Case study·Social housing

Follow-on work prevention

From closing jobs that came back — to catching follow-on work before the operative left site

12%
Repeat visits reductionSame-property, same-trade follow-on jobs vs the 60-day baseline
“It caught the jobs that looked closed but were not really finished.”
Repairs Manager · UK social housing provider
UK social housing provider·Around 9,000 homes across one regional repairs operation·5-week pilot·United Kingdom
Talk to us about this workflowTry Unify
01Pilot envelope
Pilot length5 weeks
First signal7 days
First ROI35 days
Team alongside5 seats · 2 colleagues

5-week quality pilot; joined job notes, parts and photos; the colleague flagged likely follow-on work before closure. The Co-ordinator ran read-only against 60 days of historical jobs first, then graduated to approval-gated pre-closure prompts owned by supervisors.

02What it owns
Reports toRepairs Manager, with a dotted line to the Procurement Lead on parts-related closures.
Owns
  • Pre-closure risk score — every job carries a likelihood-of-follow-on rating drawn from past job notes, trades and property history
  • Evidence completeness check — required photos, notes and parts records reconciled against the agreed closure rules before sign-off
  • Follow-on flag and reason — short note explaining why a job is at risk and what the suggested second visit or trade is
  • Supervisor approval queue — flagged closures held in Service Connect until the supervisor approves or overrides with a recorded reason
  • Weekly recall review — repeat-visit patterns by trade and property type, source-linked back to the original job
Does not do
  • Booking the follow-on appointment — supervisors and planners retain scheduling decisions
  • Overriding closures — flags only, the supervisor makes the call to hold or release the job
  • Operative performance judgements — surfaces patterns; the Repairs Manager owns those conversations
Done looks like

Operatives close jobs once. Supervisors see likely follow-on work before the van leaves the property, with notes, parts and photos already lined up. Repeat visits stop being a monthly KPI surprise and become a controlled, evidenced exception.

03The team
AI teammates2
NoraReads job notes, parts orders and photos against historical follow-on patterns by trade and property, scores closure risk and surfaces likely follow-on work to the supervisor before the operative presses complete.
ReubenChecks the closure pack against the agreed evidence rules, drafts the missing-evidence prompts back to the operative on the mobile app, and queues the closure for supervisor approval once parts, photos and notes are aligned.
Human team5
  • Repairs ManagerRepairs leadership
  • 4 SupervisorsRepairs operations
  • 5 OperativesField repairs
  • Procurement LeadProcurement
  • Data AnalystRepairs insight
04Connected stack
TotalmobileService ConnectOperative mobile appPurchasing systemSharePointMicrosoft Teams
05What it returned
12%Repeat visits reductionSame-property, same-trade follow-on jobs vs the 60-day baseline
44%Missing evidence at closure reductionClosures with absent photos, notes or parts records vs baseline
38Jobs prevented from false completionHeld at the supervisor approval queue during the pilot window
  • Day 0
    Co-ordinator sessionRepairs Manager, supervisors and the Procurement Lead align on closure rules, evidence requirements and the trades most prone to follow-on work.
  • Day 7
    First signalNora completes a read-only pass over 60 days of historical jobs; the recall pattern by trade and property is sized for the first time.
  • Day 14
    Read-only flags livePre-closure risk scores land in the supervisor view in Teams; supervisors correct two trade groupings and one evidence threshold.
  • Day 24
    Approval-gated closuresReuben holds flagged closures in Service Connect; supervisors approve or override each case with a recorded reason.
  • Day 35
    ROI reviewSponsor signs off the three baseline metrics and approves the pre-closure check as a standing step in the repairs workflow.
↗Get started

Want this for your team?

Each design-partner pilot starts the same way: one workflow, the minimum useful context, and a first ROI signal measured in days.

Talk to us about this workflowTry Unify