start silently deciding which tasks matter. Better that you decide, deliberately, once a quarter, with the finding data in front of you. The metrics that make this review possible are in the FM KPI framework pillar.

10. Coordinating with housekeeping and front office

This is where hospitality PM is actually won or lost, and it gets almost no coverage in general maintenance literature because no other sector has quite this dependency. The engineering department needs the cooperation of two other departments with different incentives, different reporting lines and, frequently, a different view of whose building it is.

What works, in my experience of watching properties that do this well:

• Housekeeping is your largest inspection workforce, so equip them. A room attendant enters every room every day. They see the dripping tap, the flickering lamp, the lifting silicone and the noisy fan coil long before any PM cycle reaches that room. Give them a genuinely frictionless way to raise a fault, a mobile CMMS or a simple app, and close the loop by telling them what happened. If reported faults disappear into silence, reporting stops within a month and you lose the most valuable early-warning system you have.
• Agree the room release protocol in writing. Who blocks the room, at what time, who releases it, who triggers the make-up clean, and what the cut-off is for returning a room to sale. Ambiguity here produces the worst possible outcome, which is a guest checking into a room with a technician's toolbox still in it.
• Give front office a same-day escalation path. A guest complaint about a room fault needs a response measured in minutes, not in the next PM cycle. Define the response target, staff to it, and make the room-move decision quickly rather than trying to repair under pressure with a guest standing in the doorway. Response and resolution targets belong in a documented service level structure, which is the subject of the SLA matrix design pillar.
• Publish the schedule where the other departments can see it. The coming week's room allocation, the noisy works, the plant isolations and the statutory test times, visible to housekeeping and front office without anyone having to ask. Most of the friction between these departments is surprise, not disagreement.
• Attend the morning meeting, every day, personally. The chief engineer who is in that room is part of the hotel's operating team. The one who sends a report is a service provider to it, and service providers do not get the room allocation they need.

11. What the CMMS needs to do for a hotel

Hotels have a handful of requirements that general maintenance software handles poorly if you do not check for them during selection, and I would put these near the top of any hospitality evaluation.

• The room as an asset with children. Room 412 needs to be an asset in its own right, with the fan coil, the door lock, the television and the bathroom fittings beneath it, so that you can see the full history of that room rather than a scatter of unlinked work orders. A flat asset list makes room-level analysis impossible.
• Bulk PM generation and cycle scheduling. Generating and rescheduling hundreds of near-identical room PMs against a moving allocation, without someone doing it by hand each week.
• Mobile fault reporting for non-technical staff. Room attendants and front office agents need to raise a fault in seconds, with a photograph, from a phone, in their own language where the workforce is multilingual.
• Property management system awareness. At minimum, visibility of room status. At best, an integration so an engineering block and a room status change do not have to be keyed twice. In practice this integration is rarer and harder than vendors imply, and it is worth testing the claim carefully before it becomes a business case assumption.
• Multi-property reporting. For groups, the ability to compare completion rates and fault trends across properties on a consistent asset taxonomy. Without a shared taxonomy imposed at the outset, the group-level reporting is worthless no matter how good the tool is.

Platform choice matters less than people assume. Large groups and owner-operated portfolios with heavy plant tend toward IBM Maximo, Infor EAM or Planon. Individual properties and mid-sized groups are usually better served by a lighter mobile-first system such as MaintainX, Limble, Fiix, eMaint or UpKeep, where the room attendant fault-reporting experience is genuinely good out of the box. What matters more than the badge is whether the room hierarchy, the cycle scheduling and the non-technical reporting path work the way the property actually operates. My neutral view on the market is in the CMMS buyer shortlist.

The idea to walk away with

A hotel preventive maintenance checklist is the easy part of this problem, and it is the part most properties spend all their effort on. The hard part is access. You can write the best guest room PM checklist in the region, build it beautifully in the CMMS, and still complete a third of it, because the rooms were never blocked, the allocation was never agreed, and the cycle was never sized against the number of working days in the year.

So the sequence I would follow is the reverse of the usual one. First do the arithmetic: room count, cycle frequency, working days, rooms per day. Then negotiate that allocation with front office and revenue management in commercial language, using guest-reported faults per occupied room as the evidence. Then, and only then, design the checklist to fit the visit duration the allocation supports. A checklist sized to the access you actually have will outperform an ideal checklist sized to the access you wish you had, every single year.

Final thoughts

Hospitality maintenance rewards a particular temperament: someone willing to run a repetitive cycle with discipline for years, to negotiate patiently with departments that do not report to them, and to accept that the measure of success is the absence of complaint rather than any visible achievement. It is not glamorous work and it is rarely celebrated when it goes well. But the properties that hold their rate through a refurbishment cycle, that pass brand audits without a scramble, and that do not have a chiller failure in the peak month, are almost always the ones with a quiet, boring, relentlessly completed PM cycle underneath them.

If you are inheriting a hospitality maintenance programme and want one place to start, start with the number. Work out how many room PMs a day your stated cycle actually requires, compare it honestly with how many rooms you are given, and put that single comparison in front of the general manager. In most properties that one slide changes the conversation faster than any checklist ever will. For how this pattern compares with maintenance in other sectors, the PM examples across industries pillar sets hospitality alongside manufacturing, healthcare and utilities, and the preventive maintenance strategies pillar covers the time, meter and condition choices underneath all of it.

Related reading: Preventive maintenance: the complete guide, PM checklists, templates and examples, How to build a PM schedule, PM examples across industries, Asset criticality classification, FM KPI framework.