Before I explain that specific board, I should explain a pattern I watched repeat across dozens of associations over fourteen years — because the incident doesn't make full sense without it.
Boards regularly misclassify capital repairs as capital improvements. Not out of confusion. Out of incentive.
The distinction matters because of where the money comes from. Repairs draw from the operating account — the annual budget every board is trying to protect. Capital improvements can be funded from reserves, which accumulate year over year specifically to cover large, predictable expenditures. For a board staring at a budget overrun, the temptation to call something an improvement rather than a repair is both understandable and completely inappropriate.
I had this conversation more times than I can count. Replacing two pieces of vinyl siding is not a capital improvement. It is a repair. The reserve fund is not a contingency account for items that didn't fit in the operating budget. The governing documents, in virtually every case I managed, were clear about this — not always in the way boards wanted them to be. Boards that pushed back hardest were usually the ones whose financial position was most precarious. The misclassification was a way of kicking the financial can of worms down the road. A future board would inherit the reserve shortfall. A future special assessment would cover it. The current board's annual numbers would look fine.
The Board That Couldn't Decide
One association came to my firm after years of mismanagement. The governance habits were bad. The institutional memory was worse. When a unit owner's roof developed a significant leak — substantial enough to cause real interior damage, but not a full-replacement scenario by any standard definition — the board couldn't agree on what to call it.
Some members argued it was a repair. Others argued that because the scope of the work was more extensive than a typical patch job, it might constitute a partial replacement — which would let them draw from reserves rather than the operating account. The roof wasn't actually being replaced. But calling it a replacement solved a budget problem, and the financial incentive to make that argument was sitting in the room with everyone who was making it.
What followed was the specific chain of events I find myself returning to every time I work on BoardPath's hierarchy resolution logic:
Every link in that chain was a document hierarchy resolution failure. Not a legal failure, not a communication failure, not a leadership failure — though it was eventually all of those things too. At its root, it was an information retrieval failure: a board that couldn't access the authoritative answer to a governance question fast enough to make a decision before the consequences of not deciding became worse than whatever decision they'd been avoiding.
What the Documents Actually Said
Here is what I found when I read the governing corpus in the correct hierarchical order — the thing nobody in those board meetings had done in the room, in real time, with enough precision to stop the argument:
Three documents. Read in the correct authority order. The answer was there — not ambiguous, not in conflict, not requiring legal interpretation. The Declaration established the obligation. The reserve fund study established what reserves were for. The prior resolution established precedent. Read together, in the right order, they resolved the question.
Nobody assembled them in the room. The seven-year-old resolution existed in a file nobody knew to look in. The reserve fund study was seventeen pages long and three years old. The Declaration was the kind of document boards receive at closing and never open again.
The Design Response
BoardPath's hierarchical document weighting algorithm was built specifically to prevent this kind of failure — not by making boards smarter, but by doing the assembly work before the meeting starts.
When a board asks "is this a repair or a capital improvement," the platform doesn't treat it as an ambiguous philosophical question. It treats it as a hierarchy resolution problem: what does the highest-ranking document in this corpus say about maintenance obligations, what does the reserve fund study enumerate as capital components, and does any prior board resolution establish applicable precedent? The answer gets assembled from the corpus in authority order, with every source ranked and every citation made explicit.
Declaration Article [X]: Association responsibility for common element maintenance and repair confirmed. Reserve Fund Study: Roof enumerated as full-replacement capital component — partial repairs not within defined reserve scope. Resolution [Year]: Prior board concluded substantial roof repair constitutes repair, not replacement. Funded from operations. Confidence: HIGH. No conflict detected between sources. Corpus is unambiguous on classification. Recommended action: fund from operating account and proceed.
That output — generated before the first vote, visible to every person in the room simultaneously — doesn't end the argument because it's authoritative in some abstract sense. It ends the argument because it shows its work. Every board member can see which document said what, how those documents rank relative to each other, and what the accumulated precedent of the corpus actually is. There's nothing left to interpret. There's nothing to argue about that isn't already on the table.
The owner doesn't hire his own contractor. The reimbursement question never gets asked. The court record doesn't exist.
What This Revealed About the Actual Problem
Boards don't make expensive governance mistakes because the rules don't exist. The rules almost always exist. They're in the Declaration. They're in the reserve fund study. They're in a board resolution from seven years ago that nobody remembers passing.
The problem is retrieval friction. Consulting those documents in real time — reading them in the correct hierarchical order, cross-referencing across sources, surfacing applicable precedent — takes longer than taking a vote. The friction of finding the answer is higher than the friction of proceeding without it. So boards proceed without it, and the consequences land years later on someone else's watch.
Document hierarchy resolution is the missing infrastructure layer in HOA governance. Not legal counsel on retainer. Not better board training. A system that reads the architecture correctly, assembles the relevant provisions in authority order, and surfaces the answer in the two minutes before the vote happens — not the three weeks after.
The binders had the answer the whole time. What they needed was something that could read them all at once.