Maintenance Backlog Management: How to Regain Control
- Trefnus
- Aug 7
- 14 min read
Published: 5 August 2026 | Last reviewed: 5 August 2026
Every maintenance team has a list of jobs it has not got round to yet. That list is your maintenance backlog, and its existence is not a problem in itself. Backlogs become a problem when nobody knows how big the list is, how old the oldest item is, or which jobs on it could stop the business trading.
Maintenance backlog management is the discipline of measuring, prioritising and steadily working through outstanding work so the backlog stays a planning tool rather than a growing liability. For small and medium businesses, where one or two people often cover an entire site, the risk is sharper. A backlog nobody is tracking tends to be discovered by a breakdown, an insurance inspection or a failed audit rather than by a manager.
What a maintenance backlog actually is
A maintenance backlog is all identified maintenance work that has been raised but not yet completed. It usually includes overdue preventive tasks, corrective jobs from defect reports, work identified during inspections, and anything postponed while waiting for parts, budget or access. It is often called a work order backlog, because in most systems outstanding maintenance tasks are tracked as open work orders.
Counting all of that as one number hides the most useful information. It is worth splitting the backlog into three categories.
Ready backlog: Everything needed to do the job is available: the parts are on the shelf, the labour exists, access is possible and the spend is approved. These jobs could be scheduled tomorrow.
Waiting backlog: The job is defined but blocked. A part is on order, a contractor is not booked, the asset cannot be taken out of service until a quiet period, or a quote is sitting with a director.
Deferred maintenance: A deliberate decision to postpone work, usually on cost grounds, with a consequence somebody has accepted in writing.
Why the distinction matters
These three categories point at completely different problems. A team buried in ready backlog has a resourcing problem and needs hours. A team buried in waiting backlog does not need another technician, it needs better parts planning and faster approvals. A team accumulating deferred maintenance has a budgeting conversation to have, not a scheduling one.
If all three sit in one undifferentiated pile, the obvious response is to hire or to work overtime, which may fix nothing at all.
Why maintenance backlog management matters for smaller businesses
Larger organisations usually have a planner whose job includes watching the backlog. In an SME, that responsibility tends to sit with someone who also runs the site, orders the stock and answers the phone. The consequences of losing track are the same, but they arrive faster.
Costs escalate quietly
Deferred work rarely stays the same size. A worn bearing left for six months becomes a motor replacement. A leaking roof detail becomes a ceiling. The cost difference between doing a job when it is identified and doing it after failure can be substantial, and the failure version usually arrives at the worst possible moment, with production stopped and somebody charging emergency rates.
Safety and legal exposure
Some backlog items carry legal weight. Under the Provision and Use of Work Equipment Regulations 1998, work equipment must be maintained in an efficient state, in efficient working order and in good repair. The same regulations require that where machinery has a maintenance log, that log is kept up to date. There is no blanket legal requirement to keep a maintenance log for every piece of equipment, but if you keep one, it has to be current.
Thorough examination under the Lifting Operations and Lifting Equipment Regulations 1998 is a separate duty from routine maintenance. Lifting equipment must be thoroughly examined by a competent person at the intervals set out in those regulations or under a written scheme of examination. Passing a thorough examination does not discharge the maintenance duty, and keeping up with maintenance does not remove the need for examination.
Neither set of regulations prescribes a maintenance frequency for most equipment, so the right interval depends on the equipment, how it is used, the environment it operates in and the manufacturer’s guidance. HSE takes a comparable line on electrical equipment, where the type and frequency of inspection and testing depends on the equipment, its environment and the results of previous checks.
The practical consequence is that an overdue safety-critical inspection or maintenance task can represent a significant safety and compliance risk rather than a purely administrative one. It should not be sitting in the same queue as a request to repaint a door.
Insurance and warranty expectations
Insurers, landlords, manufacturers and warranty providers may expect evidence that a maintenance regime is being followed. These are contractual expectations rather than statutory duties, and the specifics vary considerably between policies, leases and warranty terms, but a backlog with no records behind it makes them awkward to demonstrate. Check your own wording rather than assuming.
Team morale and retention
A permanently growing backlog is demoralising. Technicians who spend every day firefighting, knowing the list behind them is getting longer, tend to disengage or leave. In a two-person team, losing one person doubles everyone else’s problem overnight.
How to measure your maintenance backlog
You cannot manage a backlog you have not sized. Counting open jobs is the weakest possible measure, because one item might take twenty minutes and the next two days. Measure in estimated labour hours instead.
It is also worth separating backlog from overdue work. A backlog is not automatically an overdue backlog. Some of it may be correctly scheduled for a future date and entirely under control. The sharper question is how much work has passed its required or agreed completion date, and how much of that carries safety, statutory or operational consequences. That figure, rather than the headline total, is the one to lose sleep over.
Backlog in weeks
The standard approach expresses the backlog as the number of weeks it would take to clear at current capacity:
Backlog weeks = total estimated labour hours outstanding ÷ available labour hours per week
A worked example for a small facilities team:
Two technicians at 37.5 hours a week gives 75 hours of total capacity.
Roughly 40 per cent is absorbed by reactive callouts and scheduled preventive work, leaving about 45 hours a week available for backlog.
Outstanding work is estimated at 180 labour hours.
180 divided by 45 gives four weeks of backlog.
Many maintenance teams treat somewhere between two and four weeks of ready backlog as a comfortable working range: enough to keep the schedule full, not so much that jobs go stale before anybody reaches them. This is a rule of thumb passed around the trade rather than a published standard. It is not drawn from any regulation, standard or recognised benchmark, and it should not be adopted as a target in its own right. The right figure for your site depends on your asset base, shift pattern and how seasonal your workload is.
What matters far more than the number is the direction it is moving in. A backlog sitting steadily at six weeks is under control. A backlog that has moved from two weeks to four over a quarter is telling you something important, whatever the absolute figure and whatever range anybody says is normal. Record it monthly and watch the line rather than the reading.
Backlog by age
Track the age of the oldest open item alongside the total. A ninety-day ageing threshold works well as a review trigger for general work, but it is a prompt to look rather than a universal standard, and the right threshold depends on the type and criticality of the job. For safety-critical work it should be considerably shorter. Anything that has sat untouched past whatever threshold you set is usually one of two things: not actually important, in which case close it honestly, or blocked by something nobody has escalated. Both need a decision rather than another month of drift.
Backlog by asset criticality
Forty hours of outstanding work on a single production line matters far more than two hundred hours spread across office lighting and car park signage. Splitting the backlog by asset criticality stops the total figure from flattening out the difference and helps you argue for resource where it actually counts.
What to record for every backlog item
Backlog measurement only works if the underlying records are consistent.
As a minimum, capture:
The asset and its location.
Estimated labour hours and the trade or skill needed.
Parts required and whether they are in stock, on order or unsourced.
The reason the job is still outstanding.
The consequence of continued delay, in plain language.
The date the job was raised, so ageing can be tracked.
How to prioritise a maintenance backlog you cannot clear
Most small teams will never reach zero, and chasing zero is the wrong objective anyway. A healthy backlog is a queue of planned work waiting for the right slot. The goal is to make sure the right things are at the front of it.
Score by consequence, not by who complained loudest
Backlog order in many businesses is decided by whoever asks most persistently. A simple scoring approach removes that. Score each item on the likelihood it will fail or worsen before the next planned window, and on the consequence if it does, covering safety, statutory obligations, production, cost and reputation. Two five-point scales multiplied together are enough. Anything more elaborate tends to be abandoned within a month.
Account for time sensitivity as well as risk
Risk and consequence are not the whole picture. Some jobs are only moderately serious but have a window that closes. Work that can only be done during a planned shutdown, a school holiday, a seasonal lull or a scheduled isolation either gets pulled forward or waits until the window comes round again, which may be a year.
A moderate-consequence job with a closing window can legitimately outrank a higher-consequence job that could be done on any quiet Tuesday. This does not need a third scoring scale: flagging window-bound jobs and reviewing them before each shutdown is usually enough.
Give statutory and safety-critical work its own lane
Statutory inspections and safety-critical work should not compete with routine repairs for schedule space. Put them in a separate queue that is planned first and is not traded away for convenience jobs. This is also the queue an inspector or an insurer will ask about.
What belongs in that queue depends on your site and the duties that actually apply to it. Thorough examination of lifting equipment, gas and pressure system work, fire safety equipment servicing and inspections required by your own risk assessments are common examples. These sit under different regulations with different requirements and different intervals, so identify which apply to you rather than treating them as one undifferentiated category.
Separate quick wins from deep work
Jobs under about half an hour behave differently from jobs that need half a day, staged parts and an isolation. Batch the short jobs into a scheduled block, ideally grouped by area, and clear ten of them in a morning. Plan the deep work individually, with parts confirmed before the slot is booked.
Build a list of work you will not do
Some backlog items should be closed rather than scheduled. Jobs on assets due for replacement within the year, duplicates raised twice by different people, cosmetic requests with no operational impact, and work on equipment that has already been decommissioned. Closing these honestly, with a recorded reason, is legitimate backlog management and often removes a surprising share of the list in an afternoon.
The distinction matters. The aim is not to make the number smaller by hiding work, it is to make the backlog reflect work that genuinely still needs doing. A list quietly pruned to look healthy is worse than a large honest one, because it destroys the evidence you would need to argue for budget or headcount.
How to reduce a maintenance backlog without starving planned maintenance
The classic mistake is to declare a backlog blitz, pull everybody onto outstanding work for a month, and let the preventive schedule slip. Preventive work missed today becomes corrective work in three months, so the backlog comes back larger than it started. Reduction has to happen alongside the planned schedule, not instead of it.
Ring-fence planned maintenance time first
Decide how many hours a week the preventive schedule genuinely needs and protect them as a fixed commitment. Backlog reduction gets what is left. If what is left is nothing, that is a resourcing conversation to have openly rather than a problem to solve by quietly skipping servicing.
Batch by location and by trade
Travel, set-up, isolation and paperwork often cost more than the work itself. Grouping backlog jobs by building, floor or plant room, and by the trade required, can produce a large increase in jobs completed per day without anybody working faster.
Fix parts availability
If most of your backlog is waiting rather than ready, the constraint is procurement. Set minimum and maximum stock levels for spares on critical assets, consolidate orders with fewer suppliers, and check whether long-lead items for known future work could be ordered in advance rather than after the failure.
Use contractors for the batch, not the crisis
Calling a contractor out for a single emergency is the most expensive way to buy their time. If you have six or eight similar jobs on the list, package them into one visit. The day rate covers far more work, and the backlog drops in a visible step rather than a trickle.
Attack repeat causes, not repeat symptoms
If the same asset keeps generating corrective jobs, the backlog is a symptom rather than the disease. Ask whether the preventive interval is wrong, whether the asset is being operated outside its design conditions, whether the last repair was a bodge, or whether the equipment has simply reached the end of its economic life. One replacement decision can permanently remove a recurring stream of work.
All of this rests on one condition: a single source of truth. Every technique above assumes you can see the whole backlog in one list, with hours attached, dates raised and status recorded. If outstanding work is spread across a whiteboard, an inbox and two spreadsheets, you cannot size it, age it or order it, and the first job is consolidation rather than reduction.
Keeping the backlog visible Getting to that single list is usually the hardest part of the job. Until every open item sits in one place with an estimated duration, a date raised and a current status, the size and age of the backlog stay guesswork, and so does the order you work through it in. Trefnus CMMS records activities, defects, assets and service contracts in a single system, with recurring schedules, severity levels, completion history and reporting built in. It runs offline as a progressive web app, so a technician in a plant room without signal can still close jobs down. It is a one-time purchase with a five-device licence rather than a monthly subscription. Find out more at: |
How to stop the backlog rebuilding
Clearing a backlog once is a project. Keeping it under control is a routine, and the routine is what most businesses skip.
Control the quality of what comes in
Vague requests generate rework and inflate the list. Require every reported job to identify the asset, the location, what is actually wrong and what it is preventing. A ten-word request that nobody can act on will sit on the backlog for months purely because it is unclear.
Hold a short weekly backlog review
Fifteen to thirty minutes a week is usually enough. Look at what came in, what is ageing, and what is blocked. The single most valuable output is assigning an owner and a date to each blocked item, because blocked work is what quietly turns a manageable backlog into an unmanageable one.
Make the preventive schedule honest
If your maintenance schedule demands sixty hours a week and you have forty-five available, the backlog is arithmetic and no amount of prioritising will fix it. Review intervals against manufacturer guidance and your own failure history. Some tasks will be too frequent and can be safely extended. Others will be too infrequent and are generating corrective work. Document the reasoning either way.
Report the number to somebody who can act on it
Backlog hours, backlog weeks and the age of the oldest open job should be visible to whoever controls budget and headcount. A backlog that only the maintenance team can see is a backlog that will only ever be addressed after something breaks.
Frequently asked questions
What is a maintenance backlog?
A maintenance backlog is the total amount of identified maintenance work that has been raised but not yet completed. It includes overdue preventive tasks, corrective repairs from defect reports, jobs found during inspections and work postponed while waiting for parts, budget or access. It is normally measured in estimated labour hours rather than in the number of open jobs, because job sizes vary too widely for a simple count to be meaningful.
How do you calculate a maintenance backlog?
Calculate backlog in labour hours rather than job counts. Add up the estimated labour hours for every outstanding job, then divide by the labour hours genuinely available each week once reactive callouts and preventive maintenance have been accounted for. The result is the backlog expressed in weeks. For example, 180 outstanding hours against 45 available hours a week gives four weeks of backlog. Recalculate monthly, because the direction of travel is more informative than any single reading.
How many weeks of maintenance backlog is acceptable?
Many maintenance teams treat two to four weeks of ready backlog as a comfortable working range, which is enough work to keep the schedule full without jobs going stale before they are reached. This is an industry rule of thumb rather than a formal standard, and the appropriate figure varies with asset base, shift patterns and seasonality. The trend matters more than the absolute number: a backlog that is stable is under control, and one that is growing quarter on quarter is not.
What is the difference between a maintenance backlog and deferred maintenance?
A maintenance backlog is all outstanding work, including jobs that are simply queued and waiting for a slot. Deferred maintenance is the subset that has been deliberately postponed, usually on cost grounds, with somebody accepting the consequence of the delay. Backlog is largely a scheduling matter. Deferred maintenance is a budgeting decision and should be recorded, with its risk, so it can be reviewed rather than forgotten.
How do you prioritise a maintenance backlog?
Score each item on the likelihood that it will fail or worsen before the next planned opportunity, and on the consequence if it does, taking account of safety, statutory obligations, production impact, cost and reputation. Keep statutory and safety-critical work in a separate queue that is scheduled first. Batch short jobs together by location, plan long jobs individually with parts confirmed, and close items that will never be done rather than leaving them on the list.
Should you aim to clear a maintenance backlog completely?
No. A backlog of zero usually means either that maintenance needs are not being identified or that the team is significantly over-resourced. A healthy backlog is a queue of planned work waiting for the right slot, which gives schedulers flexibility to batch jobs efficiently. The aim is a backlog that is measured, correctly ordered, ageing slowly and free of overdue safety-critical work, not a backlog that is empty.
Further Reading and Official Guidance
The following sources cover the legal and technical background to maintenance obligations in the UK.
Providing and using work equipment safely: a brief guide (INDG291), Health and Safety Executive. A short introduction to PUWER duties, including maintenance and inspection.
Safe use of work equipment (L22), Health and Safety Executive. The Approved Code of Practice and guidance supporting PUWER 1998.
Lifting equipment at work: a brief guide (INDG290), Health and Safety Executive. Covers LOLER duties, including thorough examination.
Work equipment and machinery, Health and Safety Executive. The main hub for equipment safety guidance by topic and industry.
Maintenance of electrical equipment, Health and Safety Executive. Guidance on maintaining electrical systems and portable equipment.
The Institute of Asset Management. Professional body publishing guidance on asset management practice, including the ISO 55000 series.
Conclusion
A maintenance backlog is not evidence of a failing team. It is evidence that maintenance needs are being identified, which is the first requirement of any functioning maintenance regime. The problem is not that the list exists, it is that in most small businesses nobody knows its size, its age or its risk profile.
Measure it in labour hours rather than job counts. Separate ready work from blocked work and from deliberately deferred work, because each points to a different fix. Protect the statutory queue, batch the small jobs, plan the large ones properly, and close the items that will never be done. Then watch the trend rather than chasing zero.
None of that is possible without a single source of truth. If you do one thing after reading this, get every job, asset and defect into one list where they can be counted, aged and ordered. Everything else here depends on it.
Disclaimer
The information in this article is intended for general guidance only and does not constitute professional legal, financial, or regulatory advice. Always consult a qualified professional for advice specific to your circumstances.
