Chatbot human handoff: how to design smooth transfers

When AI assistants should hand conversations to people, how to transfer context so customers don't repeat themselves and how to handle out-of-hours handoffs.

By the UK Technology Hub teamPublished 4 minute read

The short answer

Good handoff means customers reach a person quickly when they need one, without repeating themselves. Trigger handoff when the bot can't help, when customers ask, when frustration appears or when topics are sensitive. Pass the full conversation and key details to the right team, set clear expectations about wait times and handle out-of-hours requests by collecting details and promising follow-up.

Customers don't hate chatbots; they hate being trapped by them. The quality of handoff to people often decides whether customers see your assistant as helpful or obstructive.

When to hand off

The bot can't help

When questions fall outside its knowledge or permissions, it should say so and offer a person.

The customer asks

Requests such as "talk to a person" or "speak to someone" should always be honoured promptly.

Frustration or confusion

Repeated questions, negative language or rephrasing the same request suggest the bot isn't helping. Detect these signals and offer handoff.

Sensitive topics

Complaints, bereavement, financial hardship, health concerns, safety issues and legal matters usually need people.

High-value opportunities

Complex sales enquiries or large orders often benefit from human attention.

Pass context, not just the conversation

Agents should receive:

  • The full conversation history.
  • Customer details already provided.
  • The detected intent or topic.
  • Actions already attempted.
  • Account or order information, where authenticated.

Customers shouldn't need to repeat themselves.

Route to the right team

Use topic, customer type and urgency to route conversations to the right people: sales, support, billing or specialists.

Set expectations

Tell customers what happens next: "I'm connecting you to our support team. The current wait is about five minutes." If wait times are long, offer callbacks or email follow-up.

Out-of-hours handoff

When no one is available:

  • Explain when the team will respond.
  • Collect contact details and a description.
  • Create a ticket automatically.
  • Provide self-service resources meanwhile.
  • Give emergency routes where relevant.

Agent tools

Agents need a good interface: conversation history, customer data, suggested replies and the ability to return the conversation to the bot for routine follow-up. Our guide to chatbots versus live chat explains how bots and agents work together.

Measuring handoff

Track:

  • Handoff rate and reasons.
  • Time to reach a person.
  • Resolution after handoff.
  • Customer satisfaction for handed-off conversations.
  • Conversations abandoned during handoff.

Our chatbot analytics guide covers metrics in detail.

Improving over time

Review handoff reasons regularly. Frequent handoffs for the same topic may mean the bot needs better content or new capabilities, or that the topic genuinely needs people.

Common mistakes

  • Hiding the option to reach a person.
  • Losing conversation history at handoff.
  • Long waits with no updates.
  • Out-of-hours dead ends.

A worked example

A broadband provider's chatbot often frustrated customers by repeating the same answers when it couldn't help. It redesigned handoff so the assistant transferred conversations after two failed attempts, immediately when customers asked for a person and whenever messages showed strong frustration. Agents received a summary of the problem, account details after verification and steps already tried. When agents weren't available, the assistant explained wait times and offered a callback. Customer satisfaction improved, and agents said conversations started faster because they didn't need to ask the same questions again.

Handoff message examples

  • "I'll connect you with a member of our team. They'll see our conversation, so you won't need to repeat yourself."
  • "Our team is available from 8am tomorrow. Would you like a callback or an email reply?"
  • "I'm not able to help with that, but a specialist can. Connecting you now."

Training agents

Agents should know how to read handoff summaries, check what the assistant already said and correct any errors politely. Their feedback helps improve both the assistant and the handoff process.

Learning from handoffs

Review handoff reasons regularly. Common reasons show where the assistant needs better knowledge, new actions or clearer boundaries.

Designing for peak times

When agents are busy, handoff queues grow. Show realistic wait times, offer callbacks or email follow-up and let the assistant continue helping with simpler parts of the problem while customers wait. Review peak periods in advance and plan staffing around them.

Handoff across channels

Customers may start in chat and need a phone call or email follow-up. Make sure the conversation history travels with them, so the next person can continue without asking the customer to explain everything again. Shared case records across channels make this possible.

How we help

Our AI chatbot development team designs handoff flows and triggers, our AI solutions team integrates help desks and routing and our web app developers build agent dashboards with full context.

Frequently asked questions

When should a chatbot hand over to a person?

When it can't help, when the customer asks, when frustration appears or for sensitive topics.

How do I stop customers repeating themselves?

Pass the full conversation and collected details to the agent automatically.

What happens out of hours?

Collect details, create a ticket, set clear expectations and provide self-service resources.

This guide is general information, not legal, tax or financial advice, and details such as prices, features and rules can change. Product and company names mentioned are trademarks of their respective owners and are used only to identify those products; UK Technology Hub is not affiliated with or endorsed by them.

Related guides

All AI chatbots & assistants guides