The invoice said $14,200. A chiller compressor replacement at a mid-size office building, handled over three days, vendor paid on completion. What nobody calculated: this was the third repair on that unit in 22 months. The total spend across those three events was $31,800. The unit was 17 years old. The decision to repair rather than replace had been made each time based on the same incomplete reasoning: it’s working, the repair cost is lower than replacement, and nobody had documentation showing the full picture.
A replacement chiller for that building would have run $38,000 to $52,000 installed. The operation spent $31,800 over 22 months to avoid a decision it didn’t have the data to make confidently. Then made the replacement anyway — after the unit failed completely during an August heat wave, with guests in the building.
The operation spent $31,800 over 22 months to avoid a decision it didn’t have the data to make confidently. Then made the replacement anyway — after the unit failed completely during an August heat wave.
Why the Real Number Never Appears in Any Report
Asset tagging exists in most commercial operations. The chiller had a tag. It was in the system. The asset register showed model number, location, and installation date — when that data had been entered correctly, which is not always the case.
What the tag didn’t have: service history tied to spend. Each of the three repair events lived in three separate work orders, billed by two different vendors, recorded in a system that had no mechanism to aggregate the spend by asset over time. The FM who approved the third repair didn’t know about the pattern. The ownership group that asked about capital planning that quarter didn’t know the unit had cost $31,800 in the last two years.
This is not a data entry problem. It is a structural problem. Most CMMS implementations record work orders. They don’t surface lifecycle spend by asset as a visible, actionable metric. The data exists in disaggregated form. The intelligence — the pattern that changes the decision — is never assembled.
According to IFMA benchmarking data, facility teams that lack consolidated asset maintenance history consistently underestimate equipment replacement costs and overextend asset useful life beyond manufacturer recommendations. The consequence is not just the repair cost. It’s the emergency premium, the downtime cost, and the deferred CapEx that arrives at the worst possible time.
The Operation That Has That Number
An operation with lifecycle intelligence tied to each asset doesn’t look dramatically different from the outside. It has the same vendors, the same equipment, the same building. What’s different is one number that appears without anyone having to calculate it: total maintenance spend on this asset, over its operating life, against its current estimated replacement cost.
When that number is visible — when the FM reviewing the third repair quote can see that this unit has consumed $31,800 in corrective maintenance in 22 months and is operating 3 years past its expected useful life according to ASHRAE guidelines — the repair vs. replace conversation changes. It moves from intuition to data. From “the repair is cheaper” to “the repair is cheaper than replacement today, but more expensive than replacement was 18 months ago, and significantly more expensive than replacement will be after the next failure.”
IoT condition monitoring accelerates this further. Runtime hours tracked continuously, current draw trending over time, temperature differential patterns logged against baseline — these are the inputs that convert an asset tag into asset intelligence. Not the tag itself. The history that accumulates around it.
Repair records and costs live in separate work orders and invoices, hiding the cumulative pattern of failure.
Total maintenance spend is automatically aggregated against the asset’s operating life and replacement cost.
Decisions shift from reactive intuition to a strategic investment schedule based on asset condition and repair patterns.
The ROI Nobody Is Calculating
McKinsey documents that organizations operating with predictive maintenance programs — which require exactly this kind of asset-level spend and condition visibility — reduce maintenance costs 30 to 45 percent compared to reactive operations. That range isn’t primarily about catching failures earlier. It’s about making the repair vs. replace decision with actual data instead of incomplete information.
The CapEx conversation also changes. A facilities director who can show ownership a portfolio-level view of asset age, cumulative maintenance spend, and projected replacement timeline is not asking for capital budget. They’re presenting a data-driven investment schedule. The conversation is different. The approval rate is different. The timeline is different.
The Lifecycle Lesson
The chiller that failed during the August heat wave wasn’t a surprise in retrospect. The signals were in the data. Three repair events in 22 months on a 17-year-old unit is a pattern. The operation just didn’t have the infrastructure to see the pattern as a pattern — until it became an emergency. The next time you approve a repair on a critical asset, what does the cumulative spend history on that unit tell you?
Sources:
- IFMA Benchmarking — Asset lifecycle and maintenance spend data: https://www.ifma.org/resources/research-benchmarking/
- McKinsey — Predictive maintenance reduces M&O costs 30-45%: https://www.mckinsey.com/capabilities/operations/our-insights
- ASHRAE — Equipment useful life guidelines by asset type: https://www.ashrae.org
- BOMA International — EER operational cost benchmarks: https://www.boma.org/BOMA/Resources/EER