What are Tags?

Tags are the classification system that runs through all of Atender. They organize conversations, power filtering and saved views, drive Automations, and help AI agents and routing decide where a conversation belongs.

A tag is short — usually one or two words — and reusable across thousands of conversations. Once a tag is in your taxonomy, every part of the product can read from it.

Two types

Every tag is one of two types, set on the root tag:

• Topic — Red — What is this conversation about? (returns, billing, shipping, technical issue)
• Metadata — Blue — What’s notable about this conversation? (VIP, escalated, follow-up needed, first-contact)

The red/blue distinction makes a conversation scannable at a glance — you immediately see both subject matter and operational status without reading any tag names.

Hierarchical

Tags support parent–child relationships for granular categorization:

Returns (parent — Topic)
  ├── Defective product
  ├── Wrong item
  ├── Changed mind
  └── Size exchange

Sub-tags inherit their root’s type. Reporting works at either level — count all “Returns” or specifically “Defective product” returns.

Where tags are used

Tags aren’t just for organizing the inbox. They’re shared vocabulary that other parts of the product read from:

• Conversations — Tag manually, in bulk, or via AI; filter inbox views by tag
• Automations — Trigger on tag added / tag removed; filter by conversation.tags; add_tag / remove_tag as actions
• Agent Stacks — Specialist agents associated with tags so the router prefers them when those tags are present
• Customer Tiers — Each customer level can carry tags that auto-apply to conversations from contacts at that level
• Analytics — Filter and group every report by tag

Because so much builds on tags, the time you spend designing a clean taxonomy pays back across every other feature. Get this right early.

AI auto-tagging

Each tag has an AI Auto-Tag switch. When on, Sidekick is allowed to apply that tag automatically as conversations evolve. The tag’s description is what the AI matches against — tags without a description don’t participate.

See Enable AI auto-tagging on a tag for setup.

Where to start

• Build your taxonomy: Create a tag, then organize them into hierarchies.
• Decide Topic vs Metadata: Topic vs Metadata reference.
• Make the AI help: Enable AI auto-tagging.
• Sweep an existing inbox: Bulk-tag conversations from the Monitor.