Destination was part of the device configuration
The organization’s departments and users had different requirements. The delivery team introduced hospital-specific labels and shipping boxes holding 30 phones. These changes helped receiving staff route devices to the intended teams while reducing packaging overhead.
At this scale, a correctly configured phone delivered to the wrong department is still an incomplete deployment. The labeling work connected the technical identity of the device to its physical destination. It made the receiving process part of the delivery system rather than leaving local staff to infer where a box of apparently similar phones belonged.
Automation ended with a check of the actual device state
Python and Swift scripts handled preparation, updates, enrollment profiles, Workspace ONE enrollment and additional API tasks. An automated health check verified the desired device state before the workflow completed. Provisioning took place centrally or onsite as required.
The final check is significant because the sequence contained dependent operations. Requesting enrollment and obtaining a usable, managed endpoint are different states. A process can finish sending commands while the device has not reached the condition the user needs. The project’s explicit health check connected the completion decision to the endpoint rather than only to the execution of the script.
Existing devices required a local exchange
The healthcare organization required refreshed phones to be replaced onsite. Staff credentialed for healthcare deployment performed that work. Central preparation therefore supported the local exchange instead of eliminating it.
The distinction captures the physical side of a mobility program. A new device can be prepared far from the hospital, but the working user still needs to receive it in the environment where the old device is being replaced. Keeping those responsibilities separate gives each stage a clear purpose: provisioning establishes the intended state, and the onsite exchange connects that state to the actual user and location.
The operating program extended beyond the rollout
The January 2020 account described continued orders in waves of 1,000 to 5,000 devices and support for approximately 100,000 iOS endpoints. A mobile provisioning cart was still in development. The reported annual savings of more than $3 million were a comparison with the previous provider’s proposed costs, not a published audit of cumulative cash savings.
Cartennas’s assessment is that the project’s strongest feature was the connection between automation, destination accuracy and local delivery. None of those elements substitutes for the others. Together, they explain how a large healthcare fleet could be prepared consistently while still accommodating hospital-specific recipients and mandatory onsite exchanges. The completed operating model, including continuing order waves and lifecycle support, provides stronger evidence than a claim that the initial shipment alone constituted success.