Answer with Iris
Freemium
HVAC, plumbing, electrical, roofing contractors
Running support across email, WhatsApp, Instagram, and Slack means your team is context-switching between five tabs before they've had coffee — and half the responses are still late. Relvios collapses that into one inbox where an AI either handles the ticket or drafts the reply for a human to approve.
The vendor describes a single workspace that ingests Gmail, Outlook, SMTP/IMAP, WhatsApp, Instagram, Messenger, Telegram, Slack, Discord, and webhooks, then runs each message through Gemini-based triage — topic tagging, sentiment, priority, suggested action — before routing to either autonomous AI or a human review queue. You configure that boundary per inbox, not globally, which means you can let AI auto-resolve billing FAQ threads while keeping refund requests in human approval mode. Knowledge grounding pulls from docs, PDFs, Google Docs, or markdown, so answers stay tied to verified content rather than model improvisation. The REST and MCP API lets external agents read threads, draft, and reply inside the same guardrails. Teams hitting edge cases in autonomous handling tend to fall back to draft-only mode, which adds a review step but preserves the unified queue.
Bottom line: Relvios is the right call when your support volume spans four channels and your team is losing tickets in separate inboxes — but teams that need on-premise data residency or a perpetual free tier will hit a hard wall immediately, because neither option exists.
For growing teams
Complete multi-channel support
High volume and scale
View full pricing on relvios.com →
Pricing may have changed since last verified. Check the official site for current plans.
Find your fit
More options in Customer Support / Helpdesk. Compare pricing, platforms, and features side by side.
Freemium
HVAC, plumbing, electrical, roofing contractors
Paid
Australian service businesses
Customer messages arrive from a dozen directions — email threads, Instagram DMs, a Discord server, a website chat widget — and most support tools force you to triage them separately. Relvios is a cloud-only AI support inbox that normalizes all of those channels into one workspace. The core workflow: connect your channels via OAuth, SMTP/IMAP, bot integrations, or webhooks; attach your knowledge sources (docs, FAQs, PDFs, Google Docs, markdown); then choose, per inbox, whether AI resolves high-confidence tickets automatically or prepares drafts for a human to approve before anything sends. Every inbound message becomes a numbered ticket moving through Open, Pending, Resolved, and Closed states, with audit logs and analytics attached.
The differentiating feature is the per-inbox autonomous/HITL toggle. Most AI support tools make you choose one operating mode globally — you are either running an autonomous bot or a draft assistant. Relvios lets you run both in the same workspace: autonomous resolution on your public-facing FAQ inbox, human approval required on your enterprise account inbox. That boundary is configurable without engineering work, which matters when your risk tolerance is not uniform across customer segments.
The API and MCP protocol layer is worth noting for teams building agent pipelines. The vendor describes exposing the inbox to external AI agents with explicit guardrails — those agents can list threads, read context, draft replies, and respond, but within defined constraints. For teams integrating Relvios into a broader automation stack rather than using it as a standalone tool, this is the mechanism that makes that possible.
The fit is clearest for growing support teams managing multi-channel volume without a dedicated platform. Where it breaks: there is no self-hosted option, so teams with strict data residency requirements cannot route around the vendor’s cloud. The freemium model has no perpetual free tier — the vendor states a 7-day trial, after which continued use requires a paid subscription. Teams that exhaust the trial evaluating fit and then need procurement approval face a gap.
Sign in to rate this tool or suggest an edit.