Developers handing over 120 or so units a year drown in snag lists tracked by email and memory. A photo-form pipeline uses AI to classify each defect by trade, routes it to the right subcontractor and runs SLA timers to signed-off resolution.
Typical results
6 days
Average fix time
-71%
Time to resolution
0
Snags lost once live
120
Units handed over yearly
Figures reflect typical results for this kind of automation.
The challenge
Every handover produced a snag list — a dripping tap, a scratched door, a socket without power — reported by phone, email, and WhatsApp to whoever answered. Items were forwarded to subcontractors and forgotten; roughly one in seven vanished entirely. Fixes averaged three weeks, buyers called daily to chase, and warranty goodwill evaporated fast.
Before and after
Average fix time
−71%Snags lost or forgotten
−100%Buyer chasing calls
−80%The automation
Buyers now report every defect through a photo form linked to their unit. AI reads the description and images, classifies the snag by trade — plumbing, electrical, carpentry, finish — and opens a work order with the responsible subcontractor, complete with photos and access details. SLA timers track each trade's committed deadline, overdue items escalate to the project manager automatically, and a snag only closes when the buyer confirms the fix with a photo sign-off.
How it works
- 01
Snag reported
The buyer photographs the defect and submits a short form linked to their unit.
- 02
AI classifies
The snag is categorized by trade — plumbing, electrical, carpentry, or finish.
- 03
Routed
A work order lands with the right subcontractor, with photos, unit, and access details.
- 04
SLA clock
Timers track each trade's committed deadline; overdue snags escalate to the project manager.
- 05
Signed off
The buyer confirms the fix with a photo — only then does the snag close.
Starts from
- Buyer photo reports
- Handover snag lists
Delivers
- Routed to subcontractor
- SLA timers & escalations
- Photo sign-off closure
The impact
Average fix time typically falls from twenty-one days to six, and snags stop getting lost. Buyer chasing calls drop by about four fifths, subcontractors work from clean photo-documented orders instead of forwarded emails, and the snag process becomes a point buyers mention favorably at handover.
Questions about this automation
How does the AI decide which trade a snag belongs to?
It reads the buyer's description and the attached photos and assigns a trade, such as plumbing, electrical, carpentry or finish. When the description is vague or confidence is low, the snag goes to the project manager for a quick manual call before any work order is sent, so subcontractors only receive jobs that belong to them.
What happens when a subcontractor misses the deadline?
Each trade has a committed fix time, and a timer starts the moment the work order is sent. Before the deadline the subcontractor gets a reminder. Once it passes, the snag escalates to the project manager with its photos, unit and history, so overdue items surface on their own instead of waiting for the buyer to chase.
When is a snag considered closed?
Only when the buyer confirms the fix. After the subcontractor marks the job done, the buyer receives a link to approve it with a photo or reopen it with a note. Reopened snags go back to the same subcontractor with the original timer history, and every step is logged against the unit.