Late to this one, but since the thread lands on the right answer we wanted to add a concrete example. @aleclerc had it in posts 3/10: call hunting can’t drop an unanswered call into a specific mailbox (support confirmed in post 9 – it only ever falls back to the DID’s own voicemail), so the way to get “ring device A, then device B, then a chosen voicemail” in one object is a Ring Group with its voicemail field set.
We build an (unofficial) MCP server that lets an AI assistant drive a VoIP.ms account through the same public API, so we tried lemko’s exact scenario in plain language – home SIP phone first, then the cell, then voicemail – without naming any VoIP.ms feature:
On my home line 647-478-2641, I want incoming calls to ring my home SIP phone (sub-account 552036_demo) first, then my cell (forward that to my second number 647-478-5907), and if neither picks up send the caller to voicemail box 101.
The interesting part: it reached the same conclusion this thread did, on its own. It read the current routing, noted the DID has no native 3-step chain, and picked a Ring Group – then asked how long each leg should ring before building anything:
After we picked ring times, it wired up the whole chain end to end – created the forwarding for the cell, built the ring group (account:552036_demo 20s → fwd: cell 20s, voicemail = 101), and pointed the DID at the group:
Net result is exactly what lemko wanted: home SIP for 20s → cell for 20s → mailbox 101. It even flagged that the DID’s own 60s dial timeout and none: failover fields don’t matter here, because the ring group owns the whole chain including the voicemail fallback.
Nothing here you can’t do by hand in the portal – it’s the same ring-group approach @aleclerc and Adam already confirmed works. Just a different way to get there if you’d rather describe what you want than click through the menus.
(We’re not affiliated with or endorsed by VoIP.ms.)