All prompts
Operations

Customer support operations desk

Design an internal support tool around triage speed, conversation context, and safe actions.

Build prompt

Copy as-is, then replace the variables

Create a responsive customer support operations desk for [COMPANY TYPE]. The primary user is a support agent handling [TICKET VOLUME] conversations per day.

Optimize for fast, accurate triage while preserving enough customer context to avoid careless replies.

Workspace layout:
- Left navigation: Inbox, Mine, Unassigned, Waiting, Resolved, and saved views with counts.
- Ticket list: requester, subject, channel, priority, status, assignee, last activity, and SLA risk. Include search, filters, sort, and bulk selection.
- Conversation: chronological messages, internal notes, attachments, events, and a reply composer with public reply/internal note modes.
- Customer context: identity, plan, account health, recent activity, previous tickets, and relevant orders or projects.

Core interactions:
- Assign, change priority/status, add tags, snooze, merge duplicate, and resolve.
- Provide confirmation or undo for risky actions.
- Add keyboard shortcuts for common navigation and clearly expose them in the UI.
- Use realistic sample tickets for [COMPANY TYPE], including one angry customer, one billing issue, one technical bug, and one unclear request.

States and safety:
- Loading, empty queue, no search results, offline, stale ticket, send failure, and success.
- Warn when another agent is viewing or replying to the same ticket.
- Never fabricate account actions; visually distinguish customer facts from AI-suggested summaries.

Visual direction: [VISUAL DIRECTION]. Prioritize density, scanability, and clear status over decorative cards. Adapt to a focused single-pane mobile workflow.

Expected outcome

A three-pane support workspace with realistic tickets, filters, reply tools, status changes, and robust states.

Why it works

  • Frames the interface around throughput and reply accuracy rather than dashboard decoration.
  • Includes concurrency, stale-data, offline, and send-failure cases common to real support tools.
  • Separates verified customer facts from AI-generated suggestions.
Variables to customize
[COMPANY TYPE]Example: B2B project management SaaS
[TICKET VOLUME]Example: 35–50
[VISUAL DIRECTION]Example: calm professional utility with compact typography

Method sources

Traceable, not copied blindly

This prompt is an original editorial adaptation. The references below informed its structure and review criteria; their text is not presented as our own.

Keep building