Conversation list

Platica's Inbox intelligently organizes thousands of conversations into logical categories so you can find what you need in seconds. The left side panel is your main navigation map.

Main Conversations

This section is your daily operations center, letting you manage interactions that need your direct attention and constant oversight.

• Assigned to Me

This is where every conversation under your direct responsibility is centralized. Platica's system guarantees an efficient workflow through smart assignment rules based on your role and specialty:

• Responsibilities The system will automatically assign you conversations tied to things you own:
  • Agents: If you are the owner of a specific AI Agent, any message it receives will be assigned to you so you can supervise the interaction.
  • Customers: If a customer has previously been assigned to your user, any new conversation they start will go straight to your inbox.
  • Tags: If you manage specific categories, applying a tag (for example, "Angry Customer" or "High Priority") will route the conversation to your immediate attention.
• Manual Assignment: An admin can delegate specific cases to you that require your expertise or authority level.
• Taking Control: When you proactively jump into a conversation in "Assistance" state, the system marks it as yours to avoid duplicate effort.

• Conversations

This is the master view: every conversation regardless of status, agent, or assignment. It's your audit and advanced search tool.

When to use this view:

• You need to find a specific conversation with multiple filters
• You want to review the entire team's work
• You're doing custom analysis or reports

Conversations by Status

The heart of the hybrid attention system. Each conversation is automatically classified by its attention state, using an intuitive color system for quick visual identification.

Assistance

Critical state: Conversations that need human intervention right now.

They land here when:

• The customer explicitly asked to speak with a human
• The AI detected a situation outside its scope (refunds, serious complaints)
• An escalation rule was triggered (keywords like "cancel", "fraud")
• The AI agent ran out of attempts to resolve the inquiry

Visual indicator: A red number on the bell icon shows how many conversations are waiting for an urgent reply.

Supervision

Active human control: An operator is handling the conversation personally.

Features:

• The AI stays silent until you give control back
• You have access to assisted drafting tools
• You can consult the AI Assistant privately, without the customer seeing it

In Progress

Autopilot: The AI is handling the conversation from start to finish without issues.

These conversations naturally flow toward resolution. You only need to step in if:

• You spot an error from the agent
• You want to review how the AI is replying
• The customer mentions something that requires human follow-up

Started

Waiting on a customer reply: Messages your team or the AI sent proactively.

Common cases:

• Marketing campaigns sent via WhatsApp templates
• Scheduled follow-ups ("Hey, how did things go with the product?")
• Appointment or payment reminders

Once the customer replies, the conversation automatically changes to In Progress.

Closed

Mission accomplished: Conversations that reached a successful resolution.

They get marked as closed when:

• An agent (AI or human) resolved the inquiry and closed the conversation
• 24 hours have passed with no customer reply (automatic expiration)
• An operator archived them manually

Conversations by Agent

A strategic view organized by which AI agent is handling each conversation.

Why is it useful?

If you have multiple specialized agents (Sales, Support, Collections), this view lets you:

• Monitor each agent's individual performance
• Identify which one needs more training
• Distribute the workload evenly
• Analyze which agent has the most assistance requests

Conversations by Topic

Automatic classification based on the reason for contact from the customer.

Every time a conversation starts, the AI agent analyzes the context and categorizes it under one of the topics you defined beforehand.

Strategic benefits:

• You identify trends in your customers' inquiries
• You group conversations by topic for analysis
• You prioritize recurring topics to improve your knowledge base

Dangerous Conversations

Your automatic safety filter. This category protects your operation from unwanted interactions.

Conversations identified as the following show up here:

• ✋ Potential spam (mass messages, suspicious links)
• 🤖 Malicious bots (aggressive automation patterns)
• ⚠️ Harassment or abusive language
• 🚫 Contacts manually blocked by your team

What happens to these conversations?

Automatic block:

• AI agents won't reply to their messages
• They won't consume resources or messages from your plan
• They stay archived for auditing

Manual review:

• You can review each case to identify false positives
• If a legitimate customer was flagged by mistake, you can unblock them
• The full history is preserved for analysis

Quick Navigation

Keyboard shortcuts and tips:

• Use the top search bar to find conversations by customer name
• The number next to each category shows unread or pending conversations
• The colored dots let you identify the state visually without even reading the text
• Click the notification bell to jump straight to conversations needing urgent assistance

Now that you understand how conversations are organized, the next step is to master the chat interface so you can interact with your customers effectively.