What 5,000 Maintenance Requests Taught Us About Preventive Scheduling

When you have 5,000 maintenance requests with consistent metadata — unit, date, category, cost, vendor, resolution time — patterns emerge that aren't visible when you're looking at tickets one at a time. The request that lands in your queue on a Tuesday afternoon looks like an individual event. In aggregate, it's part of a seasonal cycle, a building-specific failure pattern, or a unit-age correlation that, once you see it, changes how you approach the whole maintenance calendar.

Here's what the data actually shows.

HVAC Failures Cluster in the Second and Third Week of June

Florida HVAC failures spike predictably. The timing isn't random — it tracks closely with the first sustained stretch of 90-plus degree days combined with high humidity after a mild spring. Units that have been running intermittently through the spring don't fail when they're first turned on in late April. They fail when they've been running at capacity for two consecutive weeks in the summer heat.

In the 5,000-request dataset, 23 percent of all HVAC calls in a given year occurred between June 10 and July 5. Of those, 67 percent involved units that were 7 years or older and had not had a service visit in the prior 14 months. The failure wasn't unpredictable. It was entirely predictable. Those units could have been serviced in April and May at standard rates, with vendor availability and no tenant disruption from heat. Instead, they failed during peak demand when every HVAC contractor in the region is already booked and emergency rates apply.

The cost differential on an HVAC failure versus a pre-season service visit runs roughly 4x to 6x depending on whether there's component replacement involved. On a portfolio with 30 HVAC units approaching end of service cycle, a May service program costs roughly $4,500. Emergency replacements and repairs on units that fail in June run $18,000 to $28,000 in an average failure year. The math isn't subtle.

Water Heaters Signal Failure Three to Six Months in Advance

Water heater failures generate a recognizable complaint sequence before the unit actually fails. The sequence, when you know to look for it: initial complaint is reduced hot water temperature or longer wait time for hot water. Second complaint, typically 6 to 12 weeks later, is inconsistent hot water — works sometimes, not others. The third event is failure or significant leak.

In the data, 74 percent of water heater failures were preceded by at least one prior service call or tenant complaint involving the same unit within the preceding six months. The prior complaint was recorded, the visit was made, the symptom was addressed or the tenant was told it was normal operating behavior, and then the unit failed anyway a few months later.

That first complaint is not a nuisance ticket. It's a replacement forecast. A unit that starts producing inconsistent hot water temperature in a building with average water mineral content, after seven or more years of service, has a predictable remaining service life. The replacement should be scheduled at the next vacancy — or at the first available window where the unit can be swapped without tenant disruption — not after it fails and causes water damage.

Units with Five or More Requests in 18 Months Drive Disproportionate Costs

High-frequency units are a well-known concept in maintenance operations. What the data clarifies is the cost concentration: in the 5,000-request set, units generating 5 or more requests in any 18-month period accounted for 9 percent of total units but 31 percent of total maintenance spend. That's a 3.4x cost ratio versus average units.

The high-frequency units cluster around two explanations: older buildings with aging systems across the board, and units that had an underlying issue (water intrusion, electrical fault, foundation settling) that was addressed symptomatically rather than at the root cause.

The reactive approach to a high-frequency unit is to respond to each ticket as it arrives. The predictive approach is to flag units crossing a frequency threshold for a root cause inspection — specifically, a visit aimed at identifying what is causing the volume of requests, not just addressing the current one. That inspection typically costs $300 to $600 and either finds the root cause, which gets addressed in a single larger intervention, or rules out structural issues and reclassifies the unit as a candidate for targeted preventive work.

Seasonal Predictability by Category

Beyond HVAC, the data shows clear seasonal patterns by maintenance category. Plumbing calls rise 34 percent in the 45 days following Thanksgiving and Christmas — kitchen disposal failures, increased usage, grease buildup. Pre-holiday plumbing checks on buildings with disposal histories are a measurably cost-effective intervention.

Roof and water intrusion calls spike in the two weeks following major storm events, predictably. Less predictable: they also spike in early October, when thermal cycling between warm days and cooler nights accelerates small cracks in roofing material after a summer of expansion. Roof inspections in late September, before the October spike, catch and seal what would otherwise become emergency water intrusion calls.

Window and door seal failures peak in January and February in Florida — not from cold, but from the thermal contraction that occurs during the occasional cold snaps, combined with the summer expansion that preceded it. Buildings with aluminum-framed windows over 12 years old generate window-related requests at 2.7x the rate of newer buildings in January through March.

Building a Preventive Calendar From the Data

The practical output from this kind of analysis isn't a complex algorithm. It's a maintenance calendar built around known failure patterns rather than tenant complaints. For a typical Southeast multifamily portfolio:

March through May: HVAC pre-season service for all units over 6 years, prioritizing those with prior performance complaints. Water heater replacement for any unit with prior performance complaints plus age over 8 years.

September: roof inspection across all buildings, targeted sealing of any identified cracks before October thermal cycling begins.

November: plumbing checks on buildings with disposal histories, kitchen fixture review for buildings with known grease buildup patterns.

Ongoing: flag any unit crossing 4 maintenance requests in 12 months for root cause inspection rather than continued reactive response.

The result isn't zero reactive maintenance — some things will always fail without warning. But the percentage of maintenance spend that's reactive versus scheduled drops significantly, emergency billing declines, and the vendor relationships you've built through regular scheduled work typically deliver better pricing than emergency calls to whoever is available.

The data to build this calendar exists in your ticket history right now. Most operators have it recorded somewhere. The gap is that it's never been organized in a way that surfaces the patterns. Once it is, the preventive schedule writes itself.