Support clarification and current status
I would like to correct and clarify the support status described in my earlier updates.
I previously said that Athom had escalated the issue to engineering. Athom’s first-line support has since clarified exactly what this means in practice.
The issue has been recorded on Athom’s internal problem board since 28 August, with my test evidence, screenshots and the independent reproduction attached. First-line support has also applied a high-priority tag.
However:
-
The priority tag is not an engineering-reviewed severity classification.
-
Development has not yet picked up the case.
-
Engineering has not yet reviewed or confirmed the technical findings.
-
There is currently no engineering-triage date or investigation milestone.
-
First-line support has advised that it cannot initiate a separate management, product-owner or service-owner review.
Therefore, for accuracy, this issue has been recorded and prioritised by first-line support, but it has not yet entered active engineering investigation.
The earlier posts in this thread already document:
-
Two current reproductions on separate Homey Pro Early 2019 systems.
-
Similar historical reports involving Homey Pro Early 2019 and Early 2023, including a closely matching failure reported to Athom support in early 2025 without a known resolution.
-
The controlled Virtual Switch test and the evidence supplied to Athom.
I am currently awaiting confirmation of the model and OS associated with @Olivier67’s recent shutters report.
I also still need one comparison result from a Homey Pro (2026). If anyone with that model already uses the official Homey Alexa skill, please try the short test described in post 13: create a temporary native Virtual Switch classified as a light, ensure it is not excluded from Alexa, run discovery and report whether it appears. A successful result is just as valuable as a failure.
Alternatively, if you have recently added any physical or virtual Homey device that Alexa either did or did not discover, that result would also be useful. Please include your Homey model, Homey OS version, Alexa region and whether your previously established Homey devices continue to work through Alexa.
I appreciate the assistance provided by Athom’s first-line support and recognise that it has taken the actions available within its process. My continuing concern is with what happens afterwards.
This affects an important official Homey integration rather than a community-maintained third-party app. Affected users cannot add newly installed or replacement Homey devices to Alexa, and no supported, non-destructive workaround has been identified. Despite that impact, the case currently has no active engineering owner or defined review date and could remain in the internal queue for an indefinite period.
I work as an IT Solutions Architect and have experience across development and technical-support environments. I fully appreciate that engineering teams have competing priorities and that identifying and resolving a fault can take time. What I find difficult to understand is the apparent absence of a time-bound triage or ownership mechanism after first-line support has identified a reproducible failure affecting official product functionality.
In comparable support environments, I would expect such a failure to receive a service-impact assessment, a reviewed priority and a scheduled engineering-triage point, even if a resolution date could not yet be provided. My concern is not with the assistance provided by first-line support, but with the apparent process gap after the issue enters the internal development board.
If an Athom representative—or anyone familiar with Athom’s support process—knows the appropriate route for requesting a product-owner, service-owner or management priority review in these circumstances, I would appreciate that guidance.
I am not asking for an immediate permanent resolution. However, I believe an official integration failure with no supported workaround should receive engineering triage, a priority assessment and a defined next-review date rather than remaining without any timescale.
Thanks,
Hung