Know When to Escalate

Even the best AI chatbot can't handle everything. Knowing when to hand a conversation to a human agent is just as important as knowing how to answer — it's what keeps customers happy and prevents frustration.

Why Escalation Matters

Lobot handles up to 98% of customer inquiries automatically. But the remaining 2% — complex complaints, edge cases, emotional situations — are often the most critical interactions. A smooth handoff to a human turns a potentially negative experience into a positive one.

The worst thing a chatbot can do is keep trying to help when it clearly can't. Customers forgive a bot for not knowing everything. They don't forgive it for wasting their time.

When Should the Bot Escalate?

Define clear triggers in your custom instructions. Common escalation scenarios:

Always Escalate

• Payment issues — Failed transactions, double charges, refund requests
• Defective products — Broken items, wrong orders, quality complaints
• Legal or compliance questions — GDPR requests, data deletion, warranty claims
• Angry or frustrated customers — When tone analysis suggests the customer is upset and not getting resolution
• High-value orders — Custom or bulk orders that need personal attention

Consider Escalating

• Repeated failures — The bot couldn't answer the same customer's question after 2–3 attempts
• Complex product questions — Highly technical compatibility or customization queries
• Price negotiations — Bulk discount requests or special pricing
• Shipping complications — International customs, special delivery requirements

How to Set Up Escalation in Lobot

In your custom instructions, define escalation behavior clearly:

Making the Handoff Smooth

A good escalation message should:

• Acknowledge the issue — "I understand this is important to you"
• Explain why — "This needs personal attention from our team"
• Provide clear next steps — Contact email, phone number, or expected response time
• Set expectations — "Our team typically responds within 2 hours during business hours"
• Summarize the conversation — So the human agent has context and the customer doesn't have to repeat themselves

What Happens After Escalation

When a conversation is escalated, Lobot:

• Logs the escalation in your conversation overview
• Preserves the full chat history for your support team to review
• Tags the conversation so you can filter and analyze escalation patterns
• Allows you to export the conversation to share with your team

Use escalation data to improve your bot. If the same type of question keeps getting escalated, add better instructions or FAQ content so the bot can handle it next time.

Key Takeaways

• Define specific escalation triggers — don't leave it to the bot to guess
• Always escalate payment, defective product, and legal issues
• Make handoffs smooth: acknowledge, explain, provide next steps
• Track escalation patterns and use them to improve your bot's coverage
• A well-handled escalation builds trust — a poorly handled one destroys it