TG-Staff 团队 avatar TG-Staff 团队

How to Set Up a TG Bot Customer Service: A Step-by-Step Tutorial from Scratch (Bind Bot, Open Agent, Configure Routing)

tg-bot-cs Getting Started Customer Service System Telegram Tutorial

How to Set Up a TG Bot Customer Service: Step-by-Step Tutorial from Scratch (Bind Bot, Add Agents, Configure Routing)

Want to use a Telegram Bot to build a customer service system but don’t know where to start? You’re not alone. Many teams quickly discover that the native solution has several pain points when trying to reply to users directly with a Bot: only single-user login, no conversation history backend, and chaos when user volume spikes. This article is a complete TG Bot customer service setup tutorial, guiding you from binding a Bot to configuring agents and routing, step by step to launch a production-ready customer service system. Whether you’re a cross-border team, Web3 project, or community operator, this workflow applies.


Why Do You Need a TG Bot Customer Service System?

Consider a typical scenario: your project has a community or Bot on Telegram, with dozens or even hundreds of user inquiries daily. You log into the Bot with your personal account and reply one by one. As user volume grows:

  • Messages are easily missed—you don’t know which ones have been replied to
  • No division of labor—multiple people replying to the same Bot? Not supported natively
  • No backend—conversation records, user profiles, and data statistics are all blank
  • No routing—all messages flood to one person, causing crashes during peak hours

That’s why you need a third-party TG Bot customer service platform. Such platforms turn your Telegram Bot into a complete customer service backend: agents log in via a web console, multiple people handle conversations simultaneously, messages are automatically distributed, and user sources can be tracked. TG-Staff is one such no-code solution, and the following steps will use it as an example.


What Do You Need to Prepare Before Setup?

Before you begin, ensure you have these basics ready. The good news: you don’t need to buy a server or write code.

Essential Tools and Account Checklist

ItemDescription
Telegram accountFor creating a Bot and logging into TG-Staff
Bot TokenA secret key obtained after creating a Bot via BotFather
TG-Staff accountRegister for a free 3-day trial
Customer service staff Telegram accountsFor adding as agents (need Telegram ID or username)

Steps to Get a Bot Token (if you haven’t created a Bot yet):

  1. Search for @BotFather in Telegram and click Start
  2. Send /newbot, then follow the prompts to enter the Bot name and username
  3. After creation, BotFather will return a Token like 123456:ABC-DEF1234ghIkl-zyx57W2v1u123ew11
  4. Copy and keep it safe — the Token is the Bot’s “admin password”; if leaked, others can control your Bot

Common Preparation Mistakes

  • Do not share the Token in public groups or GitHub — if leaked, immediately use /revoke in BotFather to regenerate
  • No need to manually set Bot permissions — TG-Staff automatically handles privacy mode, group permissions, etc.
  • No need for your own server — all features run on TG-Staff’s cloud; you only need a browser

Preparation Checklist

□ Register TG-Staff Account → □ Obtain Bot Token → □ Confirm Plan Agent Limit → □ Prepare Staff Telegram Accounts


Step 1: How to Bind Your Bot to the Customer Service Platform?

The core of this step is to hand over the “control” of your Bot to TG-Staff, allowing it to receive user messages and distribute them to the agent dashboard.

  1. Log in to the TG-Staff console (https://app.tg-staff.com/))
  2. Click “Projects” in the left menu, then click “Add Project”
  3. Enter the project name (e.g., “Main Customer Service Bot”), and paste the Token you obtained from BotFather into the “Bot Token” field
  4. Click “Save” or “Create”

After successful binding, you will see the project status change to “Online.” At this point, any user who sends a message to this Bot will have their message appear in TG-Staff’s conversation list, not in your personal Telegram account.

Note: After binding, your Bot will no longer reply directly through your personal Telegram account—all replies must be made in the TG-Staff console. If you need to retain a personal reply channel, you can disable “Takeover Messages” in the project settings (see documentation for specific steps).


Step 2: How to Add Customer Service Agents and Assign Permissions?

“Agent” is a core concept of TG-Staff—each agent has an independent Web login account and can log in to the console to serve users. Unlike sharing a single Telegram account, agents have their own operation records and permissions.

Specific Steps to Add an Agent

  1. Click “Team Management” → “Members” in the left menu of the console
  2. Click “Add Member”
  3. Enter the agent’s Telegram ID (pure numbers, obtainable via @userinfobot) or Telegram username (e.g., @zhangsan)
  4. Set the role:
    • Administrator: Has full project permissions (add/delete projects, manage agents, view all conversations)
    • Regular Agent: Can only handle assigned conversations, cannot modify project settings
  5. Click “Confirm Add”

After adding successfully, the agent will receive a notification from @tgstaff_robot and can log in by clicking it.

Agent Permission Configuration Suggestions

RoleSuitable ScenarioPermission Scope
AdministratorTeam leader or technical operationsProject management, agent management, view all conversations
Regular AgentFrontline customer serviceOnly handle assigned conversations, cannot modify settings

If you have multiple Bot projects (e.g., one for pre-sales consultation, one for after-sales tickets), you can specify during agent addition that the agent can only access specific projects to avoid permission overreach.

Seat Quota Reminder

Free trial and Standard plans have a seat limit (3/5/20). Check your current plan quota before adding seats. The Pro plan supports more seats; see the official website’s pricing page for details.


Step 3: How to Configure Conversation Routing for Efficient Customer Service?

When multiple agents are online simultaneously, who should receive messages from new users? This requires configuring “Conversation Routing Rules.”

Two Routing Modes

  • Round Robin (default): New conversations are assigned to all authorized agents in order, regardless of whether they are online. Suitable for teams with a fixed number of agents and consistent working hours.
  • Online First: New conversations are prioritized for agents currently online. If all agents are offline, it falls back to round robin. Suitable for teams with variable agent shifts or part-time support staff.

Configuration Steps

  1. In the console left menu, click “Project” → select your project
  2. Go to the “Routing Settings” tab
  3. In the “Routing Rule” dropdown, select “Round Robin” or “Online First”
  4. Under “Project Agent Scope,” choose “All Agents” or “Specified Agents”:
    • All Agents: Every agent under the project can receive conversations for this project
    • Specified Agents: Only selected agents handle this project (useful for multi-project division)
  5. Click Save

Best Practice: If your team has morning/evening shifts or cross-timezone agents, use “Online First” to avoid assigning messages to offline agents and causing long wait times.


This is a practical feature of TG-Staff, especially useful for ad traffic generation or multi-channel attribution. A routing link (also called a magic link) is a short URL (e.g., https://app.tg-staff.com/{code}). When users click it, they are redirected to your Bot, and TG-Staff captures the visitor’s IP, browser information, and custom parameters carried in the URL.

  1. In the console left menu, click “Project” → select your project
  2. Go to the “Routing Links” tab
  3. Click “Create Routing Link”
  4. In the “URL Parameters” field, add custom tracking parameters (e.g., utm_source=twitter&utm_campaign=launch)
  5. Copy the generated short link

Example Use Cases

  • Twitter/Instagram Ads: Use routing links in ad landing pages or tweets to track which channel generates the most inquiries
  • YouTube Video Descriptions: Use different parameters for different videos to analyze which content has higher conversion rates
  • Community Promotion: Share routing links with different parameters across multiple communities to compare engagement levels

Shunt Link Best Practices

Use shunt links in ad URLs or social media posts, combined with bot auto-replies and human agent handoff, to form a complete conversion chain. User clicks the link → jumps to Bot → auto-reply with welcome message → agent takes over, the entire process requires no extra steps from the user.


FAQ

Q: Do I need to write code to set up a TG Bot customer service?
A: No. Using a no-code platform like TG-Staff, you only need to add a Bot Token in the console, configure agents and routing rules, and go live — no programming required.

Q: Can I test all features during the free trial?
A: TG-Staff offers a 3-day free trial, during which you can use all Standard features (including routing links and agents). Professional features (such as content moderation and unlimited translations) require an upgrade to test.

Q: Can one Bot be assigned to multiple customer service agents simultaneously?
A: Yes. With session routing, multiple agents (3/5/20 depending on plan) can handle different user sessions at the same time, supporting session transfer and collaboration.

Q: What user information can routing links track?
A: Routing links can capture visitor IP, browser information, and custom parameters in the URL (e.g., utm_source), making them suitable for ad attribution and multi-channel performance analysis.

Q: Will user messages be lost if an agent is offline?
A: No. Messages sent by users are stored in session records, and agents can continue replying after coming online. In online-preferred routing mode, messages are prioritized to online agents.


The above four steps complete the TG Bot customer service setup process. From binding the Bot to configuring agents and routing, to using routing links for source tracking, each step is actionable. If you encounter any issues, you can get help through the following channels: