How to build a unit turn checklist that actually gets followed

Every portfolio has a turn checklist. Most of them are a single long list, and a single long list gets skipped under pressure because there is no point at which skipping is visible.

The version that survives a Friday move-out has stages, and each stage has a gate.

Eight stages, because the handoffs are where days disappear

Move-out. Scope. Paint. Carpet. Punch. Cleaning. Quality inspection. Move-in ready.

The reason to structure it as stages rather than tasks: three of the four causes of a long turn are handoff failures, not work failures. Paint finishes and nobody tells carpet, so the unit sits two days between trades with nobody responsible.

A stage-gated checklist makes that visible. A flat list does not, because the list does not know who is waiting.

Eight stages, and the gate that has to close before the next one opens

The gate is what names the next trade. A flat checklist has the same items and no answer to who is waiting.

Move-out
Keys in, unit secured, meters read.
keys returned, date recorded
Scope
Damage by location, category and severity, with a photo per area and a decision on what is chargeable.
scope approved, materials ordered against it
Paint
Walls, ceilings, trim.
paint signed off, which is what tells carpet to start
Carpet
Replace or clean, per the approved scope.
floors done, unit clear for punch
Punch
The list of small defects a walk turns up.
punch list closed, not just written
Cleaning
Final clean, after every trade is out.
clean after the last trade, not before
Quality inspection
Someone other than the person who did the work walks it.
inspected by a second person
Move-in ready
Available to lease, with the date the P&L will use.
ready date stamped

Work time inside a stage is fairly consistent across properties and techs. The variance lives in the gaps between them, which is why the gate is the thing worth building and the timestamp is the thing worth reading.

Timestamp every stage, not just the endpoints

This is the single most valuable thing you can change today.

Without stage timestamps you know a turn took 11 days. With them you know it spent six of those between paint and carpet, which is a scheduling conversation rather than a performance one.

Most portfolios record move-out and ready. That gives you a duration and no diagnosis.

Scope on day one, properly, or the schedule resets on day four

The most expensive checklist failure is a cursory move-out inspection.

Real damage surfaces on day four, the scope changes, materials have to be reordered, and the whole schedule resets. Every downstream stage was planned against the wrong scope.

What the scope stage has to produce: damage by location, category, and severity, with a photo per area, and a decision on what is chargeable. GetDone's move-out inspection produces exactly that and then fans damages out into real make-ready work orders automatically, so the scope becomes the schedule rather than a document somebody references.

That also means the charge conversation with the departing resident is built from the same record, against a charge catalog with per-property and per-market defaults, rather than reconstructed later.

Eight items per stage is about the limit

A stage with 20 checkboxes gets pencil-whipped. Eight is roughly where attention holds.

Which means the checklist has to be edited down rather than expanded, and the test for each item is whether anybody has ever been harmed by it being missed. If not, cut it.

One real constraint worth knowing about board design: adding tasks by hand is expensive at volume. On some board interfaces, adding ten tasks takes about 20 minutes per unit, which at 190 turns a year is 63 hours of somebody's time. Templates per property exist for that reason, so a turn inherits its task set rather than being built each time.

Require a photo only where a photo settles a dispute

Photo requirements are the fastest way to make a checklist resented, so be selective.

Worth requiring: the unit condition at move-out, per area, because that is the charge dispute. Anything you are charging the resident for. The quality inspection before move-in ready, because that is the one that prevents a new resident's first day being a complaint.

Not worth requiring: every completed task. That is documentation theatre and it teaches people the checklist is not serious.

Different asset classes need different checklists

A 1966 Class C unit and a 2025 lease-up unit do not have the same turn.

The older asset generates more scope discovery and more plumbing surprises. The newer one turns faster and needs less. Running one checklist across both means it is either too heavy for the new asset or too light for the old one.

GetDone's turn task templates are configurable per property, and preventative schedules can target by year built, vintage, and market. That matters at 86 properties, where maintaining the difference by hand is how templates end up on the wrong assets.

What the checklist is actually protecting

At $1,500 a month, a vacant day costs $50. Ayla, a 340-unit Class C community, lost $601,807 in make-ready vacancy loss in a single year, and cut average make-ready time 51% after the work order moved to the tech's phone and the turn moved onto a board.

Two days off an average turn is $100 back per turn. Multiply by your own turn count rather than by a per-100-units figure, because turnover rate is the variable that makes those extrapolations wrong.

Do this with your current checklist this week

Count the items. If it is over 40, it is being skipped somewhere and you do not know where.

Then check whether it has stages with gates, or is one list. And check whether anything is timestamped beyond start and finish.

Those three answers tell you whether you have a checklist or a schedule.

How many stages does yours have?