TG Customer Service System Agent Seat Model: 3/5/20 Quota, Project Permissions, and Concurrent Sessions

When your Telegram community or e-commerce bot starts having multiple customer service agents handling user inquiries in shifts, a core question arises: How to avoid information chaos, unclear responsibilities, and uncontrolled permissions caused by multiple people sharing the same Telegram account? The answer is to introduce the “agent seat” model. TG-Staff, as a customer service and operations SaaS platform for Telegram Bots, helps teams upgrade from “shared-device customer service” to “professional division of labor customer service” through independent web-based agent accounts, flexible quota configurations, and granular project permissions. This article will delve into the 3/5/20 quota options, project permission configuration, concurrent session mechanism, and collaboration features of the TG customer service system’s agent seats, helping you avoid detours during deployment.

Why Does the TG Customer Service System Need the “Agent Seat” Model?

In traditional mode, teams usually handle Telegram customer service in the following ways:

• Multi-account rotation: Each agent uses their own Telegram account to join the bot group or directly reply to user private messages, but it is easy to confuse who replied to whom, and users often experience agent switching.
• Shared device: Everyone logs into the same Telegram account of the bot, message records are chaotic, and specific agent operations cannot be traced.
• No permission isolation: All agents can see all conversations, including sensitive information or customers not under their responsibility.

The agent seat model maps each agent to an independent web-based login account. Each agent has their own identity, operation logs, and permission scope. This means:

• Agent A cannot see Agent B’s private notes (Pro version);
• Agent C can only handle conversations for project X and cannot access project Y;
• Administrators can audit each agent’s response quality and risk behavior.

The core value of this model is: Shift customer service management from “rule of man” to “permission and process-driven”, especially suitable for teams in cross-border, Web3, e-commerce, and other fields with high compliance and efficiency requirements.

TG-Staff Agent Quota: How to Choose 3/5/20?

TG-Staff’s plan directly determines how many independent agent accounts you can have simultaneously. The quota comparison is as follows (subject to the latest information on the official website):

Plan Type	Agent Quota	Suitable Scenario
Free Trial	1 agent	Personal testing or feature experience
Standard	3 agents	Small team (1–2 agents for shift work, or 3 agents online simultaneously)
Pro	5 agents	Medium team (3–5 agents for shift work or online simultaneously)
Higher Quota	20+ agents	Large customer service team (contact sales for customization)

Selection Suggestions:

• Small customer service team (1–2 agents for shift work): If your team only has 1–2 agents working in shifts (e.g., 1 agent for morning shift, 1 for evening shift), the Standard plan with 3 agents is sufficient, leaving one for an administrator or backup.
• Medium team (3–5 agents): If you need more than 3 agents online simultaneously, or three shifts with 1 agent each, it is recommended to directly choose the Pro plan. 5 agents can cover most shift needs for teams of 5–8 people.
• Large team (20+ agents): If the number of customer service agents exceeds 20, the fixed quotas of Standard and Pro plans may not be enough, and you need to contact TG-Staff sales for a customized solution.

Project Permission Configuration: Controlling What Agents Can Do and See

TG-Staff’s project permission management is highly flexible. You can assign permissions to each agent in “Project Settings → Agent Scope.” The following three common scenarios cover the needs of most teams.

Scenario 1: Full Agent Permissions (Default)

Suitable for small teams or single business lines. All agents can access all projects, ideal for unified management in a customer service center. For example, in a cross-border e-commerce Bot project, all agents can handle all order inquiries without permission isolation.

Scenario 2: Specified Agent Permissions

Suitable for multi-business line teams. For instance, if you run both an NFT project Bot and an exchange Bot, you may want agents 1 and 2 to handle only the NFT project, and agents 3 and 4 to handle only the exchange project. In project settings, set the agent scope for the NFT project to “Specified Agents” and select agents 1 and 2; do the same for the exchange project. This way, agent 3 will only see conversations from the exchange after logging in, preventing misreading or misresponse.

Scenario 3: Permission and Routing Rule Integration

This is a more advanced usage. In “Conversation Routing” rules, you can further restrict which agents can receive conversations from specific sources. For example:

• Set up a routing link specifically for Facebook ad traffic;
• In the project corresponding to that routing link, set the agent scope to “Specified Agents”;
• Meanwhile, in the routing rule, select “Online First” distribution.

The final effect: users from Facebook ads will only be assigned to specified online agents (e.g., agents A and B), while other agents cannot see these conversations. This achieves a “Project + Agent + Source” three-level permission control, ideal for ad attribution and channel isolation.

Concurrent Session Mechanism: How Many Users Can an Agent Handle Simultaneously?

A common misconception is that agent quotas determine how many users can be served simultaneously. In reality, TG-Staff does not impose a hard concurrency limit—in theory, an agent can open any number of conversation windows at once. However, practical experience shows that unlimited concurrency harms response quality.

Recommended reasonable concurrency range:

• 3–5 active sessions: For most agents, this is the ideal range for maintaining both response speed and reply quality.
• Over 10 active sessions: Response times will noticeably increase, leading to user frustration and potential complaints.

The system’s built-in session assignment logic helps manage concurrency:

• Round-robin (default): New conversations are assigned to all agents with permission in sequence, ensuring load balancing.
• Online First: Prioritizes agents currently online. If all agents are offline, it falls back to round-robin. Suitable for peak periods requiring quick responses.

Session Transfer and Collaboration: Mechanisms for Agent Collaboration

Even with reasonable permissions and routing, agents need to collaborate. TG-Staff provides three key features to prevent information gaps.

When to Transfer a Session?

• Agent off-duty or shift change: Before Agent A leaves work, they transfer unfinished sessions to Agent B, so the user does not need to re-explain the issue.
• Encountering a specialized issue: Agent A finds that technical expert intervention is needed midway through handling a session, and directly transfers the session to Agent C.
• User complaint escalation: User dissatisfaction escalates, and the session is transferred to a team leader or administrator.

After transfer, the receiving agent can view the complete chat history and assignment records (who handled the user and at what time). The user side is unaware of the transfer, ensuring a seamless experience.

Suggestions for Using Private Notes

Private notes are a Pro feature that allows agents to add internal remarks to a session, invisible to the user. Recommended uses:

• Record user preferences (e.g., “Prefers to communicate in English,” “Previously complained about logistics”);
• Mark to-do items (e.g., “Need to follow up on refund status”);
• Leave handover information (e.g., “Order number confirmed, please follow up on shipping”).

This way, even if the session is transferred, the new agent can quickly understand the context and avoid asking repetitive questions.

Common Misconceptions: Relationship Between Agent Count, Login Devices, and Concurrency

Here are three common misunderstandings that need clarification:

1. Agent quota ≠ number of simultaneous online devices: One agent account can log in on multiple devices (e.g., Web, mobile) but counts as only 1 agent quota. The system records the latest login session activity and does not consume additional quota due to multi-device login.

2. Agent quota ≠ total number of users that can be served: One agent can handle an unlimited number of users simultaneously (limited only by human capacity). Agent quota only limits “how many people can log into the system at the same time,” not “how many users can be served in total.”

3. Agents during the free trial cannot be used in production: After the trial ends, bound Bot projects and data are retained, but agents will not be able to receive new sessions. If you do not renew after the trial, it is recommended to migrate data or upgrade your plan in advance.

Frequently Asked Questions

Q: Is the Standard plan (3 agents) sufficient?

A: If your team has only 1–2 people working shifts or online simultaneously, 3 agents are usually sufficient. However, if you need one person per shift (morning, afternoon, night) or more than 3 customer service agents online at the same time, it is recommended to upgrade to the Pro plan (5 agents) or higher. You can first try the 3-day trial and evaluate based on actual load.

Q: Can agent permissions be set to “read-only”?

A: Currently, TG-Staff supports project-level permissions (accessible/inaccessible) but does not yet support “read-only mode.” All agents with permission can send messages. For content auditing, it is recommended to enable the Pro content moderation feature to monitor and restrict agent message content. This feature can configure risk word detection to prevent agents from mistakenly or illegally sending sensitive information.

Q: If one agent logs into multiple devices, does it consume multiple agent quotas?

A: No. One agent account can log in on multiple devices (e.g., Web, mobile) but counts as only 1 agent quota. The system records the latest login session activity to avoid message confusion. However, it is recommended that each agent stays active on only one device to avoid message conflicts.

Q: What happens if the number of agents online exceeds the plan quota?

A: The system will not forcibly log out, but the excess agents will not receive new session assignments. It is recommended to check current agent usage in the console under “My Subscription” and upgrade your plan in time. For example, with the Standard plan (3 agents), if a 4th person logs in, they can see the console interface but will not receive any new session assignments.

Q: Will data be lost after the free trial ends?

A: No. After the trial ends, your projects, user data, and chat history will be retained. However, agents will not be able to receive new sessions or log into the console. All features will be restored upon renewal. It is recommended to fully test agent permissions and routing rules during the trial to ensure correct configuration before going live.

Conclusion and Action Recommendations

The core value of the TG customer service system’s agent model lies in: independent accounts, controllable permissions, efficient collaboration. It resolves the chaos of multiple people sharing a Telegram account, making customer service management standardized and professional. Whether your team has 2 or 20 people, TG-Staff provides matching agent quotas and permission plans.

Next Steps:

• Sign up for a trial: Free 3-day trial to experience agent configuration and project permission management → https://app.tg-staff.com/
• Read the documentation: Learn more about agent management, routing rules, and permission settings → TG-Staff Docs - Agent Management
• Contact support: If you have questions, you can directly contact the official support Bot → @tgstaff_robot

Start configuring your first agent now, and let your customer service team move away from chaos towards efficient collaboration.