A Foundational Guide for Facility Management (FM-008) · By David Gray | DavidGrayProjects.com
Introduction: The Invoice You Don't See
Deferred maintenance rarely arrives with a warning. It looks like a capital request moved to next year, a pump allowed to run another season, or a roof repair postponed until the budget improves. The savings are immediate and visible. The risk is delayed and much less visible.
That asymmetry makes deferral easy to justify and hard to manage. The goal is not to eliminate every deferral; some are sound decisions. The goal is to make the consequences explicit, rank them by business risk, and tell the difference between disciplined asset management and quiet underfunding.
I have walked plenty of mechanical rooms where a "temporary" fix had quietly become permanent: a bypass line, a rented unit that never went back, a hand-written tag on a valve that no one on the current team could explain. None of those started as a bad decision. Each was a reasonable call made under budget pressure. The problem was that no one had written down what risk was being accepted, so no one knew when it was time to revisit it.
FM-004 made the case for moving from break-fix to proactive maintenance. This article deals with what happens when that shift has not happened yet and a backlog has built up: how to measure it honestly, how to rank it, how to present it to finance, and how to work it down without pretending you can fix everything at once.
1. Deferral Is a Financial Decision
Every deferral reallocates money to another use. Sometimes that is sensible: a redundant, low-consequence asset in a building scheduled for renovation may not justify immediate investment. Sometimes it is reckless: a single-point-of-failure component serving a critical process may be cheap to repair and very expensive to lose.
The right question is not "Can this wait?" It is "What risk, cost escalation, and service impact are we accepting by waiting, and who has agreed to accept it?"
Finance should see three distinct numbers:
The currently identified cost to repair or replace.
The expected cost if intervention is delayed: escalation, secondary damage, emergency premiums.
The downside exposure if the asset fails: outage, safety, regulatory, and reputational consequences.
Do not merge them into one backlog total. A backlog measures identified work. It is not automatically a liability, and it is not a forecast of failures. Presenting a single large number invites one of two bad reactions: paralysis ("we can never fund that") or dismissal ("it has been that big for years and nothing has happened").
Deferral versus disciplined risk acceptance
| Disciplined risk acceptance | Quiet underfunding |
|---|---|
| Asset-level decision with a named owner | Line item cut from the budget total |
| Consequence of failure documented | Consequence assumed to be low |
| Interim mitigation in place (inspection, spares, temporary equipment) | No mitigation; "we'll deal with it if it breaks" |
| Trigger points defined for re-evaluation | Revisited only after a failure |
| Recorded in the risk register | Invisible outside FM |
The same asset can sit in either column. What separates them is documentation and accountability, not the amount of money spent.
2. Build an Asset-Level Backlog
Start with a reliable asset register: location, system, criticality, installation date, condition, remaining useful life, maintenance history, known deficiencies, replacement cost, and operational dependencies. Reconcile it with CMMS work orders, facility condition assessments, commissioning records, and current capital plans. Remove duplicates, and separate routine preventive tasks from capital renewals.
Then segment the backlog:
Life-safety and regulatory obligations. Fire protection, egress, accessibility, environmental compliance, and statutory inspections.
Business-critical reliability work. Systems whose failure stops or materially degrades operations.
Lifecycle renewals. Planned replacement of systems approaching the end of service life.
Efficiency improvements. Projects justified primarily by energy or operating-cost savings.
Discretionary enhancements. Improvements to appearance, amenity, or convenience.
Segmentation prevents an attractive energy project from crowding out a life-safety deficiency. Record the evidence behind every estimate, including the inspection date, estimator, and escalation basis, so the number can be defended when finance challenges it.
Facility Condition Index: useful, but not a risk measure
A common portfolio indicator is the Facility Condition Index (FCI): deferred maintenance cost divided by current replacement value [1]. It is useful for comparing buildings and tracking trends over time. It is not a risk measure. Two buildings with the same FCI can carry very different exposure: one may have cosmetic deficiencies spread across many systems; the other may have a single failing switchgear lineup that could take the whole building down. Never use FCI as the only basis for ranking work, and always state which deficiencies are included in the numerator.
Data quality checklist for the backlog
Is every deficiency tied to a specific asset ID, not just a building?
Is the condition rating based on an inspection within a defined period?
Does each cost estimate state its basis (quotation, unit rate, parametric) and date?
Are escalation assumptions explicit?
Have duplicate entries from different assessments been removed?
Are items already funded in the current capital plan flagged?
If the answer to most of these is no, the first project is not a capital program. It is a backlog clean-up.
3. Rank by Consequence, Not Age Alone
Age is useful but insufficient. Two identical chillers can warrant different priorities if one serves a redundant office plant and the other serves a temperature-sensitive process. Consider probability of failure, consequence of failure, detectability, redundancy, procurement lead time, repairability, and available contingency.
A risk matrix helps, but calibrate its categories to your actual operations, and do not present an ordinal score as if it were a measured failure probability. A "4 x 5 = 20" score tells you relative priority. It does not tell you there is a 20% chance of anything.
For each high-priority asset, write down the consequence pathway: what stops, who is affected, how long restoration would take, and what mitigation exists. This gives FM, operations, finance, risk, and executives a shared language. It is also consistent with the risk-based, life-cycle decision-making emphasized in the 2024 revision of ISO 55001, the international asset management standard, which added explicit requirements on decision-making, risk and opportunity, and life-cycle management [2].
A practical prioritization framework
| Factor | Questions to answer | Why it matters |
|---|---|---|
| Consequence | What stops if this fails? Safety, operations, revenue, compliance? | Drives the size of the downside |
| Likelihood | What does condition, history, and monitoring data indicate? | Drives the timing |
| Redundancy | Is there a standby unit or alternative supply? | Reduces consequence |
| Lead time | How long to procure and install a replacement? | Long lead times amplify consequence |
| Detectability | Will we see failure coming? | Monitoring can buy time cheaply |
| Mitigation | What can we do now for a fraction of the replacement cost? | Allows defensible deferral |
Lead time deserves special attention. Large electrical equipment, custom air handlers, and specialized process equipment can take many months to procure. An asset that would take two weeks to replace is a very different risk from one that would take a year, even if both are equally likely to fail. Confirm current lead times with suppliers rather than relying on historical experience.
4. Understand the Cost of Waiting
Delay can increase cost through emergency labor, premium freight, secondary damage, energy waste, shortened equipment life, and unplanned downtime. None of these outcomes is automatic. Some assets operate safely well past nominal useful life with inspection and targeted maintenance.
Use your own service records, supplier quotations, reliability data, and facility-specific exposure to build scenarios. Generic percentage multipliers circulating in the industry vary widely and rarely transfer between portfolios; if you use one, cite it and test the decision against a range.
A practical comparison is the risk-adjusted present value of three options:
Repair or replace now.
Monitor and defer with defined mitigation and trigger points.
Replace within a planned capital program, bundled with related work.
Include downtime, energy, maintenance, residual value, and tax or accounting treatment. In the U.S., whether spending is expensed as a repair or capitalized as an improvement (a betterment, restoration, or adaptation) is governed by the IRS tangible property regulations, and the answer can change the business case [3]; involve your tax and accounting teams. Show a base case and a downside case. Uncertainty does not disappear inside a single NPV figure.
Worked illustration: three options for an aging pump set
Illustrative example only. All figures are hypothetical, chosen to show the method, and do not describe any actual facility.
A chilled-water pump serving a single critical cooling loop has no installed standby. A condition assessment rates it poor; vibration readings are trending upward. Replacement cost (equipment and installation) is estimated at $120,000. Lead time for a matching pump is 14 weeks. If the pump fails, the operations team estimates a temporary rental pump could be connected within 3 days, at a cost of $25,000 including rental, labor, and temporary piping, and that each day of reduced cooling would cost the business $40,000 in lost production.
Assume an annual failure probability of 15% if nothing is done, falling to 5% with enhanced monitoring and a pre-staged rental agreement (both judgments, not measurements; test them).
| Option | Cost now | Expected annual failure cost | Notes |
|---|---|---|---|
| A. Replace now | $120,000 | Near zero | Planned outage required |
| B. Monitor and defer 2 years | $15,000/yr monitoring + standby rental agreement | 5% x ($25,000 + 3 days x $40,000) = $7,250/yr | Replacement later at escalated cost |
| C. Do nothing for 2 years | $0 | 15% x ($25,000 + 3 days x $40,000) = $21,750/yr, plus emergency premium on eventual replacement | Downside: a failure during peak season with no rental available |
On expected value alone, Option B looks attractive: about $22,000 per year against a large upfront cost. But the downside case matters. If no rental pump is available and restoration takes 14 weeks rather than 3 days, the loss is measured in millions, not thousands. That single scenario may justify Option A or, at minimum, adding a permanent standby pump. The analysis does not make the decision; it makes the trade-off visible to the people who must own it.
Escalation is a real cost - but estimate it, don't assume it
Waiting usually means paying a later price. Construction and equipment costs change with labor markets, commodity prices, tariffs, and supply conditions, and the direction is not guaranteed in any given year. Rather than applying a flat assumption, use a published construction cost index or recent supplier quotations, state the rate you chose and why, and show how the option ranking changes if escalation is higher or lower. If the ranking flips within a plausible range, say so: that is a judgment call for leadership, not a calculation result.
Also account for the cost of a compressed schedule. Work that is planned can be competitively bid, scheduled into a low-impact outage window, and coordinated with other trades. The same work done after a failure is typically procured under time pressure, with fewer bidders and less leverage. That difference is real even when it is hard to quantify precisely.
5. The Portfolio View
At portfolio scale, backlog competes with leases, workplace changes, growth investments, and business-unit priorities. A building scheduled for exit, consolidation, or redevelopment changes what is economically rational to fund.
CRE should bring lease expiries, owned-versus-leased responsibilities (including landlord repair obligations), disposition plans, and future demand to the capital discussion.
FM brings asset condition and operational risk.
Finance reconciles both with available funding and risk tolerance.
The lease timeline in FM-005 is where these conversations should be anchored. Before investing in a leased building, read the lease: some repair and replacement obligations belong to the landlord, some to the tenant, and many are disputed. Restoration clauses can also turn a deferral into a liability at lease end.
This is why a rolling multi-year capital renewal plan matters. Annual budgeting on its own encourages repeated deferral and year-end crises. A rolling plan, typically five years or longer, makes long-lead procurement, bundled outages, and coordinated construction possible.
Hold-period logic
| Expected hold period | Typical renewal stance |
|---|---|
| Under 2 years (exit planned) | Life-safety, compliance, and critical reliability only; mitigate rather than replace where safe |
| 2-5 years | Repair and targeted renewal; avoid long-payback efficiency projects unless they transfer value |
| 5-10 years | Full lifecycle renewal on critical systems; efficiency projects with payback inside the hold |
| 10+ years / owned core asset | Lifecycle-optimized renewal; invest for total cost of ownership |
These are starting points for discussion, not rules. A building held for two years that houses a critical operation may still warrant major investment.
6. How to Catch Up Without Pretending You Can Fix Everything
Isolate immediate life-safety and statutory issues and fund them first.
Address unmitigated single points of failure on critical systems.
Bundle related work where it reduces outage or mobilization cost.
Put condition-based monitoring on assets that can reasonably wait, with defined triggers.
Establish an annual renewal funding envelope tied to a transparent, risk-ranked program.
Be disciplined about scope. A roof replacement may justify coordinated insulation upgrades; it should not automatically become a building renovation. Every addition needs its own incremental business case and owner.
Funding the catch-up
Organizations usually combine several sources rather than relying on a single large appropriation:
A baseline renewal allocation in the annual capital budget, sized to stop the backlog from growing.
A catch-up tranche over several years, prioritized by risk.
Bundling with planned projects, such as fit-outs or energy upgrades, to share mobilization and outage costs.
Lease and transaction events, where renewal negotiations can shift some work to landlords or fund it through allowances.
Disposition, where exiting a building with a large backlog may be cheaper than renewing it.
Be explicit about which of these reduce risk and which only reduce the reported number.
7. Metrics Executives Can Trust
Report:
Backlog by risk category, facility, and planned funding year
Share of assets with current condition assessments
Overdue statutory and life-safety actions
Unplanned downtime on critical systems
Planned versus reactive work ratio
Forecast versus actual renewal cost
Be careful with "backlog reduction." Selling a building, closing stale work orders, or reclassifying items can reduce the reported total without improving the condition of anything you still own. Show those movements separately. A decline in critical, unmitigated risk means more than a decline in total dollars.
A backlog bridge
Present year-over-year change as a bridge, the way finance presents a variance:
Opening backlog
Plus newly identified deficiencies
Plus escalation
Less work completed
Less assets disposed
Plus or minus reclassification
Closing backlog
This one chart answers most of the questions an audit committee or CFO will ask, and it prevents disposals or reclassification from being mistaken for progress.
8. Governance and Accountability
Assign an accountable owner for every major deficiency, with a due date, interim mitigation, and funding decision. Require a documented, signed exception when high-risk work is deferred. Finance should challenge the business case, but not treat the absence of a failure this year as proof the deferral was harmless. FM should challenge unsupported failure assumptions, but not hide uncertainty behind technical language.
A quarterly cross-functional capital review usually works better than an annual scramble, and it applies the same cost-control discipline that keeps capital projects predictable. Use stage gates to refresh estimates, approve designs, confirm outage windows, and release procurement.
What a deferral exception should contain
Asset, location, and deficiency description
Consequence pathway and current risk rating
Reason for deferral and alternatives considered
Interim mitigation and its cost
Trigger points that will force re-evaluation
Approving executive and review date
If no one is willing to sign the exception, that is a signal the deferral should not happen.
9. Presenting the Backlog to the CFO and the Board
Most backlog presentations fail because they lead with the total. Lead with the decision instead.
One of the more productive capital conversations I have been part of started when our team stopped leading with the backlog total. For years the number had been presented annually, it had grown steadily, and the response had settled into polite acknowledgment and a modest allocation. When we instead brought a short list of critical systems, each with its consequence pathway, lead time, and what we were doing in the meantime, the discussion changed. Finance asked better questions, operations leaders recognized their own risks in the list, and the top items were funded that cycle. The total did not get smaller that year. The exposure did.
Start with exposure, not dollars. "Three critical systems have no standby and long replacement lead times" is more actionable than "$X million of deferred maintenance."
Show the trend. Use the backlog bridge to show whether the position is improving, stable, or deteriorating, and why.
Offer funding scenarios. Present at least three levels (for example, hold the line, steady reduction, accelerated reduction) with the risk profile each leaves behind.
Name the residual risk. For every scenario, state which high-risk items remain unfunded and who has accepted them.
Commit to measurement. Tell leadership which metrics will show progress and when they will see them next.
Boards and audit committees increasingly view physical-asset risk as part of enterprise risk, and the same risk-management habits used on capital programs apply here. Framing the backlog in that language, with owners, mitigations, and review dates, makes it far easier to fund than a list of repairs.
10. Common Mistakes to Avoid
Treating the backlog as a fixed number. It changes daily with new findings, escalation, and completed work. Report it as of a date, with a bridge.
Ranking by age. Age is one input. Consequence and lead time often matter more.
Using borrowed multipliers. A figure from someone else's portfolio is not evidence about yours.
Ignoring the lease. Spending tenant capital on landlord obligations, or renewing a building you are about to exit, is a common and avoidable error.
Counting disposals as progress. Selling a building removes its backlog from the report; it does not improve the condition of anything you keep.
Funding enhancements before obligations. Visible improvements are easier to approve. They should not displace life-safety or critical reliability work.
Conclusion: Make the Risk Visible
Deferred maintenance becomes dangerous when an organization confuses avoided spending with created value, the opposite of the long-term asset-value view set out in FM-001. A credible program does not demand unlimited funding. It makes asset condition, business consequences, alternatives, and trade-offs visible enough for leaders to decide deliberately, and to own the decisions they make. That moves the conversation from backlog anxiety to capital discipline.
Next in the series, FM-009 looks at how FM involvement in capital projects prevents tomorrow's backlog from being built into today's new facilities.
Related from Albers Management:
Risk Management: Anticipating the Unexpected – Albers Management
Cost Control – Planning for Predictability – Albers Management
Sources
APPA, "Deferred Capital Renewal and Deferred Maintenance"; NCES, Planning Guide for Maintaining School Facilities, ch. 3 (FCI definition). appa.org; nces.ed.gov
ISO 55001:2024, Asset management - Asset management system - Requirements (published 3 July 2024). ISO/TC 251
IRS, Tangible Property Final Regulations; 26 CFR 1.263(a)-3. irs.gov
Worked examples labeled "illustrative" use hypothetical figures to explain methods.
About the Author
David Gray is a capital delivery strategist, owner's representative, and founder of DavidGrayProjects.com. With over two decades of experience helping organizations bring complex projects to life, from data centers and healthcare facilities to higher-ed campuses, David blends practical delivery with forward-thinking strategy.
He writes about project controls, capital planning, and real estate development to help leaders deliver smarter, faster, and more sustainably.
📩 Connect on LinkedIn | 🌐 Explore More at DavidGrayProjects.com | 🌐 Explore More at Albers Management