1) Let the agent answer safely
It answers routine questions using grounded knowledge.
SupportHQ is built for human handoff: the agent answers routine questions fast, and escalates cases to your team with context so customers don't get bounced between replies.
When handoff happens, your team doesn't start from scratch. SupportHQ keeps the conversation context so your next response is fast, accurate, and consistent.
The agent handles common questions using your knowledge base.
The agent escalates the cases that need a person to resolve, and steps back the moment a teammate replies.
How it works
A clear workflow from agent to team.
It answers routine questions using grounded knowledge.
It escalates when a case needs human judgment, writing its own reason, urgency, and summary.
Your team gets alerted, sees the conversation history, and can respond immediately.
In depth
The agent decides when to escalate, explains why, and steps back the moment your team steps in.
Three triggers are built in. Confidence is low because the answer is not in the knowledge base. The customer asks for a person. Or the request needs an action the agent cannot take, such as a refund, an account change, or a bug that needs investigating.
You can shape this with plain-English rules per project, for example escalating anything that mentions billing. Read the handoff workflows guide for patterns that work.
The agent writes its own reason for escalating, sets an urgency, and summarizes the conversation so far. Your teammate opens a thread that already says what the customer wants, what the agent tried, and how urgent it is.
Your team is notified by email, Telegram, or SMS, per project, so nobody has to watch an inbox to catch an escalation.
The customer stays in the same conversation, on the same channel, whether that is the website widget, Telegram, or Discord. They are not sent to a form or asked to email someone. When a person replies, it appears in the thread they were already in.
The agent tells the customer it is bringing in a person, so there is no silent gap where they wonder if anyone is there.
The moment a teammate replies, the agent goes quiet on that conversation and does not talk over them. When the issue is resolved, the answer your teammate wrote can be turned into a knowledge base article in one step, so the agent handles it next time.
Escalations are tracked in insights alongside auto-replies and resolutions, so you can watch the handoff rate fall as your knowledge base grows. All of this runs from the unified inbox.
When confidence is low, when the customer asks for a person, or when the request needs an action the agent cannot take. You can add plain-English rules per project to escalate specific topics.
The agent's written reason, an urgency level, and a summary of the conversation so far, plus the full thread in the unified inbox.
By email, Telegram, or SMS, configured per project.
No. They stay in the same conversation on the same channel. A human reply appears in the thread they were already in.
No. The moment a teammate replies, the agent goes quiet on that conversation.
Yes. Anyone on your team can join any conversation from the inbox at any time.
Usually. Answers your team writes can become knowledge base articles in one step, so the agent covers more questions and escalates fewer. Insights show the trend.