Work-in-process carry and draw timing model

Every draw driven by work actually complete, every finished job ranked by what it costs to leave open. Move dates on records that already exist and watch the interest line.

$2.6M

Reference-scale value

Data sets in the model

7 weeks

Build and deploy

Derived

Evidence grade

The decision this model makes

Draws go in when the paperwork is ready, not when the work is done. Jobs close when somebody gets to them. Neither is a decision anybody makes on purpose, and both cost money every day.

One hundred and eighty-five homes sit in work in process against an average balance of $52.9M, financed at 8.2%. Nineteen days of completed work sits on the facility before a draw goes in.

This is the one model in the segment where the mechanism is pure timing. Nothing here asks anyone to build faster, spend less or sell for more.

It moves dates on records that already exist, which is why it carries the highest return in the segment and why a CFO can verify it from the following month's interest line rather than from an attribution argument.

Cause 01

Completed, not drawn

Nineteen days of finished work carried on the facility before anyone submits.

Cause 02

Complete, not closed

Twenty-four days between completion and closeout, on every home. Part finance, part construction.

Cause 03

Drawn ahead of need

Eleven per cent of draws sitting idle. The mirror problem, and it costs the same rate.

Cause 04

Retention off schedule

Holdback moving in both directions. Early release is more common than late, which surprises people.

Want more detail?

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.