Getting Started with Your AI Customer Service Agent

Written by:MMMalaz Madani
-

Share Article

1. Create Your Agent App

  • Go to your workspace overview and click “Create a new app.”

  • Choose Customer Service as the app type. Add a name and a short description.

  • Once created, you’ll land on the agent dashboard, where you can see performance stats like conversations, resolved queries, and lead captures.


2. Add It to Your Website

Copy the embed code and paste it into the <head> of your website. That’s it – your chatbot is now live.

Currently, Mando supports web chat. Support for WhatsApp, Messenger, and Telegram is coming soon.


3. Customise the Look

From your app dashboard:

  • Head to Channels > Web Messenger

  • Adjust branding: colours, icons, logos

  • Toggle visibility for linked apps like Help Center and Newsroom

You can preview changes instantly or live on your site with a refresh.

Check the Web Messenger Customisation Detailed Guide


4. Add Quick Prompts

Quick prompts are clickable icebreakers. They help users start conversations – like asking for a discount code.

You can:

  • Set their position (start of chat, after responses, or both)

  • Link them to specific articles


5. Fine-Tune Settings

General Settings

Turn your agent on/off or delete it.

AI Model

Pick your preferred LLM and adjust the tone or prompt if you know what you’re doing.

User Capture

Choose when the bot should collect user info:

  • Mandatory: At the start of the chat

  • On Handoff: Only when escalating to a human

  • Disabled

Human Handoff

Enable it so unresolved chats escalate to real agents. You’ll be notified and can assign tickets to team members.


6. Feed It Knowledge

Think of your Data Library as your AI’s brain.

You can:

  • Crawl your website (like we did for Chelsea Dogs)

  • Upload files (PDFs, docs, etc.)

  • Use Mando’s AI editor to generate support articles

  • Sync content from Google Drive, Notion, etc. (on request)

Make sure your content is linked to the correct app by ticking the box when saving.

Check the Data Library Guide


7. Test It Live

  • Visit your website or click on the preview button

  • Start a chat

  • Try your quick prompts (e.g., “Discount Code”)

  • Check if user capture and escalation work as expected


8. Manage Team & Tickets

When a chat is escalated:

  • You’ll get a notification

  • You can assign the ticket to teammates

  • Add team members from the Members tab

Use AI Analysis to summarise conversations instantly – no need to read through the full chat history.


You’re All Set!

Your AI agent is now ready to handle real customer queries. Don’t forget:

  • Use the free trial to test it fully

  • Send us feedback – we’re building this with you

Did this answer your question?

More Support

Get more support from us

Made by Mando AI