An anonymised NZ logistics operator brought Aivid's Audit Agent in on week one. By week three, a dispatch-scheduling AI employee was live and the operations team was working fewer hours on a faster cycle.
The customer operates four distribution centres across New Zealand and runs on a roster of roughly 120 staff in dispatch, customs brokering, exception handling, and finance ops. Throughput had been growing year-on-year, but cycle times were inconsistent — and the recovery slips only became visible once they had already hit a downstream customer.
The leadership team had hired two integration consultancies in the previous eighteen months and walked away from both engagements with slides and no live systems. They brought Aivid in to get a process map, not a sales deck.
Dispatch coordinators reconciled driver availability, live order volume, and equipment readiness by hand every shift. They bounced between a shared spreadsheet, a shared inbox, and back-and-forth phone calls with drivers on the road.
Recovery slips were invisible until they hit a customer. The ops team had no shared view of which steps actually took the time — and no baseline to measure any automation effort against.
Customs broker handoffs missed SLA more often than not. Dispute responses averaged 9 minutes, and finance ops closed the month having written off lost invoice value three quarters in a row.
Week 1Live
Process discovery & roadmap
Catalogued 18 repeatable processes across the operations function and scored each on a 0–100 automation opportunity scale. Dispatch-to-dispatch surfaced as the highest scorer.
Week 3Roadmap
Live dispatch decisions
Reads live order volume, driver availability, and equipment readiness. Surfaces the next dispatch decision for a human reviewer and reports daily into the customer's ops inbox.
Week 6In build
Customs SLA & dispute triage
Triage customs broker handoffs to stay inside the SLA window and answer disputes fast. Brought typical dispute triage response time down from a 9-minute baseline to 38 seconds.
The Audit Agent runs end-to-end across dispatch, customs, exception handling, and finance ops. It catalogues 18 repeatable processes and publishes a scored roadmap.
A dispatch-scheduling AI employee is stood up against the top-scoring process. It reads live order volume, driver availability, and equipment readiness, and surfaces the next dispatch decision for a human reviewer.
A second AI employee takes over dispute triage and customs broker handoff triage. The team gets back 11 hrs/week previously lost to manual reconciliation.
An SLA customs-handoff AI employee is stood up. Within-shift customs broker handoffs completed inside the SLA window climb to 94%.
A finance ops reconciliation AI employee closes the loop on lost invoice value. The first month on agent swings the prior three-quarter trend.
Dispatch-to-dispatch cycle time is down 62%. Ops-side weekly hours reclaimed: 11. Customs SLA handoffs within window: 94%. Dispute triage response: 38s. Finance ops reconciliation write-offs: $0 lost invoice value in 90 days.
“We stopped reconciling dispatch by hand on the third week, and the ops team noticed the hours within the first month. The dashboard tells us what's real — it isn't a sales pitch.”