Bad handoffs feel jarring; great ones feel invisible. A field guide to writing the rules that decide when your AI steps aside.
The single moment that decides whether customers love AI support is the handoff. A graceful transition — where the human agent picks up with full context and the customer doesn't have to repeat themselves — is invisible. A clumsy one feels like being passed around at the DMV.
Tickki triggers a handoff automatically when one of four signals fires:
When the AI passes a conversation, the message it sends sets the customer's expectations. We A/B tested 40 variants. The two principles that won:
1. Don't apologise for the AI. "Sorry the bot couldn't help" frames the experience negatively. "Let me get a teammate" reframes it as an upgrade. 2. Give a concrete time. "Someone will be with you shortly" loses to "Adaeze will be with you in under 2 minutes."
The agent who picks up should never have to ask "what was your order number again?". Tickki ships a one-paragraph summary to the agent the moment they open the conversation: customer's stated intent, what's been tried, what's still unresolved.
This is the single biggest UX win. We measured a 22% reduction in agent handle time on handed-off conversations after we shipped the summary.
The often-forgotten move: when a human resolves something the AI almost handled, capture that as training data. Tickki's retrace feature replays the conversation as if the AI had been able to answer, and surfaces the missing knowledge or tone gap. Over a quarter, this is what makes your AI quietly competent.
Get the handoff right and the AI–human boundary disappears for your customers. They just remember being helped.
Spin up a Tickki workspace in 5 minutes. No credit card needed.
AI drafts can sound generic and corporate. Here's the playbook our customers use to get drafts that sound like their best agent on her best day.
WhatsAppWhatsApp can be your highest-converting channel — or a black hole for support tickets. The difference comes down to five setup decisions you make on day one.
ResearchMost support dashboards measure the wrong things. After analysing 50M conversations, here are the four numbers that correlate with customer lifetime value.