Someone claims the exception
Over-threshold is no longer only a pop-up. A ticket lands on the person on shift; close leaves a trail. Duty moves from staring to a queue.
We do not count “looks high-tech” as a gain. These six things a person on shift can feel after go-live: one fewer trip, one fewer reconcile, one fewer “why does that card still swipe.”
Rights follow the role
Exit, role change, visitor expiry: gates and doors die together. Nobody walks a vendor console deleting cards one by one.
Crews copy fewer meters
kWh and air enter MES or crew costing. Handover no longer copies a notebook. One layer of mismatched definition drops.
A scan books stock
PDAs and guns bind wave and location. One less hand-copied slip. Book-vs-stock gaps show the same day.
Patrols you can verify
Scan arrival, a ticket on a miss, an exportable closed-loop list for inspection—not a week of yellow paper.
A board you can drill
Next to uptime is handle latency. Open it and you get a document, not another picture you cannot drill.
These gains share one premise: hardware events must enter the business system you already use—not another closed cloud. A box that locks data empties every efficiency card: When a closed IoT box will not release your data. Devices only online, tickets still empty, and staring does not vanish: Devices are online. Why the floor still stares at screens.
You need not finish all six in phase one. A park can start with access and HR: Access control tied to HR: access dies when employment does; A line can start with energy into MES: Energy into MES: crews stop copying meters by hand; A warehouse can start with scan into WMS: Scan into WMS so books and stock stay closer.
Gains must map to a role before a project can stand
Maintenance wants fewer missed tickets, HR wants a card that dies now, warehouse wants fewer count losses, safety wants an exportable list. Write “what you can see in two weeks” by role. That passes budget more easily than talking IoT. Alerts to tickets: Fewer screens to watch: how alerts become assignable tickets. How to estimate cost and cycle: What hardware-software projects cost, and how long they take.
DaXi Technology writes those abilities as acceptably testable results: dispatch, death of rights, book, export—not a demo cabinet. To land gains in phase one for your site, go to the Hardware-software service page or compare the gap with and without a loop With devices in the business vs. without: what exception handling looks like.
What two weeks can show, and what a quarter must settle, are different claims
Whether dispatch arrives, whether an exit card refuses on the spot—two weeks can prove that. Energy definitions, scan book-vs-stock, patrol miss rate usually need a full roster cycle. Do not use three days of wall dwell time to kill a ticket loop. False-alert rate also needs threshold work: The system never alerts, or alerts too much. How to diagnose.
You can also write gains as internal metrics: mean ticket response, fake-close samples, failed access syncs, whether handover still copies meters. With metrics you know whether to add points or tighten rules. Without them, gains stay on a kickoff slide. Patrols from paper to a loop: Device patrols: from paper checklists to a closed loop.
Visit-path efficiency is easiest to perform: green lamps, moving curves, a voice prompt. Whether the night crew can use it is the real gain. Split visit accounts from on-shift accounts so a demo does not turn suppress off. Fewer screens and auto tickets are two links on the same chain: Fewer screens to watch: how alerts become assignable tickets and After exceptions open tickets, missed handling drops.