Nobody re-measures a takeoff before close. There is not time, and a second full pass would consume the hours the estimate needs for pricing and review.
That is the wrong ambition anyway. Quantity errors are not distributed evenly across a takeoff. They cluster in a small number of predictable places, and a deliberate pass through those places costs an hour and catches most of what a full re-measure would.
Re-checking is not re-measuring
The distinction matters. Re-measuring asks whether the number is right. Re-checking asks whether the number is still answering the question it was taken off to answer.
Most quantity failures at close are not arithmetic. They are quantities that were correct when taken and have since been invalidated by something else — a revised sheet, a changed assumption, a scope boundary that moved.
1. Anything taken off a superseded sheet
Every re-issued drawing invalidates the takeoff derived from it, whether or not the revision cloud touches your scope. The cloud marks what the designer changed; it does not mark what your quantities did in response.
Keep a record of which sheet revision each quantity came from. Without it, this check is not possible at all, and reconstructing it under deadline pressure is exactly the work nobody has time for.
2. Quantities at the scope edges
Interior quantities are usually fine. The errors live at boundaries: where your scope meets another trade's, where a building meets its site, where a phase meets the next one.
These are where double-counting and gapping both occur, and they are invisible in a total. Two trades each pricing the same slab edge produces a number that is too high and looks reasonable. Neither pricing it produces one that is too low and looks equally reasonable.
3. Quantities that came from somewhere else
Any quantity your firm did not take off — supplied by a subcontractor, carried from a similar past project, or read off the owner's own estimate — carries an assumption you did not make and probably did not see.
A sub's quantity reflects the sub's scope boundary, not yours. A quantity carried from a past job reflects that job's details. Both are legitimate inputs and both need the assumption behind them stated before they are trusted at close.
4. Quantities the bid form transformed
The bid form frequently asks for a quantity in a different unit, a different aggregation, or a different measurement basis than the takeoff produced.
Every conversion is an opportunity for a factor-of-ten error, and unit conversions are the single most common source of catastrophic estimate mistakes precisely because the resulting number still looks plausible. Check the unit of measure on the bid form against the specification, not against your takeoff — the bid form and the specification disagree more often than either disagrees with you.
5. Quantities nobody owns
On a split takeoff, some items sit between two estimators, and those are the ones that get taken twice or not at all. The same is true of items that arrived by addendum after the takeoff was divided.
The check is not a measurement. It is asking, for each addendum and each scope area, who took it off — and treating "I assumed you had it" as a finding.
Schedule the pass
The reason this does not happen is not that estimators disagree with it. It is that the last day before close is fully committed before it starts, and an unscheduled review is a review that does not occur.
Put it in the schedule as its own block, before pricing rather than after, and give it a fixed scope: the five categories above and nothing else. An hour spent on the quantities most likely to be wrong is worth considerably more than the same hour spread evenly across quantities that are mostly right.
Related reading: when the drawings and the specifications disagree.
