This is a piece in defense of the person who copied last year's budget. Not because the copy is good, but because copying is the rational move inside the process they were handed. The failure is a lack of tools, not a lack of desire.
In this guide
The rational shortcut
Picture the drafting moment honestly. Eight weeks, a book of associations, and a blank grid. The information that should drive next year's numbers exists, but not one piece of it is in front of the drafter. Under those conditions, "last year plus a percent" is not negligence. It is triage. The person is fine. The starting position is impossible (I wrote about the season's arithmetic in Nobody Chose 3%. The Calendar Did.).
Where the year actually lives
The year's truth is written down. That is what makes this failure so maddening. It lives in board minutes recording the contract renegotiated in March. In a loan agreement whose final payment lands in June. In a lease that ends in October. In the reserve study's funding schedule. In the renewal your broker flagged in an email.
Nothing in this process ever reads these. The draft starts context-blind, so the draft starts as a copy.
When I took over my own association's books, the biggest corrections did not come from fixing arithmetic. They came from documents the budget had never been told about. Every one was sitting in a folder the whole time.
The half-fix managers already found
Here is the generic method, and it costs nothing: before a single number gets typed, drop the year's documents, the minutes, the contracts, the loan agreements, the reserve study, into an AI session and ask what changes next year's budget. Let the machine do the remembering.
Plenty of managers already do this informally. And they are right to. Reading a year of unstructured documents used to be the advice nobody had time to take. Now the reading is cheap. That sentence was not true two years ago: what changed is the arrival of language models that can interpret meaning and extract context from documents never structured for software, and it is the entire reason this method exists now instead of a decade ago. The manager who does this walks into drafting with the March contract, the June payoff, and the October lease already surfaced, which is most of what boards catch drafts missing.
But stop there and you have exactly half a fix.
The other half: the math must be boring
An AI reading documents is interpretation, and interpretation is probabilistic. Useful, fast, and not the thing you can take into an annual meeting. The other half is that the derivation must be deterministic: the same inputs producing the same assessments every time, computed the way the governing documents prescribe, auditable line by line.
An interpretation you cannot audit is an opinion. A derivation you cannot reproduce is indefensible in the meeting where it matters.
So the shape of the real fix is targeted AI: the AI's one job is translating the meaning and context of the documents into a system, and the system runs the computations according to those documents. AI interprets; the algorithm derives. Never AI doing the math, and never a chat window off to the side of the real work.
Some associations already budget this way
Inputs gathered before numbers. Interpretation assisted, derivation deterministic. A draft that starts from the year instead of from last year. None of this requires heroism, and the associations doing it are not smarter. Their process just reads.
In the interest of transparency: this is the gap my company works on. We offer a free, easy-to-use tool governed by your association's own documents. It ensures every budget line is properly coded, who is responsible and on what basis it is shared, as the documents direct, or as an explicit board decision where the documents leave that to the board. The math underneath is deterministic. The method above works with or without it.
Related: The Three Ways Every HOA Budget Goes Wrong · Seven Inputs Your HOA Budget Ignored.
