A reserve study is only as reliable as the fieldwork, pricing, and honesty behind it — and a weak one is worse than none, because it lends a professional cover page to numbers that don't deserve it. You don't need to be a reserve specialist to spot trouble. The six signs below are all visible to a board member reading closely, and any two of them together are reason to get a second opinion.
- The most dangerous reserve study isn't the pessimistic one — it's the rosy one, because it licenses underfunding.
- The biggest tells are visible to non-experts: no recent site visit, template numbers, costs that don't match real bids, implausibly healthy results.
- The CAI National Reserve Study Standards require specific disclosures — service level, funding goal, cost sources, inflation and interest assumptions. Their absence is itself a red flag.
- A second opinion doesn't mean starting over: an update with a site visit from a different firm usually settles it.
1. No site visit behind the numbers
Reserve studies come in three service levels: a full study with on-site measurement (Level I), an update with a site visit (Level II), and a desk update with no visit at all (Level III). Desk updates are legitimate — in an alternating cadence. What's not legitimate is a chain of them: a community that has bought Level III updates for six straight years is paying for extrapolations of an inspection nobody remembers.
Check the cover page of your current study for its level, then work backwards: when did a preparer last actually walk the property? If the answer is more than three years ago, the condition assessments — and every remaining-life estimate built on them — are guesses wearing old data. In California this isn't merely sloppy: the law requires a diligent visual inspection at least every three years (Civ. Code §5550, verified September 2026), so a chain of desk updates can put the association out of compliance, not just out of date.
2. Suspiciously round, template numbers
Real fieldwork produces messy numbers. A component list where every replacement cost lands on a clean multiple of $5,000 and every useful life matches the industry table exactly — 25 for every roof, 20 for every boiler, regardless of climate, usage, or what the preparer saw — reads like a template with your association's name typed in.
Useful lives should start from industry tables, but a careful preparer adjusts them: a coastal roof weathers differently from an inland one, a heavily used pool differs from a decorative one. If nothing in the study reflects anything specific about your property — no photographs, no component-level condition notes, no local adjustment — the "study" may be a spreadsheet exercise.
A related tell hides in updates: compare the new study's remaining lives to the last one's. If every component's remaining life is exactly the old number minus the years elapsed, the preparer rolled the file forward without judging anything — real components age unevenly, and a genuine reassessment shows it.
3. Costs that don't survive contact with a bid
The sharpest test of a study is a real bid. When your association actually goes to market for a project, compare the winning bid to the study's replacement cost for that component. One overrun is life; systematic 30–40% gaps mean the study's pricing is stale or sourced from nowhere — and the damage isn't limited to one line.
Work the arithmetic: your study prices the roof at $180,000, but the real bid comes in at $260,000. The roof is 15 years into a 25-year life, so its share of the fully funded balance was booked at 15/25 × $180,000 = $108,000 when it should be 15/25 × $260,000 = $156,000. If the study's total fully funded balance was $500,000, the truth is closer to $548,000 — which quietly drops your reported 50% funded ($250,000 balance) to about 46%. Underpriced components don't just misprice projects; they overstate your health.
4. A 40-year-old building at 90% funded
It happens, but rarely — and when a study reports an implausibly healthy number for an aging property, the explanation is usually in what's missing: components left off the inventory, remaining lives stretched heroically, or a fully funded balance computed against a thin component list. Percent funded is a fraction, and shrinking the denominator flatters it as effectively as growing the numerator.
Be equally suspicious of the reassuring flatline: a 30-year projection where the balance never dips near zero and no year ever stacks two major projects together. Buildings don't age that politely. The rosy study is more dangerous than the grim one, because a board that believes it will underfund with a clear conscience.
5. Missing or phantom components
Read the inventory against reality. Two failure modes:
- Missing: the boiler, the elevator modernization, or the retaining wall your CC&Rs make the association responsible for simply isn't on the list. Every omission understates the fully funded balance and the needed contribution.
- Phantom: items that don't belong in reserves at all — routine landscaping, annual servicing, minor repairs that are properly operating expenses. Padding the list with operating items muddies the plan and double-counts costs your dues already cover.
Walk the property with the component list in hand once per board term. It's the cheapest audit your association will ever do.
6. The required disclosures aren't there
The Community Associations Institute's National Reserve Study Standards (verified September 2026) require a study to disclose, among other things: which service level was performed, the funding goal and methods behind the plan, the sources for its cost estimates, the interest and inflation assumptions in the projection, and any other business the preparer has with your association.
These disclosures are what make a study auditable. If you can't find the inflation assumption, can't tell which funding goal the recommended contribution aims at, or can't tell whether anyone visited the site — the standards say you should be able to, and a preparer who follows them will have put those answers where you can find them.
What to do if you counted two or more
First, ask before you conclude. Take the six questions from our reading guide to your current preparer — a good one can defend every line, and a defensive non-answer is information too.
If the answers don't hold up, commission an update with a site visit (Level II) from a different firm and compare the two studies' component lists, costs, and percent funded. You're not paying for the same document twice; you're paying to find out which one to trust — usually a fraction of a full study's cost, and a fraction of a fraction of the special assessment that a bad plan matures into.
And keep the distinction straight: a study that delivers bad news is not a bad study. The preparer who tells you you're 35% funded with a boiler due in four years is doing their job. The one who tells a 40-year-old building it's fine might not be.