Repair or Replace? The Math HOA Boards Skip — and the $38,000 Patch That Bought Two Years
Here's a conversation I've had more times than I can count, in some version, at some community, in some month of budget season.
The roof is leaking again. The board gets two numbers: $38,000 to repair the failing sections, or $210,000 to replace it. The room does the arithmetic everyone does — replacement is five and a half times the cost — and votes for the repair. Reasonable. Responsible, even. Nobody wants to be the board that spent $210,000 when $38,000 would do.
Two years later it's leaking again.
I want to be careful here, because the board didn't do anything stupid. They compared the two numbers in front of them. The problem is that those two numbers aren't comparable, and nobody in the room said so out loud.
Almost everything I've written in this newsletter is about running a project well — bidding it, contracting it, documenting it, closing it out. This one is about the decision that happens before any of that: whether the project should exist at all, and in what form. Get this one wrong and every good decision after it is applied to the wrong project.
1. Stop Comparing Prices. Compare Cost Per Year of Service.
One line of math changes the whole conversation.
Take the price and divide it by the years of service it buys you.
Repair: $38,000 ÷ 3 years = $12,700 per year
Replacement: $210,000 ÷ 25 years = $8,400 per year
The "expensive" option is a third cheaper. It just doesn't look that way on a single year's budget, and a single year's budget is what boards are staring at when they vote.
I'm not claiming the math always lands this way — often it doesn't, and I'll get to when repair genuinely wins. The point is that most boards never run the second calculation at all. They compare $38,000 to $210,000, feel responsible for choosing the smaller one, and never ask what each number actually buys.
Run the division. It takes thirty seconds and it will occasionally reverse your vote.
2. Ask the Contractor One Question: Does This Reset the Clock?
The math above only works if you know the years. So make the bidder tell you.
Ask every repair bidder to state, in writing, the expected service life of what they're proposing. Not the warranty term — the service life. Those are different, and the gap between them is where boards get hurt.
You'll find that "repair" covers two very different products:
The repair that resets the clock. Full-system work, manufacturer materials, a manufacturer's warranty behind it, and a genuine service life measured in years — sometimes most of a new component's life at a fraction of the cost. These are excellent buys and boards underuse them.
The patch. Localized, labor-warrantied for a year, addressing the symptom that's visible today. It has value — it stops the leak — but it does not extend the life of the assembly around it. And when the replacement finally happens, your patch gets demolished along with everything else. You paid twice for the same square footage.
A contractor who won't put a service-life number in writing has told you which of the two you're buying.
3. Count the Money You'll Spend Twice
Three $38,000 repairs are not $114,000.
Every mobilization gets paid every time. Setup and demobilization. Scaffolding or lifts. Permits and engineering, if the scope triggers them. Bid solicitation, leveling, and contract review — your time, or your manager's time, or mine. And the cost nobody puts in a spreadsheet: three separate rounds of resident notices, blocked parking, noise, and complaints. As I wrote in the piece on living in a construction zone, disruption is a real cost to the community even when it never shows up on an invoice. Spending it three times to avoid spending it once is a bad trade.
There's a market cost, too. Construction pricing doesn't wait for your reserve schedule. The replacement you deferred for six years is not the same price in six years — and the repairs you bought in the meantime don't come off that number.
4. When Repair Is Genuinely the Right Call
I'd be selling you something if I pretended replacement always wins. It doesn't. Repair is the correct decision when:
The failure is isolated and the component has real life left. A section damaged by a specific event isn't a system at end of life. Don't let one bad area condemn a component with fifteen good years in it.
You're bridging to a funded replacement. If the full replacement is scheduled and funded two or three years out, a repair that carries you to it is exactly right. That's not a patch — it's a bridge, and it's a legitimate use of operating funds.
The cash genuinely isn't there. Sometimes the community can't absorb a special assessment this year without real hardship, and the honest answer is to repair now, fund deliberately, and replace on a defined date. That's a plan. What isn't a plan is repairing every year because each year's assessment conversation is uncomfortable.
The distinction I'd offer: repair as a decision is fine. Repair as a default is how a community ends up with a $210,000 problem and a reserve balance that went to patches.
5. Document the Decision Before It Gets Political
Whichever way you vote, put the whole picture in the minutes: both options, both costs, both service-life estimates, the cost-per-year math, and the reserve impact of each.
Two reasons, and they both matter.
Boards turn over. In three years, a different set of volunteers inherits the consequence of tonight's vote, and they deserve the reasoning — not just the outcome. Without it, they're relitigating a decision they have no context for, usually while water is coming through a ceiling.
And it protects you. If you repaired and it failed early, the record shows a documented, reasoned decision rather than a board that reflexively grabbed the cheap number. If you replaced, the record shows why the larger expenditure was the fiscally conservative choice — which is exactly the conversation you'll be having at the annual meeting with an owner who read the assessment notice and nothing else.
This is the documentation habit I wrote about early in this series, applied to the decision instead of the construction.
The Bigger Point
The board that voted for the $38,000 repair wasn't being careless with the community's money. They were being careful with it — using the only comparison anyone put in front of them.
That's the part I'd ask you to take from this. The failure usually isn't a bad vote. It's a good vote taken on an incomplete question. "Which costs less?" has an obvious answer and it's frequently the wrong one. "Which costs less per year of service, counting every time we'll mobilize to do it?" is a harder question, and it's the one your community is actually asking you to answer.
Ask the harder question. Sometimes it confirms the repair. Sometimes it saves the community six figures. Either way, you'll be able to explain the decision three years from now — and in this business, that's most of the job.
I'd like to hear it from the field: what's the component your community has repaired one time too many? I suspect a lot of us are going to name the same three things.
Next up: if the city inspector isn't checking quality, who is? A look at third-party testing and inspection — and why the report should belong to the association, not the contractor.
Sean Davis, PMP | Civil Engineer | Current HOA Board Member (12 years) | DHA Construction Management | Broomfield, CO
Comments