Shared Account
Inbound emails are matched to your knowledge base and drafted for review, so your team spends time resolving, not writing from scratch.
How it works
Each step is handled by a specialized AI agent or by your team. Purple steps are AI, teal steps are yours.
A customer or prospect emails your shared address (e.g. info@, support@). The email appears in the shared queue, visible to the whole team.
The system sends an instant acknowledgement with a unique ticket number, so the sender knows their message was received and can reference it later.
The system checks if the sender is already in your CRM. If not, it matches the sender’s email domain to your companies and auto-creates a contact, so new inquiries from known accounts are linked automatically.
The Knowledge Agent searches your organization’s knowledge base for relevant information like product docs, FAQs, policies, and previous resolutions, selecting what’s most useful for this specific inquiry.
The Content Agent drafts a reply using the retrieved knowledge and your team’s tone guidelines. The draft is ready for review, not auto-sent.
A team member claims the conversation from the shared queue. Claiming prevents duplicate work and assigns clear ownership for the thread.
The team member reviews the AI draft, makes any edits needed, and sends the reply from the shared account. The email threads naturally under the original conversation.
When a team member edits the AI draft, the Knowledge Agent evaluates the changes. Meaningful corrections like factual updates, tone adjustments, and new information are automatically added to the knowledge base for future use.
How it compares
Traditional support tools manage tickets. Amplycom manages the thinking that goes into each reply.
| Amplycom | Traditional tools | |
|---|---|---|
| Draft replies | AI drafts grounded in your knowledge base | Canned responses selected manually |
| Contact matching | Auto-links sender to CRM by domain and creates contacts automatically | Manual contact creation or separate CRM integration |
| Knowledge base | Learns from every edit automatically | Articles maintained by hand |
| Reply quality | Improves over time as team edits accumulate | Static unless someone updates templates |
| Setup | Connect your email and start. No widget needed | Widget embed, portal setup, custom domains |
| Built for | Small teams handling mixed inbound | Dedicated support departments |
Every time a team member edits an AI draft, the Knowledge Agent evaluates whether the change contains something worth learning: a factual correction, a better phrasing, a new policy detail. Valuable edits are stored in the knowledge base and used to improve future drafts. The more your team uses the system, the less editing it needs.