Ask a property manager whether their maintenance expense per unit is high, and you'll usually get one of two answers: either "yes, we had a bad year with HVAC" or "I think it's pretty normal." The honest answer, in most cases, is that they don't have a reliable external reference point to evaluate the number against. Expense benchmarking gives you that reference — and it turns out to be considerably more useful than gut feel.
The goal of benchmarking is not to drive every cost category to the lowest possible number. It's to identify which categories are running meaningfully outside the range that comparable properties achieve, understand why, and decide whether that deviation is acceptable or actionable. Sometimes the deviation is explainable (older building stock, deferred capex being expensed, unusual insurance market conditions in your state). Sometimes it's a vendor relationship that hasn't been renegotiated in three years.
The Categories That Matter Most
Multifamily operating expenses typically break into six meaningful categories for benchmarking purposes: repairs and maintenance (R&M), utilities, property management fees, payroll and benefits (for on-site staff), marketing and leasing, and property insurance plus taxes. Capital expenditures — roofs, HVAC replacements, major unit renovations — are separate and should not be commingled with the operating expense categories for benchmarking purposes, though operators sometimes blur that line during the budget process.
Repairs and Maintenance. For stabilized Class B garden-style product in most major Sun Belt and Mountain West markets, R&M typically runs $700–$1,100 per unit per year. Mid-rise and high-rise product runs higher — $1,000–$1,500 or more — reflecting elevator, mechanical room, and common area complexity. Properties older than 1985 tend to run toward the top of those ranges or above them, driven by appliance age, plumbing condition, and HVAC vintage. Note that turn costs (unit make-ready for new residents) are sometimes categorized within R&M and sometimes separated as a line item — confirm which convention your PMS uses before comparing across properties.
Utilities. The range here is wide and highly location-dependent. A Phoenix property where residents pay all utilities directly and the owner covers common area electricity only will run $200–$400 per unit per year. A Denver property with older boilers and owner-paid heat running through a master-metered system can run $800–$1,400 per unit. Weather normalization is essential for meaningful comparison — a hard winter that pushed heating degree days 15% above average will inflate utility expense in ways that have nothing to do with operational efficiency.
Management Fees. Third-party management typically ranges from 4% to 8% of collected revenue, depending on portfolio size and market. Operators managing in-house will see this line as a salary/payroll item rather than a fee, but for benchmarking purposes, it's worth calculating an implied management cost regardless of structure — comparing in-house staffing costs against what third-party management would cost for the same portfolio is a useful check.
Data Sources and Their Limitations
There are several legitimate sources for multifamily expense benchmarks. The NMHC/IREM Operating Cost Survey has historically been the most comprehensive, covering tens of thousands of units across property types and markets, with data broken out by building type, unit count, and geographic region. Yardi Matrix and CBRE Research publish annual multifamily operating reports with expense data aggregated from their client portfolios. Local apartment associations — affiliated with the NAA — often publish regional surveys that are more granular for specific markets.
The limitation of all of these sources is timing. A published benchmark from Q3 2024 may not reflect the insurance premium spikes seen in coastal and high-risk markets through 2024–2025, or the labor cost increases in maintenance and landscaping that compressed operating margins across the Mountain West. Published benchmarks are useful for establishing a baseline framework; current market intelligence comes from conversations with peer operators and regional property management firms.
We're not saying published benchmarks are unreliable — they are genuinely useful as a sanity check and a budgeting reference. We're saying that treating a published benchmark as a fixed standard without accounting for your specific property type, vintage, and market conditions will lead to either false comfort or unnecessary alarm.
A Portfolio Benchmarking Example
Consider a regional operator running four properties in the Denver metro area — total 310 units, built between 1987 and 2002. On a 2024 trailing 12-month basis, their aggregate R&M per unit was $1,340. Compared against the published NMHC/IREM benchmark for garden-style Class B properties in the Mountain West ($850–$1,150 per unit), that's a notable variance of $190–$490 per unit above benchmark.
The question is whether the variance is explained or unexplained. Drilling into the four properties: two of them ran R&M at $1,050–$1,100 (within normal range for their vintage), one ran at $1,200 (slightly elevated, attributable to a roof repair that was expensed rather than capitalized), and one ran at $1,680 (significantly above range, driven by HVAC failures across six units). That last property is the actual problem — and without the per-property breakdown, the portfolio aggregate masked it.
This is the core value of benchmarking at the property level, not just the portfolio level. When you benchmark only at the portfolio level, outlier properties hide in the average. When you benchmark each property individually, the outlier becomes visible and actionable.
Turn Cost: The Undertracked Expense
Resident turnover generates a cost category that many operators track imprecisely: the make-ready cost per unit vacancy. This includes cleaning, paint, flooring repair or replacement, appliance checks and minor replacements, lock re-keying, and any upgrades applied during the turn. At a well-managed Class B community, typical turn cost runs $1,500–$3,500 per unit depending on the condition the previous resident left the unit in and the scope of any updates applied.
Turn cost is relevant to NOI in two ways: directly, as an operating expense; and indirectly, as a factor in the renewal vs. new-lease calculus. If retaining a resident at a rent discount of $75/month costs you nothing in turn expense, that $75/month concession needs to run for 20–47 months before it costs more than a typical turn (depending on where your turn cost falls in that range). Many renewal pricing decisions are made without this math, which leads to either excessive concessions or a higher-than-necessary turnover rate.
Tracking turn cost per unit per vacancy, rather than burying it in aggregate R&M, gives you the data to run that math accurately. Most PMS platforms can isolate turn-related work orders if the maintenance workflow is tagged correctly — a discipline that many properties implement inconsistently.
Expense Variance vs. the Prior Year: The Most Practical Benchmark
External benchmarks are useful context. Your own prior-year performance is often the most actionable benchmark for day-to-day management. If R&M at Property A ran $920/unit in 2023 and $1,180/unit in 2024, that $260/unit variance is a concrete, investigable number — not a comparison against an industry average that may not fit your specific property characteristics.
Running a monthly budget vs. actual variance report by expense category, consistently, and investigating any category running more than 10% over budget before month-end closes is a discipline that consistently surfaces expense problems earlier than year-end reviews do. By the time a December financial statement shows maintenance running 22% over annual budget, you've already incurred eleven months of overrun. Expense optimization tools that pull directly from your PMS and flag category-level variances in real time address this gap — the data already exists in your system, it's the visibility layer that's missing.
Benchmarking is most useful when it's treated as an ongoing operating discipline rather than an annual budget exercise. The question isn't only "are we in range compared to industry?" — it's "are we trending in the right direction, and do we understand why we deviate where we deviate?" Those two questions, answered consistently, are worth considerably more than any benchmark report.