Summary

  • Spark's current mobile-network page states that its 3G network shut down on 31 March 2026. That date marks the network change described by Spark, but it does not establish that every handset or connected device completed a successful migration.
  • Spark says the best 4G coverage and experience requires support for Band 28 and Band 3, while voice calling over 4G requires a compatible Voice over LTE, or VoLTE, capability. Compatibility is therefore an operating gate, not a guarantee of connection quality or uninterrupted service.
  • For connected equipment, Spark's Internet of Things guide broadens the check beyond a single device label. It calls for review of modules, antennas, bands, VoLTE where voice is used, SIM requirements, approvals and the complete use case, followed by field and end-to-end testing.
  • The New Zealand Commerce Commission says 4G is not a like-for-like replacement for 3G. It identifies older incompatible handsets, access to 111 calling and older cellular IoT equipment as transition-impact surfaces, without reporting that those risks became a measured Spark outage or an actual 111-calling failure.

A mobile network closure is easy to describe as a change of generations. One network is retired and users are directed toward newer equipment and coverage. That description is administratively neat, but it does not show whether a particular handset, alarm, meter, terminal or other connected system can complete the functions for which it is used.

Spark's own compatibility conditions make the operational question more exact. A device may need the relevant LTE bands to connect as intended. Voice use over 4G requires compatible VoLTE support. An IoT installation may also depend on its antenna, module, SIM, approval status and the way all components behave together. Each element is a gate in a working chain. Passing one does not prove that the others have passed.

The Commerce Commission's assessment reinforces why this chain matters. Its statement that 4G is not a like-for-like replacement for 3G prevents a simple substitution story from becoming the conclusion. Newer network availability and service continuity are related, but they are not the same fact. Continuity has to be evaluated through equipment capability, coverage, configuration and the full solution that reaches the user.

That is also why the article does not claim an outage, a 111-calling failure or universal migration success. The available sources describe conditions, guidance and transition-impact surfaces. They support a practical framework for checking readiness; they do not supply fleet-wide completion findings or evidence that every identified risk materialised.