Every land pro forma contains a date the municipality has not agreed to. Rezoning takes nine months, the outline plan takes six, subdivision takes four. Those numbers come from the last file, or from the consultant, or from what it took the time before that.
None of them come from a record of what that authority has actually decided, because almost nobody keeps one.
Approvals running five months past the assumption carry a million dollars of exposure across the lots in flight. Avoidable correction cycles cost consultant time and calendar. An approval landing after the construction season closes pushes a phase a full year.
The model forecasts an approval date per file per authority from that authority's own decision record, with the sample size and the spread published beside it, and reads resubmission risk against the corrections that authority has issued before.
Decisions running past the pro-forma assumption. Seventy per cent forecastable; the rest is appeal.
Correction cycles that a completeness check would have caught. The line you control outright.
An approval landing after servicing closes. A phase waits for spring.
Charges rising between filing and decision. A municipal decision — the model times it, it cannot change it.
The rest of this model — the data sets it runs on, the cause split, the working behind the reference figure and a sample output — goes out by email. Two fields, nothing else.